To see the other types of publications on this topic, follow the link: YOLOv8.
Journal articles on the topic 'YOLOv8'
Create a spot-on reference in APA, MLA, Chicago, Harvard, and other styles
Consult the top 50 journal articles for your research on the topic 'YOLOv8.'
Next to every source in the list of references, there is an 'Add to bibliography' button. Press on it, and we will generate automatically the bibliographic reference to the chosen work in the citation style you need: APA, MLA, Harvard, Chicago, Vancouver, etc.
You can also download the full text of the academic publication as pdf and read online its abstract whenever available in the metadata.
Browse journal articles on a wide variety of disciplines and organise your bibliography correctly.
1
Sharma, Pravek, Dr Rajesh Tyagi, and Dr Priyanka Dubey. "Optimizing Real-Time Object Detection- A Comparison of YOLO Models." International Journal of Innovative Research in Computer Science and Technology 12, no. 3 (2024): 57–74. http://dx.doi.org/10.55524/ijircst.2024.12.3.11.
Full textAbstract:
Gun and weapon détection plays a crucial role in security, surveillance, and law enforcement. This study conducts a comprehensive comparison of all available YOLO (You Only Look Once) models for their effectiveness in weapon detection. We train YOLOv1, YOLOv2, YOLOv3, YOLOv4, YOLOv5, YOLOv6, YOLOv7, and YOLOv8 on a custom dataset of 16,000 images containing guns, knives, and heavy weapons. Each model is evaluated on a validation set of 1,400 images, with mAP (mean average precision) as the primary performance metric. This extensive comparative analysis identifies the best performing YOLO variant for gun and weapon detection, providing valuable insights into the strengths and weaknesses of each model for this specific task.
2
Tahir, Noor Ul Ain, Zhe Long, Zuping Zhang, Muhammad Asim, and Mohammed ELAffendi. "PVswin-YOLOv8s: UAV-Based Pedestrian and Vehicle Detection for Traffic Management in Smart Cities Using Improved YOLOv8." Drones 8, no. 3 (2024): 84. http://dx.doi.org/10.3390/drones8030084.
Full textAbstract:
In smart cities, effective traffic congestion management hinges on adept pedestrian and vehicle detection. Unmanned Aerial Vehicles (UAVs) offer a solution with mobility, cost-effectiveness, and a wide field of view, and yet, optimizing recognition models is crucial to surmounting challenges posed by small and occluded objects. To address these issues, we utilize the YOLOv8s model and a Swin Transformer block and introduce the PVswin-YOLOv8s model for pedestrian and vehicle detection based on UAVs. Firstly, the backbone network of YOLOv8s incorporates the Swin Transformer model for global feature extraction for small object detection. Secondly, to address the challenge of missed detections, we opt to integrate the CBAM into the neck of the YOLOv8. Both the channel and the spatial attention modules are used in this addition because of how well they extract feature information flow across the network. Finally, we employ Soft-NMS to improve the accuracy of pedestrian and vehicle detection in occlusion situations. Soft-NMS increases performance and manages overlapped boundary boxes well. The proposed network reduced the fraction of small objects overlooked and enhanced model detection performance. Performance comparisons with different YOLO versions ( for example YOLOv3 extremely small, YOLOv5, YOLOv6, and YOLOv7), YOLOv8 variants (YOLOv8n, YOLOv8s, YOLOv8m, and YOLOv8l), and classical object detectors (Faster-RCNN, Cascade R-CNN, RetinaNet, and CenterNet) were used to validate the superiority of the proposed PVswin-YOLOv8s model. The efficiency of the PVswin-YOLOv8s model was confirmed by the experimental findings, which showed a 4.8% increase in average detection accuracy (mAP) compared to YOLOv8s on the VisDrone2019 dataset.
3
Tresnawati, Dewi, Shopi Nurhidayanti, and Nina Lestari. "A Comparison of YOLOv8 Series Performance in Student Facial Expressions Detection on Online Learning." Jurnal Online Informatika 10, no. 1 (2025): 93–104. https://doi.org/10.15575/join.v10i1.1390.
Full textAbstract:
Student engagement in online learning is an important factor that can affect learning outcomes. One indicator of engagement is facial expression. However, research on facial expression detection in online learning environments is still limited, especially in the use of the YOLOv8 algorithm. This study aims to compare the performance of several YOLOv8 variants, namely YOLOv8x, YOLOv8m, YOLOv8s, YOLOv8n, and YOLOv8l in recognizing six facial expressions: happy, sad, angry, surprised, afraid, and neutral. Student facial expression data was collected through the Moodle platform every 15 seconds during the learning process. All models were trained using 640x640 pixel images for 100 epochs to improve facial expression detection capabilities. The main contribution of this study is to provide a comprehensive analysis of the effectiveness of YOLOv8 in detecting student facial expressions, which can be used to improve the online learning experience. The evaluation results show that the YOLOv8s model has the best performance with the highest mAP of 0.840 and the fastest inference speed of 2.4 ms per image. YOLOv8m and YOLOv8x also performed well with mAP of 0.816 and 0.815, respectively. Although YOLOv8x had the slowest inference speed, it was superior in detecting fear, happiness, and sadness expressions with mAP above 0.9. YOLOv8n had mAP of 0.636, while YOLOv8l achieved mAP of 0.813 with an inference speed of 9.1 ms per image. This study shows that the YOLOv8 algorithm, especially YOLOv8s, can be an effective solution to analyze student engagement based on their facial expressions during online learning.
4
Wulanningrum, Resty, Anik Nur Handayani, and Aji Prasetya Wibawa. "Perbandingan Instance Segmentation Image Pada Yolo8." Jurnal Teknologi Informasi dan Ilmu Komputer 11, no. 4 (2024): 753–60. http://dx.doi.org/10.25126/jtiik.1148288.
Full textAbstract:
Seorang pejalan kaki sangat rawan terhadap kecelakaan di jalan. Deteksi pejalan kaki merupakan salah satu cara untuk mengidentifikasi atau megklasifikasikan antara orang, jalan atau yang lainnya. Instance segmentation adalah salah satu proses untuk melakukan segmentasi antara orang dan jalan. Instance segmentation dan penggunaan yolov8 merupakan salah satu implementasi dalam deteksi pejalan kaki. Perbandingan segmentasi pada dataset Penn-Fundan Database menggunakan yolov8 dengan model yolov8n-seg, yolov8s-seg, yolov8m-seg, yolov8l-seg, yolov8x-seg. Penelitian ini menggunakan dataset publik pedestrian atau pejalan kaki dengan objek multi person yang diambil dari dataset Penn-Fudan Database. Dataset mempunyai 2 kelas, yaitu orang dan jalan. Hasil perbandingan penggunaan model yolov8 model segmentasi yang terbaik adalah menggunakan model yolov8l-seg. Hasil penelitian didapatkan Instance segmentation valid box pada data orang, mAP50 tertinggi pada yolov8l-seg dengan nilai 0,828 dan mAP50-95 adalah 0,723. Instance segmentation valid mask pada orang nilai mAP50 tertinggi pada yolov8l-seg dengan nilai 0,825 dan mAP50-95 adalah 0,645. Pada penelitian ini, yolov8l-seg menjadi nilai terbaik dibandingkan versi yang lain, karena berdasarkan nilai mAP tertinggi pada valid mask sebesar 0,825. Abstract A pedestrian is very vulnerable to road accidents. Pedestrian detection is one way to identify or classify between people, roads or others. Instance segmentation is one of the processes to segment people and roads. Instance segmentation and the use of yolov8 is one of the implementations in pedestrian detection. Comparison of segmentation on Penn-Fundan Database dataset using yolov8 with yolov8n-seg, yolov8s-seg, yolov8m-seg, yolov8l-seg, yolov8x-seg models. This research uses a public pedestrian dataset with multi-person objects taken from the Penn-Fudan Database dataset. The dataset has 2 classes, namely people and roads. The results of the comparison using the yolov8 model, the best segmentation model is using the yolov8l-seg model. The results obtained Instance segmentation valid box on people data, the highest mAP50 on yolov8l-seg with a value of 0.828 and mAP50-95 is 0.723. Instance segmentation valid mask on people the highest mAP50 value on yolov8l-seg with a value of 0.825 and mAP50-95 is 0.645. In his study, yolov8l-seg is the best value compared to other versions, because based on the highest mAP value on the valid mask of 0.825.
5
Panja, Eben, Hendry Hendry, and Christine Dewi. "YOLOv8 Analysis for Vehicle Classification Under Various Image Conditions." Scientific Journal of Informatics 11, no. 1 (2024): 127–38. http://dx.doi.org/10.15294/sji.v11i1.49038.
Full textAbstract:
Purpose: The purpose of this research is to detect vehicle types in various image conditions using YOLOv8n, YOLOv8s, and YOLOv8m with augmentation.Methods: This research utilizes the YOLOv8 method on the DAWN dataset. The method involves using pre-trained Convolutional Neural Networks (CNN) to process the images and output the bounding boxes and classes of the detected objects. Additionally, data augmentation applied to improve the model's ability to recognize vehicles from different directions and viewpoints.Result: The mAP values for the test results are as follows: Without data augmentation, YOLOv8n achieved approximately 58%, YOLOv8s scored around 68.5%, and YOLOv8m achieved roughly 68.9%. However, after applying horizontal flip data augmentation, YOLOv8n's mAP increased to about 60.9%, YOLOv8s improved to about 62%, and YOLOv8m excelled with a mAP of about 71.2%. Using horizontal flip data augmentation improves the performance of all three YOLOv8 models. The YOLOv8m model achieves the highest mAP value of 71.2%, indicating its high effectiveness in detecting objects after applying horizontal flip augmentation. Novelty: This research introduces novelty by employing the latest version of YOLO, YOLOv8, and comparing its performance with YOLOv8n, YOLOv8s, and YOLOv8m. The use of data augmentation techniques, such as horizontal flip, to increase data variation is also novel in expanding the dataset and improving the model's ability to recognize objects.
6
Khusainov, R. M. "Selection of a Neural Network Model Based on the Hierarchy Process Analysis Method." Vestnik NSU. Series: Information Technologies 22, no. 4 (2025): 62–70. https://doi.org/10.25205/1818-7900-2024-22-4-62-70.
Full textAbstract:
This article discusses the selection of the optimal neural network model YOLOv8 (YOLOv8s, YOLOv8l, YOLOv8x, YOLOv8m, YOLOv8n) using the hierarchy process analysis method, which allows structuring and systematizing complex decisions based on multi-criteria assessments. The main focus is on identifying and comparative analysis of the most significant criteria for assessing the effectiveness of neural network models, such as training time, as well as the Precision, Recall and F1-score metrics. These metrics play a key role in computer vision tasks, especially when it comes to object detection. During the study, pairwise comparison matrices were constructed, which allow not only to visually represent the relative importance of each of the selected parameters, but also to quantitatively assess their impact on the overall effectiveness of the model. The process of forming pairwise comparison matrices includes the opinion of experts in the field of machine learning and computer vision, which ensures a high degree of reliability of the results. After processing the data and performing calculations, including weighting each criterion, priorities for alternative YOLOv8 models were derived. As a result of the calculations, it was revealed that the YOLOv8n neural network model has the highest priority among all the alternatives evaluated. This emphasizes its superiority compared to other modifications of this model.
7
Podder, Soumyajit, Abhishek Mallick, Sudipta Das, Kartik Sau, and Arijit Roy. "Accurate diagnosis of liver diseases through the application of deep convolutional neural network on biopsy images." AIMS Biophysics 10, no. 4 (2023): 453–81. http://dx.doi.org/10.3934/biophy.2023026.
Full textAbstract:
<abstract> <p>Accurate detection of non-alcoholic fatty liver disease (NAFLD) through biopsies is challenging. Manual detection of the disease is not only prone to human error but is also time-consuming. Using artificial intelligence and deep learning, we have successfully demonstrated the issues of the manual detection of liver diseases with a high degree of precision. This article uses various neural network-based techniques to assess non-alcoholic fatty liver disease. In this investigation, more than five thousand biopsy images were employed alongside the latest versions of the algorithms. To detect prominent characteristics in the liver from a collection of Biopsy pictures, we employed the YOLOv3, Faster R-CNN, YOLOv4, YOLOv5, YOLOv6, YOLOv7, YOLOv8, and SSD models. A highlighting point of this paper is comparing the state-of-the-art Instance Segmentation models, including Mask R-CNN, U-Net, YOLOv5 Instance Segmentation, YOLOv7 Instance Segmentation, and YOLOv8 Instance Segmentation. The extent of severity of NAFLD and non-alcoholic steatohepatitis was examined for liver cell ballooning, steatosis, lobular, and periportal inflammation, and fibrosis. Metrics used to evaluate the algorithms' effectiveness include accuracy, precision, specificity, and recall. Improved metrics are achieved by optimizing the hyperparameters of the associated models. Additionally, the liver is scored in order to analyse the information gleaned from biopsy images. Statistical analyses are performed to establish the statistical relevance in evaluating the score for different zones.</p> </abstract>
8
De Dios Garcia, Julio Cesar, Nantai Nava Nolazco, Ernesto Monroy Cruz, et al. "Comparative study of Convolutional Neural Networks performance and efficiency with YOLOv8 models applied for pest detection purposes in bean plants." International Journal of Combinatorial Optimization Problems and Informatics 16, no. 2 (2025): 112–22. https://doi.org/10.61467/2007.1558.2025.v16i2.603.
Full textAbstract:
Neural Networks have significantly evolved, particularly in their application to computer vision. This paper presents a comprehensive comparison of different versions of YOLOv8, such as YOLOv8n, YOLOv8s, YOLOv8m, YOLOv8l, and YOLOv8x for the detection of pests in bean plants, leveraging the capabilities of Convolutional Neural Networks. To train the neural network using different versions of YOLOv8, identical conditions were applied, such as the amount of environment light, the number of labeled images, epochs, and batch size. The results indicate that, as the complexity of the YOLO model increases, the training time escalates significantly. This increase corresponds to a more detailed data processing approach in advanced models. The results also provide insight on which model emerges as the most balanced option, offering the highest precision without compromising too much on speed. One of the models achieves the highest precision, making it reliable for accurate object detection but the speed is slow compared with other models. Otherwise exceptional precision makes it ideal for tasks where accurate identification is critical. The slight reduction in speed does not significantly hinder its overall performance in contexts where precision and detection distance are prioritized.
9
Liu, Yinzeng, Fandi Zeng, Hongwei Diao, et al. "YOLOv8 Model for Weed Detection in Wheat Fields Based on a Visual Converter and Multi-Scale Feature Fusion." Sensors 24, no. 13 (2024): 4379. http://dx.doi.org/10.3390/s24134379.
Full textAbstract:
Accurate weed detection is essential for the precise control of weeds in wheat fields, but weeds and wheat are sheltered from each other, and there is no clear size specification, making it difficult to accurately detect weeds in wheat. To achieve the precise identification of weeds, wheat weed datasets were constructed, and a wheat field weed detection model, YOLOv8-MBM, based on improved YOLOv8s, was proposed. In this study, a lightweight visual converter (MobileViTv3) was introduced into the C2f module to enhance the detection accuracy of the model by integrating input, local (CNN), and global (ViT) features. Secondly, a bidirectional feature pyramid network (BiFPN) was introduced to enhance the performance of multi-scale feature fusion. Furthermore, to address the weak generalization and slow convergence speed of the CIoU loss function for detection tasks, the bounding box regression loss function (MPDIOU) was used instead of the CIoU loss function to improve the convergence speed of the model and further enhance the detection performance. Finally, the model performance was tested on the wheat weed datasets. The experiments show that the YOLOv8-MBM proposed in this paper is superior to Fast R-CNN, YOLOv3, YOLOv4-tiny, YOLOv5s, YOLOv7, YOLOv9, and other mainstream models in regards to detection performance. The accuracy of the improved model reaches 92.7%. Compared with the original YOLOv8s model, the precision, recall, mAP1, and mAP2 are increased by 10.6%, 8.9%, 9.7%, and 9.3%, respectively. In summary, the YOLOv8-MBM model successfully meets the requirements for accurate weed detection in wheat fields.
10
Sun, Daozong, Kai Zhang, Hongsheng Zhong, et al. "Efficient Tobacco Pest Detection in Complex Environments Using an Enhanced YOLOv8 Model." Agriculture 14, no. 3 (2024): 353. http://dx.doi.org/10.3390/agriculture14030353.
Full textAbstract:
Due to the challenges of pest detection in complex environments, this research introduces a lightweight network for tobacco pest identification leveraging enhancements in YOLOv8 technology. Using YOLOv8 large (YOLOv8l) as the base, the neck layer of the original network is replaced with an asymptotic feature pyramid network (AFPN) network to reduce model parameters. A SimAM attention mechanism, which does not require additional parameters, is incorporated to improve the model’s ability to extract features. The backbone network’s C2f model is replaced with the VoV-GSCSP module to reduce the model’s computational requirements. Experiments show the improved YOLOv8 model achieves high overall performance. Compared to the original model, model parameters and GFLOPs are reduced by 52.66% and 19.9%, respectively, while mAP@0.5 is improved by 1%, recall by 2.7%, and precision by 2.4%. Further comparison with popular detection models YOLOv5 medium (YOLOv5m), YOLOv6 medium (YOLOv6m), and YOLOv8 medium (YOLOv8m) shows the improved model has the highest detection accuracy and lightest parameters for detecting four common tobacco pests, with optimal overall performance. The improved YOLOv8 detection model proposed facilitates precise, instantaneous pest detection and recognition for tobacco and other crops, securing high-accuracy, comprehensive pest identification.
11
Hanna, Arini Parhusip, Trihandaru Suryasatriya, Indrajaya Denny, and Labadin Jane. "Implementation of YOLOv8-seg on store products to speed up the scanning process at point of sales." IAES International Journal of Artificial Intelligence (IJ-AI) 13, no. 3 (2024): 3291–305. https://doi.org/10.11591/ijai.v13.i3.pp3291-3305.
Full textAbstract:
You only look once v8 (YOLOv8)-seg and its variants are implemented to accelerate the collection of goods for a store for selling activity in Indonesia. The method used here is object detection and segmentation of these objects, a combination of detection and segmentation called instance segmentation. The novelty lies in the customization and optimization of YOLOv8-seg for detecting and segmenting 30 specific Indonesian products. The use of augmented data (125 images augmented into 1,250 images) enhances the model's ability to generalize and perform well in various scenarios. The small number of data points and the small number of epochs have proven reliable algorithms to implement on store products instead of using QR codes in a digital manner. Five models are examined, i.e., YOLOv8-seg, YOLOv8s-seg, YOLOv8m-seg, YOLOv8l-seg, and YOLOv8x-seg, with a data distribution of 64% for the training dataset, 16% for the validation dataset, and 20% for the testing dataset. The best model, YOLOv8l-seg, was obtained with the highest mean average precision (mAP)box value of 99.372% and a mAPmask value of 99.372% from testing the testing dataset. However, the YOLOv8m seg model can be the best alternative model with a mAPbox value of 99.330% since the number of parameters and the computational speed are the best compared to other models.
12
Çakmakçı, Cihan. "Dijital Hayvancılıkta Yapay Zekâ ve İnsansız Hava Araçları: Derin Öğrenme ve Bilgisayarlı Görme İle Dağlık ve Engebeli Arazide Kıl Keçisi Tespiti, Takibi ve Sayımı." Turkish Journal of Agriculture - Food Science and Technology 12, no. 7 (2024): 1162–73. http://dx.doi.org/10.24925/turjaf.v12i7.1162-1173.6701.
Full textAbstract:
Küresel gıda talebindeki hızlı artış nedeniyle yüksek kaliteli hayvansal ürün üretiminin artırılması gerekliliği, modern hayvancılık uygulamalarında teknoloji kullanımı ihtiyacını beraberinde getirmiştir. Özellikle ekstansif koşullarda küçükbaş hayvan yetiştiriciliğinde hayvanların otomatik olarak izlenmesi ve yönetilmesi, verimliliğin artırılması açısından büyük öneme sahiptir. Bu noktada, insansız hava araçlarından elde edilen yüksek çözünürlüklü görüntüler ile derin öğrenme algoritmalarının birleştirilmesi, sürülerin uzaktan takip edilmesinde etkili çözümler sunma potansiyeli taşımaktadır. Bu çalışmada, insansız hava araçlarından (İHA) elde edilen yüksek çözünürlüklü görüntüler üzerinde derin öğrenme algoritmaları kullanılarak kıl keçilerinin otomatik olarak tespit edilmesi, takip edilmesi ve sayılması amaçlanmıştır. Bu kapsamda, en güncel You Only Look Once (YOLOv8) mimari varyasyonlarından YOLOv8n, YOLOv8s, YOLOv8m, YOLOv8l ve YOLOv8x olmak üzere beş farklı model gerçek hayvan izleme uçuşlarından elde edilen İHA görüntüleri üzerinde eğitilmiştir. Elde edilen bulgulara göre, 0,95 F1 skoru ve 0,99 mAP50 değeri ile hem sınırlayıcı kutu tespiti hem de segmentasyon performansı açısından en yüksek başarımı YOLOv8s mimarisi göstermiştir. Sonuç olarak, önerilen derin öğrenme tabanlı yaklaşımın, İHA destekli hassas hayvancılık uygulamalarında etkili, düşük maliyetli ve sürdürülebilir bir çözüm olabileceği öngörülmektedir.
13
Arini Parhusip, Hanna, Suryasatriya Trihandaru, Denny Indrajaya, and Jane Labadin. "Implementation of YOLOv8-seg on store products to speed up the scanning process at point of sales." IAES International Journal of Artificial Intelligence (IJ-AI) 13, no. 3 (2024): 3291. http://dx.doi.org/10.11591/ijai.v13.i3.pp3291-3305.
Full textAbstract:
<p>You only look once v8 (YOLOv8)-seg and its variants are implemented to accelerate the collection of goods for a store for selling activity in Indonesia. The method used here is object detection and segmentation of these objects, a combination of detection and segmentation called instance segmentation. The novelty lies in the customization and optimization of YOLOv8-seg for detecting and segmenting 30 specific Indonesian products. The use of augmented data (125 images augmented into 1,250 images) enhances the model's ability to generalize and perform well in various scenarios. The small number of data points and the small number of epochs have proven reliable algorithms to implement on store products instead of using QR codes in a digital manner. Five models are examined, i.e., YOLOv8-seg, YOLOv8s-seg, YOLOv8m-seg, YOLOv8l-seg, and YOLOv8x-seg, with a data distribution of 64% for the training dataset, 16% for the validation dataset, and 20% for the testing dataset. The best model, YOLOv8l-seg, was obtained with the highest mean average precision (mAP) box value of 99.372% and a mAPmask value of 99.372% from testing the testing dataset. However, the YOLOv8mseg model can be the best alternative model with a mAPbox value of 99.330% since the number of parameters and the computational speed are the best compared to other models.</p>
14
Althaf Pramasetya Perkasa, Mochamad, R. Reza El Akbar, Muhammad Al Husaini, and Randi Rizal. "VISUAL ENTITY OBJECT DETECTION SYSTEM IN SOCCER MATCHES BASED ON VARIOUS YOLO ARCHITECTURE." Jurnal Teknik Informatika (Jutif) 5, no. 3 (2024): 811–20. https://doi.org/10.52436/1.jutif.2024.5.3.2015.
Full textAbstract:
In this study, a performance comparison between the YOLOv7, YOLOv8, and YOLOv9 models in identifying objects in soccer matches is conducted. Parameter adjustments based on GPU storage capacity were also evaluated. The results show that YOLOv8 performs better, with higher precision, recall, and F1-score values, especially in the "Ball" class, and an overall accuracy (mAP@0.5) of 87.4%. YOLOv9 also performs similarly to YOLOv8, but YOLOv8's higher mAP@0.5 value shows its superiority in detecting objects with varying degrees of confidence. Both models show significant improvement compared to YOLOv7 in overall object detection performance. Therefore, based on these results, YOLOv8 can be considered as the model that is close to the best performance in detecting objects in the dataset used. This study not only provides insights into the performance and characteristics of the YOLOv7, YOLOv8, and YOLOv9 models in the context of object detection in soccer matches but also results in a dataset ready for additional analysis or for training deep learning models.
15
KHUSAINOV, R. M. "ROAD INFRASTRUCTURE OBJECT RECOGNITION SOFTWARE PACKAGE." Herald of Technological University 28, no. 2 (2025): 61–65. https://doi.org/10.55421/1998-7072_2025_28_2_61.
Full textAbstract:
The article considers the problem of recognizing road infrastructure objects using the developed mobile application. The necessity of using YOLOv8 neural network models to solve this problem is substantiated. When implementing the mobile application, the Kotlin programming language and the Android Studio 2024.2.1 development environment were chosen. The Driver Assistant mobile application was developed on a laptop with an Intel Pentium CPU 3825U processor with a frequency of 1.9 GHz, 8 GB RAM, running the 64-bit Windows 10 Pro operating system. The main components of the mobile application are the camera module, the model selection module, and the experimental research module. The camera module includes permission to take photos or videos. The model selection module includes trained YOLOv8 neural network models (YOLOv8n, YOLOv8l, YOLOv8m, YOLOv8s, YOLOv8x). The experimental research module includes real-time object recognition and obtaining the result of its probability assessment. The main window of the mobile application includes the following tabs: «select YOLOv8 model modification», «processing time» (contains the object recognition time in milliseconds); «confidence threshold» (a value from 0 to 1 is set for the model's confidence that an object of a certain class is present in a given area of the image or video frame); «IoU (Jaccard coefficient)» (a value from 0 to 1 is set for the overlap between the predicted rectangle and the true frame). Examples of the mobile application functioning in real time are given. In the future, it is advisable to improve the «Driver Assistant» mobile application, as well as implement and use it in practice in companies producing land vehicles (cars and trucks), as well as public transport (buses, trolleybuses).
16
Salma, Kartika, and Syarif Hidayat. "Deteksi Antusiasme Siswa dengan Algoritma Yolov8 pada Proses Pembelajaran Daring." Jurnal Indonesia : Manajemen Informatika dan Komunikasi 5, no. 2 (2024): 1611–18. http://dx.doi.org/10.35870/jimik.v5i2.716.
Full textAbstract:
The implementation of Face Emotion Recognition (FER) technology in online classes opens up new opportunities to effectively monitor students' emotional responses and adjust the teaching approach. Through FER, instructors can monitor students' emotional responses to learning materials in real-time and enable quick adjustments based on individual needs. Additionally, this technology can also be used to detect the level of enthusiasm or lack thereof among students towards the learning process, allowing for the optimization of teaching strategies. This study focuses on the implementation of the YOLOv8 algorithm in detecting students' enthusiasm, comparing the performance of YOLOv8n, YOLOv8s, YOLOv8m, and YOLOv8l models. Test results show that YOLOv8n performs the best with an accuracy rate of 95.3% and a fast inference time of 62ms, enabling real-time object detection. Thus, the application of YOLOv8 in this context aims to detect students' enthusiasm in real-time and allows instructors to quickly adjust their approach to meet students' needs. Furthermore, this research contributes to improving the quality of online learning by providing insights into students' emotional engagement and serving as a tool to help instructors better understand and respond appropriately to students' emotions during the online learning process.
17
Lou, Haitong, Xuehu Duan, Junmei Guo, et al. "DC-YOLOv8: Small-Size Object Detection Algorithm Based on Camera Sensor." Electronics 12, no. 10 (2023): 2323. http://dx.doi.org/10.3390/electronics12102323.
Full textAbstract:
Traditional camera sensors rely on human eyes for observation. However, human eyes are prone to fatigue when observing objects of different sizes for a long time in complex scenes, and human cognition is limited, which often leads to judgment errors and greatly reduces efficiency. Object recognition technology is an important technology used to judge the object’s category on a camera sensor. In order to solve this problem, a small-size object detection algorithm for special scenarios was proposed in this paper. The advantage of this algorithm is that it not only has higher precision for small-size object detection but also can ensure that the detection accuracy for each size is not lower than that of the existing algorithm. There are three main innovations in this paper, as follows: (1) A new downsampling method which could better preserve the context feature information is proposed. (2) The feature fusion network is improved to effectively combine shallow information and deep information. (3) A new network structure is proposed to effectively improve the detection accuracy of the model. From the point of view of detection accuracy, it is better than YOLOX, YOLOR, YOLOv3, scaled YOLOv5, YOLOv7-Tiny, and YOLOv8. Three authoritative public datasets are used in these experiments: (a) In the Visdron dataset (small-size objects), the map, precision, and recall ratios of DC-YOLOv8 are 2.5%, 1.9%, and 2.1% higher than those of YOLOv8s, respectively. (b) On the Tinyperson dataset (minimal-size objects), the map, precision, and recall ratios of DC-YOLOv8 are 1%, 0.2%, and 1.2% higher than those of YOLOv8s, respectively. (c) On the PASCAL VOC2007 dataset (normal-size objects), the map, precision, and recall ratios of DC-YOLOv8 are 0.5%, 0.3%, and 0.4% higher than those of YOLOv8s, respectively.
18
Taufiqurrahman, Taufiqurrahman, Aji Prasetya Hadi, and Rully Emirza Siregar. "Evaluasi Performa Yolov8 Dalam Deteksi Objek Di Depan Kendaraan Dengan Variasi Kondisi Lingkungan." Jurnal Minfo Polgan 13, no. 2 (2024): 1755–73. http://dx.doi.org/10.33395/jmp.v13i2.14228.
Full textAbstract:
Keselamatan berlalu lintas adalah isu global yang membutuhkan perhatian serius mengingat tingginya angka kecelakaan setiap tahun. Penelitian ini bertujuan mengevaluasi performa YOLOv8, algoritma deteksi objek berbasis deep learning, dalam mendeteksi elemen-elemen penting lalu lintas seperti kendaraan, pejalan kaki, dan rambu lalu lintas. Dataset yang digunakan terdiri dari video jalan biasa dan jalan tol, direkam pada enam waktu berbeda (08:00, 10:00, 12:00, 18:00, 20:00, dan 22:00) untuk menangkap variasi pencahayaan dan kepadatan lalu lintas. Tiga varian YOLOv8, yaitu YOLOv8n, YOLOv8s, dan YOLOv8m, diuji untuk menganalisis akurasi, jumlah deteksi, dan performa dalam berbagai kondisi lingkungan. Hasil penelitian menunjukkan bahwa YOLOv8m memiliki performa terbaik dengan rata-rata confidence score tertinggi, khususnya pada kondisi pencahayaan optimal di siang hari. YOLOv8s menawarkan keseimbangan antara efisiensi dan akurasi, sedangkan YOLOv8n menunjukkan keterbatasan dalam mendeteksi objek pada kondisi pencahayaan rendah dan kompleksitas lingkungan yang tinggi. Jalan tol, dengan lingkungan yang lebih terstruktur, memberikan hasil deteksi yang lebih konsisten dibandingkan jalan biasa yang menghadirkan tantangan berupa variasi objek dan pencahayaan. Kesimpulannya, YOLOv8m adalah model yang paling efektif untuk aplikasi berbasis keselamatan lalu lintas, sementara YOLOv8n cocok untuk perangkat keras dengan sumber daya terbatas. Penelitian selanjutnya diharapkan dapat mengoptimalkan deteksi pada objek kecil dan meningkatkan performa di kondisi pencahayaan rendah melalui pelatihan ulang model menggunakan dataset yang lebih kompleks.
19
Gong, Chuang, Wei Jiang, Dehua Zou, Weiwei Weng, and Hongjun Li. "An Insulator Fault Diagnosis Method Based on Multi-Mechanism Optimization YOLOv8." Applied Sciences 14, no. 19 (2024): 8770. http://dx.doi.org/10.3390/app14198770.
Full textAbstract:
Aiming at the problem that insulator image backgrounds are complex and fault types are diverse, which makes it difficult for existing deep learning algorithms to achieve accurate insulator fault diagnosis, an insulator fault diagnosis method based on multi-mechanism optimization YOLOv8-DCP is proposed. Firstly, a feature extraction and fusion module, named CW-DRB, was designed. This module enhances the C2f structure of YOLOv8 by incorporating the dilation-wise residual module and the dilated re-param module. The introduction of this module improves YOLOv8’s capability for multi-scale feature extraction and multi-level feature fusion. Secondly, the CARAFE module, which is feature content-aware, was introduced to replace the up-sampling layer in YOLOv8n, thereby enhancing the model’s feature map reconstruction ability. Finally, an additional small-object detection layer was added to improve the detection accuracy of small defects. Simulation results indicate that YOLOv8-DCP achieves an accuracy of 97.7% and an mAP@0.5 of 93.9%. Compared to YOLOv5, YOLOv7, and YOLOv8n, the accuracy improved by 1.5%, 4.3%, and 4.8%, while the mAP@0.5 increased by 3.0%, 4.3%, and 3.1%. This results in a significant enhancement in the accuracy of insulator fault diagnosis.
20
Luo, Qinpeng, and Yinhua Liao. "Detecting Algorithm based on the Improved YOLOv8s for a Weak Feature Defect of Aviation Clamps." International Journal of Mechanical and Electrical Engineering 4, no. 3 (2025): 21–37. https://doi.org/10.62051/ijmee.v4n3.03.
Full textAbstract:
There are some weak defects on the surface of aviation clamps. Because they are very weak, it is difficult to identify them efficiently and accurately by the existing visual detecting algorithms, and the existing methods have high arithmetic power requirements for vision detecting systems. So, this work proposes a detecting algorithm for a weak feature defect detection of aviation clamps (YOLO-OGS). Firstly, in order to improve the ability of convolutional operations of extract features and decrease the model's GFLOPs, the Multidimensional dynamic convolutional ODConv is added to the backbone network of YOLOv8. Then, in order to reduce the complexity of the model while increasing the effectiveness of feature fusion at various levels by keeping more of the hidden connections in the channels, the GhostSlimFPN paradigm network structure, which contains GSConv convolution and slim-neck structure, is introduced in the neck network. Finally, the Shuffle Attention module is used to widen the image's sensory field and enhance the details of weak flaws. Based on the aviation clamp defect data set, the comparative analysis results of YOLO-OGS and YOLOv8s algorithms show that YOLO-OGS decreases the GFLOPs by 14.4% and increases precision, recall, mAP@0.5, and GFLOPs by 4%, 6.4%, and 3.4%, respectively. And compared with the other existing mainstream networks YOLOv6, YOLOv8s, YOLOv8n, YOLOv5n, YOLOv5s, YOLOv7, YOLOv3-tiny. 8.3%, 3.4%, 4.4%, 15.4%, 8%, 13.4%, and 12% improvement in mAP@0.5.
21
Kestane, Bahadir Besir, Emin Guney, and Cuneyt Bayilmis. "Real-time Recyclable Waste Detection Using YOLOv8 for Reverse Vending Machines." Jurnal Ilmiah Teknik Elektro Komputer dan Informatika 10, no. 2 (2024): 345–58. https://doi.org/10.26555/jiteki.v10i2.29208.
Full textAbstract:
Increasing challenges in waste management necessitate optimizing the efficiency of recycling systems. Reverse Vending Machines (RVMs) offer a promising solution by incentivizing recycling through user rewards. However, inaccurate waste detection methods hinder the effectiveness of RVMs. This study explores the potential of the YOLOv8 deep learning algorithm to enhance real-time waste classification accuracy in RVMs. We propose a YOLOv8-based framework for real-time detection of seven key recyclable materials. The model is trained on a combined dataset comprising the public TrashNet dataset and a study-specific dataset tailored to materials and variations encountered in RVMs. Performance evaluation metrics include F1-score, precision, recall, and PR curves.Results demonstrate the superior performance of the YOLOv8-based approach compared to other popular deep learning algorithms, including YOLOv5, YOLOv7, and YOLOv9. The YOLOv8 model achieves an accuracy rate of over 97%, significantly outperforming other algorithms. This improvement translates into enhanced recycling efficiency and reduced misclassification errors in RVMs. This research contributes to the development of more sustainable waste management systems by improving the efficiency and accuracy of RVMs. The YOLOv8-based framework presents a promising solution for real-time waste detection in RVMs, paving the way for more effective recycling practices and reduced environmental impact.
22
Richo, Richo. "Analisis Keandalan YOLOv8m untuk Deteksi Varian Produk Kemasan Kotak pada Sistem Manajemen Kesediaan Stock." Informatics and Digital Expert (INDEX) 6, no. 2 (2024): 124–31. https://doi.org/10.36423/index.v6i2.1984.
Full textAbstract:
Sistem manajemen kesediaan stock secara otomatis dapat meningkatkan efisiensi waktu ketika proses pendataan produk, mempercepat laju penjualan, efisiensi budget, hingga meminimalisir terjadinya tindakan kecurangan pendataan oleh petugas toko. Demi terciptanya tujuan tersebut sudah seharusnya sistem pendataan kesediaan stock dioptimalisasi dengan sistem otomatisasi. Sayangnya, sistem pendataan saat ini masih dilakukan secara manual dengan basis pengecekan produk pada display shelfing sehingga memicu terjadinya kesalahan pendataan oleh petugas, tentunya akar permasalahan ini harus segera diputuskan sehingga efisiensi manajemen persediaan stock dapat terorganisir. Sistem deteksi real-time menggunakan metode YOLO telah beberapa kali dilakukan dan terbukti telah berkontribusi positif pada keakuratan hasil deteksi, salah satunya yakni metode YOLOv8. Pada penelitian ini menggunakan komparasi 4 metode YOLOv8 diantaranya yakni YOLOv8n, YOLOv8s, YOLOv8m, dan YOLOv8l, sebagai langkah untuk menciptakan akurasi model terbaik. Produk deteksi yang digunakan pada penelitian ini menggunakan produk “Pocky 70gram Biscuit Sticks” dengan total varian produk sebanyak 8 class. Penelitian ini menggunakan 400 data training dan 400 data validation. Hasil temuan dari penelitian ini yakni metode YOLOv8m memberikan kontribusi akurasi training terbaik diantara model lainnya dengan persentase precision mencapai 95.45% dan mAP50 sebesar 96.32%. Adanya penyempurnaan pada model YOLOv8m yang telah dilakukan berhasil meningkatkan precision of accuracy model sebesar 97.85% dari sebelumnya yang hanya 95.45%. Hasil persentase average confidence score pada penelitian ini sebesar 86.08% atas kontribusi pengujian keseluruhan varian produk yang diujikan. Keandalan sistem deteksi kesediaan stock ini menunjukkan bahwa sistem application mampu mengenali setiap jenis produk secara tepat dan efektif sehingga output pada penelitian ini dapat bermanfaat untuk efisiensi manajemen kesediaan stock.
23
Asaju, Christine Bukola, Pius Adewale Owolawi, Chuling Tu, and Etienne Van Wyk. "Cloud-Based License Plate Recognition: A Comparative Approach Using You Only Look Once Versions 5, 7, 8, and 9 Object Detection." Information 16, no. 1 (2025): 57. https://doi.org/10.3390/info16010057.
Full textAbstract:
Cloud-based license plate recognition (LPR) systems have emerged as essential tools in modern traffic management and security applications. Determining the best approach remains paramount in the field of computer vision. This study presents a comparative analysis of various versions of the YOLO (You Only Look Once) object detection models, namely, YOLO 5, 7, 8, and 9, applied to LPR tasks in a cloud computing environment. Using live video, we performed experiments on YOLOv5, YOLOv7, YOLOv8, and YOLOv9 models to detect number plates in real time. According to the results, YOLOv8 is reported the most effective model for real-world deployment due to its strong cloud performance. It achieved an accuracy of 78% during cloud testing, while YOLOv5 showed consistent performance with 71%. YOLOv7 performed poorly in cloud testing (52%), indicating potential issues, while YOLOv9 reported 70% accuracy. This tight alignment of results shows consistent, although modest, performance across scenarios. The findings highlight the evolution of the YOLO architecture and its impact on enhancing LPR accuracy and processing efficiency. The results provide valuable insights into selecting the most appropriate YOLO model for cloud-based LPR systems, balancing the trade-offs between real-time performance and detection precision. This research contributes to advancing the field of intelligent transportation systems by offering a detailed comparison that can guide future implementations and optimizations of LPR systems in cloud environments.
24
Simanjuntak, Nurchaya, Raymond Erz Saragih, and Yonky Pernando. "Utilizing Lightweight YOLOv8 Models for Accurate Determination of Ambarella Fruit Maturity Levels." Journal of Computer System and Informatics (JoSYC) 5, no. 3 (2024): 719–28. https://doi.org/10.47065/josyc.v5i3.5123.
Full textAbstract:
In the agricultural sector, accurately determining fruit ripeness remains a crucial yet challenging task. Among intriguing Indonesian fruits, the Ambarella presents a particular difficulty. In Ambarella fruit, the peel changes from green to golden yellow as it ripens, serving as a visual indicator for optimal harvest time, thus determining the maturity is crucial for harvesting the Ambarella fruit. Traditionally, ripeness assessment relies on manual methods, which suffer from drawbacks like high labor costs, significant time investment, and inconsistency in results. This work explores the potential of employing YOLOv8, a cutting-edge deep learning model, to automate Ambarella fruit ripeness classification. This work focuses on the YOLOv8n, YOLOv8s, and YOLOv8m, lightweight models within the YOLOv8 family. Our results are promising: all three models achieved 100% accuracy on the training set, with YOLOv8s demonstrating the lowest loss at 0.00286. The web application was utilised to deploy the trained models, allowing users to upload images of Ambarella fruit and run the model for inference.
25
Gong, He, Jingyi Liu, Zhipeng Li, et al. "GFI-YOLOv8: Sika Deer Posture Recognition Target Detection Method Based on YOLOv8." Animals 14, no. 18 (2024): 2640. http://dx.doi.org/10.3390/ani14182640.
Full textAbstract:
As the sika deer breeding industry flourishes on a large scale, accurately assessing the health of these animals is of paramount importance. Implementing posture recognition through target detection serves as a vital method for monitoring the well-being of sika deer. This approach allows for a more nuanced understanding of their physical condition, ensuring the industry can maintain high standards of animal welfare and productivity. In order to achieve remote monitoring of sika deer without interfering with the natural behavior of the animals, and to enhance animal welfare, this paper proposes a sika deer individual posture recognition detection algorithm GFI-YOLOv8 based on YOLOv8. Firstly, this paper proposes to add the iAFF iterative attention feature fusion module to the C2f of the backbone network module, replace the original SPPF module with AIFI module, and use the attention mechanism to adjust the feature channel adaptively. This aims to enhance granularity, improve the model’s recognition, and enhance understanding of sika deer behavior in complex scenes. Secondly, a novel convolutional neural network module is introduced to improve the efficiency and accuracy of feature extraction, while preserving the model’s depth and diversity. In addition, a new attention mechanism module is proposed to expand the receptive field and simplify the model. Furthermore, a new pyramid network and an optimized detection head module are presented to improve the recognition and interpretation of sika deer postures in intricate environments. The experimental results demonstrate that the model achieves 91.6% accuracy in recognizing the posture of sika deer, with a 6% improvement in accuracy and a 4.6% increase in mAP50 compared to YOLOv8n. Compared to other models in the YOLO series, such as YOLOv5n, YOLOv7-tiny, YOLOv8n, YOLOv8s, YOLOv9, and YOLOv10, this model exhibits higher accuracy, and improved mAP50 and mAP50-95 values. The overall performance is commendable, meeting the requirements for accurate and rapid identification of the posture of sika deer. This model proves beneficial for the precise and real-time monitoring of sika deer posture in complex breeding environments and under all-weather conditions.
26
Kutyrev, A. I., I. G. Smirnov, and N. A. Andriyanov. "Neural network models of apple fruit identification in tree crowns: comparative analysis." Horticulture and viticulture, no. 5 (November 30, 2023): 56–63. http://dx.doi.org/10.31676/0235-2591-2023-5-56-63.
Full textAbstract:
The article presents the results of an analysis conducted from 2022 to 2023 to assess the quality of modern neural network models of apple fruit identification in tree crowns shown in images. In order to conduct the studies on identifying the best detector, the following neural networks were used: SSD (Single Shot MultiBox Detector), YOLOv4 (You Only Look Once, Version 4), YOLOv5, YOLOv7, and YOLOv8. The performance of the considered models of apple fruit identification was assessed using such binary classification metrics as precision, recall, accuracy, F-score, and AUC-ROCTotal (area under the curve). To assess the accuracy in predicting apple fruit identification, the mean absolute percentage error (MAPE) of the analyzed neural network models was calculated. The neural network performance analysis used 300 photographs taken at an apple garden. The conducted studies revealed that the SSD model provides lower speed and accuracy, as well as having high requirements for computing resources, which may limit its use in lower performance devices. The YOLOv4 model surpasses the YOLOv5 model in terms of accuracy by 10.2 %, yet the processing speed of the YOLOv5 model is over twice that of the YOLOv4 model. This fact makes the YOLOv5 model preferable for tasks related to real-time big data processing. The YOLOv8 model is superior to the YOLOv7 model in terms of speed (by 37.3 %); however, the accuracy of the YOLOv7 model is 9.4 % higher. The highest area under the Precision-Recall curve amounts to 0.94 when using the YOLOv7 model. This fact suggests a high probability that the classifier can accurately distinguish between the positive and negative values of the apple fruit class. MAPE calculation for the analyzed neural network models showed that the lowest error in apple fruit identification amounted to 5.64 % for the YOLOv7 model as compared to the true value determined using the visual method. The performance analysis of modern neural network models shows that the YOLO family of neural networks provides high speed and accuracy of object detection, which allows them to operate in real time. The use of transfer learning (tuning of only the last layers to solve highly specialized problems) to adjust the performance of models for different apple fruit varieties can further improve the accuracy of apple fruit identification.
27
Zhang, Yijian, Yong Yin, and Zeyuan Shao. "An Enhanced Target Detection Algorithm for Maritime Search and Rescue Based on Aerial Images." Remote Sensing 15, no. 19 (2023): 4818. http://dx.doi.org/10.3390/rs15194818.
Full textAbstract:
Unmanned aerial vehicles (UAVs), renowned for their rapid deployment, extensive data collection, and high spatial resolution, are crucial in locating distressed individuals during search and rescue (SAR) operations. Challenges in maritime search and rescue include missed detections due to issues including sunlight reflection. In this study, we proposed an enhanced ABT-YOLOv7 algorithm for underwater person detection. This algorithm integrates an asymptotic feature pyramid network (AFPN) to preserve the target feature information. The BiFormer module enhances the model’s perception of small-scale targets, whereas the task-specific context decoupling (TSCODE) mechanism effectively resolves conflicts between localization and classification. Using quantitative experiments on a curated dataset, our model outperformed methods such as YOLOv3, YOLOv4, YOLOv5, YOLOv8, Faster R-CNN, Cascade R-CNN, and FCOS. Compared with YOLOv7, our approach enhances the mean average precision (mAP) from 87.1% to 91.6%. Therefore, our approach reduces the sensitivity of the detection model to low-lighting conditions and sunlight reflection, thus demonstrating enhanced robustness. These innovations have driven advancements in UAV technology within the maritime search and rescue domains.
28
Alayed, Asmaa, Rehab Alidrisi, Ekram Feras, Shahad Aboukozzana, and Alaa Alomayri. "Real-Time Inspection of Fire Safety Equipment using Computer Vision and Deep Learning." Engineering, Technology & Applied Science Research 14, no. 2 (2024): 13290–98. http://dx.doi.org/10.48084/etasr.6753.
Full textAbstract:
The number of accidental fires in buildings has been significantly increased in recent years in Saudi Arabia. Fire Safety Equipment (FSE) plays a crucial role in reducing fire risks. However, this equipment is prone to defects and requires periodic checks and maintenance. Fire safety inspectors are responsible for visual inspection of safety equipment and reporting defects. As the traditional approach of manually checking each piece of equipment can be time-consuming and inaccurate, this study aims to improve the inspection processes of safety equipment. Using computer vision and deep learning techniques, a detection model was trained to visually inspect fire extinguishers and identify defects. Fire extinguisher images were collected, annotated, and augmented to create a dataset of 7,633 images with 16,092 labeled instances. Then, experiments were carried out using YOLOv5, YOLOv7, YOLOv8, and RT-DETR. Pre-trained models were used for transfer learning. A comparative analysis was performed to evaluate these models in terms of accuracy, speed, and model size. The results of YOLOv5n, YOLOv7, YOLOv8n, YOLOv8m, and RT-DETR indicated satisfactory accuracy, ranging between 83.1% and 87.2%. YOLOv8n was chosen as the most suitable due to its fastest inference time of 2.7 ms, its highest mAP0.5 of 87.2%, and its compact model size, making it ideal for real-time mobile applications.
29
Sabillah, Ainul Hakim Fizal, Hafizal Mohamad, Sagaya Sabestinal Amalathas, Hwan Won Seung, and Muhammad Hakim Ahmad Sobri. "ENHANCING OIL PALM FRUIT DETECTION: A COMPARATIVE ANALYSIS OF YOLO ALGORITHMS." International Journal of Innovation and Industrial Revolution 7, no. 20 (2025): 01–18. https://doi.org/10.35631/ijirev.720001.
Full textAbstract:
The oil palm industry is pivotal in agricultural research because of its importance. The central focus of this study, however, revolves around elevating cutting-edge intelligent techniques in agriculture, specifically for the improved detection of Fresh Fruit Bunches within oil palm plantations. Moreover, Malaysia’s economic impact on oil palm production is explored, emphasizing its position as a leading producer and exporter of palm oil. The study compares and corroborates the performances among a series of YOLO algorithm models, namely YOLOv3, YOLOv4, YOLOv7, and YOLOv8, by exploiting diverse essential metrics that embrace mean Average Precision, precision, recall, and F1-score. Through the rigorous evaluations of these models, the research contributes to the precision agricultural field, underscoring the superior performance of YOLOv8 in accurately detecting FFBs and facilitating its realization of advanced computer vision techniques for optimal oil palm plantation management and enhanced productivity.
30
Sun, Jihong, Zhaowen Li, Fusheng Li, Yingming Shen, Ye Qian, and Tong Li. "EF yolov8s: A Human–Computer Collaborative Sugarcane Disease Detection Model in Complex Environment." Agronomy 14, no. 9 (2024): 2099. http://dx.doi.org/10.3390/agronomy14092099.
Full textAbstract:
The precise identification of disease traits in the complex sugarcane planting environment not only effectively prevents the spread and outbreak of common diseases but also allows for the real-time monitoring of nutrient deficiency syndrome at the top of sugarcane, facilitating the supplementation of relevant nutrients to ensure sugarcane quality and yield. This paper proposes a human–machine collaborative sugarcane disease detection method in complex environments. Initially, data on five common sugarcane diseases—brown stripe, rust, ring spot, brown spot, and red rot—as well as two nutrient deficiency conditions—sulfur deficiency and phosphorus deficiency—were collected, totaling 11,364 images and 10 high-definition videos captured by a 4K drone. The data sets were augmented threefold using techniques such as flipping and gamma adjustment to construct a disease data set. Building upon the YOLOv8 framework, the EMA attention mechanism and Focal loss function were added to optimize the model, addressing the complex backgrounds and imbalanced positive and negative samples present in the sugarcane data set. Disease detection models EF-yolov8s, EF-yolov8m, EF-yolov8n, EF-yolov7, and EF-yolov5n were constructed and compared. Subsequently, five basic instance segmentation models of YOLOv8 were used for comparative analysis, validated using nutrient deficiency condition videos, and a human–machine integrated detection model for nutrient deficiency symptoms at the top of sugarcane was constructed. The experimental results demonstrate that our improved EF-yolov8s model outperforms other models, achieving mAP_0.5, precision, recall, and F1 scores of 89.70%, 88.70%, 86.00%, and 88.00%, respectively, highlighting the effectiveness of EF-yolov8s for sugarcane disease detection. Additionally, yolov8s-seg achieves an average precision of 80.30% with a smaller number of parameters, outperforming other models by 5.2%, 1.9%, 2.02%, and 0.92% in terms of mAP_0.5, respectively, effectively detecting nutrient deficiency symptoms and addressing the challenges of sugarcane growth monitoring and disease detection in complex environments using computer vision technology.
31
Jiang, Tao, Jie Zhou, Binbin Xie, et al. "Improved YOLOv8 Model for Lightweight Pigeon Egg Detection." Animals 14, no. 8 (2024): 1226. http://dx.doi.org/10.3390/ani14081226.
Full textAbstract:
In response to the high breakage rate of pigeon eggs and the significant labor costs associated with egg-producing pigeon farming, this study proposes an improved YOLOv8-PG (real versus fake pigeon egg detection) model based on YOLOv8n. Specifically, the Bottleneck in the C2f module of the YOLOv8n backbone network and neck network are replaced with Fasternet-EMA Block and Fasternet Block, respectively. The Fasternet Block is designed based on PConv (Partial Convolution) to reduce model parameter count and computational load efficiently. Furthermore, the incorporation of the EMA (Efficient Multi-scale Attention) mechanism helps mitigate interference from complex environments on pigeon-egg feature-extraction capabilities. Additionally, Dysample, an ultra-lightweight and effective upsampler, is introduced into the neck network to further enhance performance with lower computational overhead. Finally, the EXPMA (exponential moving average) concept is employed to optimize the SlideLoss and propose the EMASlideLoss classification loss function, addressing the issue of imbalanced data samples and enhancing the model’s robustness. The experimental results showed that the F1-score, mAP50-95, and mAP75 of YOLOv8-PG increased by 0.76%, 1.56%, and 4.45%, respectively, compared with the baseline YOLOv8n model. Moreover, the model’s parameter count and computational load are reduced by 24.69% and 22.89%, respectively. Compared to detection models such as Faster R-CNN, YOLOv5s, YOLOv7, and YOLOv8s, YOLOv8-PG exhibits superior performance. Additionally, the reduction in parameter count and computational load contributes to lowering the model deployment costs and facilitates its implementation on mobile robotic platforms.
32
Ramadhani, Zahra Cahya, and Dimas Firmanda Al Riza. "Model Deteksi Mikroalga Spirulina platensis dan Chlorella vulgaris Berbasis Convolutional Neural Network YOLOv8." Jurnal Komputer dan Informatika 12, no. 2 (2024): 110–19. https://doi.org/10.35508/jicon.v12i2.15375.
Full textAbstract:
Mikroalga merupakan organisme mikroskopis bersel tunggal yang hidup di berbagai perairan. Mikroalga seperti Spirulina platensis dan Chlorella vulgaris berpotensi menjadi sumber bioenergi sehingga mulai banyak dikultivasi. Kultivasi ini umumnya masih melakukan pemantauan jumlah/kepadatan sel mikroalga secara manual menggunakan hemositometer yang lebih lama dan rentan human error. Penelitian ini bertujuan untuk mengembangkan model deteksi mikroalga Spirulina platensis dan Chlorella vulgaris berbasis citra mikroskopis dan Convolutional Neural Network menggunakan YOLOv8. Metodologi penelitian mencakup persiapan sampel (pengenceran dan pengukuran optical density), penentuan kepadatan terbaik, akuisisi citra, anotasi citra, pembuatan dataset citra, pelatihan model YOLOv8, dan evaluasi kinerja model. Penentuan kepadatan terbaik bertujuan untuk mendapatkan citra mikroskopis yang baik. Akuisisi citra dilakukan menggunakan mikroskop binokuler dan menghasilkan 560 gambar yang kemudian dianotasi. Model YOLOv8n, YOLOv8s, dan YOLOv8m dilatih dengan default hyperparameter di Google Colaboratory untuk mengetahui pengaruh augmentasi terhadap akurasi model. Evaluasi kinerja model dilakukan pada model YOLOv8 terpilih dan dianalisis nilai mAP50. Hasil penelitian menunjukkan bahwa augmentasi (crop, brightness, dan blur) menghasilkan mAP train dan test tertinggi pada model YOLOv8m, yakni 0,945 dan 0,913. Model YOLOv8m ini dilatih kembali dengan variasi hyperparameters dan didapatkan konfigurasi terbaik pada optimizer SGD, epoch 50, dan learning rate 0,01 dengan mAP train dan test sebesar 0,934 dan 0,925. Namun, training 29 epoch dapat menghasilkan akurasi 0,8535 yang memperkecil overfitting serta pemborosan sumber daya. Kesimpulannya, penelitian ini dapat mempermudah peneliti maupun industri dalam melakukan penghitungan jumlah mikroalga secara otomatis dan lebih efisien.
33
V., Mareeswari, Vijayan R., Shajith Nisthar, and Rahul Bala Krishnan. "Traffic Sign Detection and Recognition Using Yolo Models." International Journal of Information Technology and Computer Science 17, no. 3 (2025): 13–25. https://doi.org/10.5815/ijitcs.2025.03.02.
Full textAbstract:
With the proliferation of advanced driver assistance systems and continued advances in autonomous vehicle technology, there is a need for accurate, real-time methods of identifying and interpreting traffic signs. The importance of traffic sign detection can't be overstated, as it plays a pivotal role in improving road safety and traffic management. This proposed work suggests a unique real-time traffic sign detection and recognition approach using the YOLOv8 algorithm. Utilizing the integrated webcams of personal computers and laptops, we capture live traffic scenes and train our model using a meticulously curated dataset from Roboflow. Through extensive training, our YOLOv8 version achieves an excellent accuracy rate of 94% compared to YOLOV7 at 90.1% and YOLOv5 at 81.3%, ensuring reliable detection and recognition across various environmental conditions. Additionally, this proposed work introduces an auditory alert feature that notifies the driver with a voice alert upon detecting traffic signs, enhancing driver awareness and safety. Through rigorous experimentation and evaluation, we validate the effectiveness of our approach, highlighting the importance of utilizing available hardware resources to deploy traffic sign detection systems with minimal infrastructure requirements. Our findings underscore the robustness of YOLOv8 in handling challenging traffic sign recognition tasks, paving the way for widespread adoption of intelligent transportation technologies and fostering the introduction of safer and more efficient road networks. In this paper, we compare the unique model of YOLO with YOLOv5, YOLOv7, and YOLOv8, and find that YOLOv8 outperforms its predecessors, YOLOv7 and YOLOv5, in traffic sign detection with an excellent overall mean average precision of 0.945. Notably, it demonstrates advanced precision and recall, especially in essential sign classes like "No overtaking" and "Stop," making it the favored preference for accurate and dependable traffic sign detection tasks.
34
Tverdokhlebov, A. S., A. S. Krasnoperova, A. A. Kartashov, V. Y. Kuprits, and V. I. Weber. "Effectiveness of neural network models for automatic detection and recognition of ground objects in infrared images." Ural Radio Engineering Journal 8, no. 4 (2024): 451–68. https://doi.org/10.15826/urej.2024.8.4.003.
Full textAbstract:
In this paper the study of modern neural network models designed for object recognition in infrared images acquired from unmanned aerial vehicles (UAVs) is carried out. Different YOLO (You Only Look Once) architectures including YOLOv5, YOLOv8 and YOLOv9 versions are considered. Models are evaluated on key metrics such as Precision (Precision), completeness (Recall) and mean accuracy (mAP), taking into account computational requirements. Particular attention is paid to the application of the models under resource constraints and increased data complexity, making them relevant for monitoring and analysis tasks in complex operating environments. The analysis of the effectiveness of neural network models for search and rescue operations has been conducted. The study demonstrates that the best results in object classification precision have been obtained by the YOLOv8l and YOLOv5mu models, with corresponding values of 0.912 and 0.911. The highest recall has been achieved by the YOLOv8s and YOLOv9c models, with results of 0.836 and 0.827. In terms of the mAP50–95 metric, the YOLOv9c and YOLOv8l-worldv2 models have the best performance, with scores of 0.591 and 0.566, respectively. The obtained results can be useful for selecting the optimal model for object detection and recognition tasks on infrared images. The YOLOv8s model is the best choice for search and rescue operations as it demonstrates high recall (0.836) and mAP50 (0.861).
35
Ma, Na, Yulong Wu, Yifan Bo, and Hongwen Yan. "Chili Pepper Object Detection Method Based on Improved YOLOv8n." Plants 13, no. 17 (2024): 2402. http://dx.doi.org/10.3390/plants13172402.
Full textAbstract:
In response to the low accuracy and slow detection speed of chili recognition in natural environments, this study proposes a chili pepper object detection method based on the improved YOLOv8n. Evaluations were conducted among YOLOv5n, YOLOv6n, YOLOv7-tiny, YOLOv8n, YOLOv9, and YOLOv10 to select the optimal model. YOLOv8n was chosen as the baseline and improved as follows: (1) Replacing the YOLOv8 backbone with the improved HGNetV2 model to reduce floating-point operations and computational load during convolution. (2) Integrating the SEAM (spatially enhanced attention module) into the YOLOv8 detection head to enhance feature extraction capability under chili fruit occlusion. (3) Optimizing feature fusion using the dilated reparam block module in certain C2f (CSP bottleneck with two convolutions). (4) Substituting the traditional upsample operator with the CARAFE(content-aware reassembly of features) upsampling operator to further enhance network feature fusion capability and improve detection performance. On a custom-built chili dataset, the F0.5-score, mAP0.5, and mAP0.5:0.95 metrics improved by 1.98, 2, and 5.2 percentage points, respectively, over the original model, achieving 96.47%, 96.3%, and 79.4%. The improved model reduced parameter count and GFLOPs by 29.5% and 28.4% respectively, with a final model size of 4.6 MB. Thus, this method effectively enhances chili target detection, providing a technical foundation for intelligent chili harvesting processes.
36
Khalid, Saim, Hadi Mohsen Oqaibi, Muhammad Aqib, and Yaser Hafeez. "Small Pests Detection in Field Crops Using Deep Learning Object Detection." Sustainability 15, no. 8 (2023): 6815. http://dx.doi.org/10.3390/su15086815.
Full textAbstract:
Deep learning algorithms, such as convolutional neural networks (CNNs), have been widely studied and applied in various fields including agriculture. Agriculture is the most important source of food and income in human life. In most countries, the backbone of the economy is based on agriculture. Pests are one of the major challenges in crop production worldwide. To reduce the overall production and economic loss from pests, advancement in computer vision and artificial intelligence may lead to early and small pest detection with greater accuracy and speed. In this paper, an approach for early pest detection using deep learning and convolutional neural networks has been presented. Object detection is applied on a dataset with images of thistle caterpillars, red beetles, and citrus psylla. The input dataset contains 9875 images of all the pests under different illumination conditions. State-of-the-art Yolo v3, Yolov3-Tiny, Yolov4, Yolov4-Tiny, Yolov6, and Yolov8 have been adopted in this study for detection. All of these models were selected based on their performance in object detection. The images were annotated in the Yolo format. Yolov8 achieved the highest mAP of 84.7% with an average loss of 0.7939, which is better than the results reported in other works when compared to small pest detection. The Yolov8 model was further integrated in an Android application for real time pest detection. This paper contributes the implementation of novel deep learning models, analytical methodology, and a workflow to detect pests in crops for effective pest management.
37
Ma, Shihao, Jiao Wu, Zhijun Zhang, and Yala Tong. "Application of Enhanced YOLOX for Debris Flow Detection in Remote Sensing Images." Applied Sciences 14, no. 5 (2024): 2158. http://dx.doi.org/10.3390/app14052158.
Full textAbstract:
Addressing the limitations, including low automation, slow recognition speed, and limited universality, of current mudslide disaster detection techniques in remote sensing imagery, this study employs deep learning methods for enhanced mudslide disaster detection. This study evaluated six object detection models: YOLOv3, YOLOv4, YOLOv5, YOLOv7, YOLOv8, and YOLOX, conducting experiments on remote sensing image data in the study area. Utilizing transfer learning, mudslide remote sensing images were fed into these six models under identical experimental conditions for training. The experimental results demonstrate that YOLOX-Nano’s comprehensive performance surpasses that of the other models. Consequently, this study introduces an enhanced model based on YOLOX-Nano (RS-YOLOX-Nano), aimed at further improving the model’s generalization capabilities and detection performance in remote sensing imagery. The enhanced model achieves a mean average precision (mAP) value of 86.04%, a 3.53% increase over the original model, and boasts a precision rate of 89.61%. Compared to the conventional YOLOX-Nano algorithm, the enhanced model demonstrates superior efficacy in detecting mudflow targets within remote sensing imagery.
38
Puchakayala Lokesh, Lahari, Rahul Gowtham Poola, Leela Prasad Gorrepati, and Siva Sankar Yellampalli. "Real-Time Cataract Diagnosis with GhostYOLO: A GhostConv-enhanced YOLO Model." Engineering, Technology & Applied Science Research 15, no. 3 (2025): 22945–52. https://doi.org/10.48084/etasr.10760.
Full textAbstract:
This study presents GhostYOLO, an enhanced YOLO-based model for cataract detection that incorporates GhostConv layers to offer greater accuracy, faster processing, and less memory consumption for real-time diagnosis. Initially, the performance of YOLO models, namely YOLOv5, YOLOv6, YOLOv7, and YOLOv8, was evaluated using 788 normal and 920 cataract images, with YOLOv8n emerging as the best standard model with excellent precision, speed, and efficiency. GhostYOLO models were developed to further improve speed and accuracy. GhostYoloV8n obtained the highest accuracy, speed, and lowest memory usage, while GhostYoLoV7-tiny also performed well. Incorporating GhostConv layers substantially improved cataract detection, increasing efficiency and real-time usage. Real-time tests using a Jetson Nano board demonstrated its efficiency, with 33.5 FPS in live detection, simplifying diagnosis. GhostYoloV8n, with only 1.6 million parameters, is a small but powerful cataract detection tool that allows for faster and more precise medical intervention. This study highlights the benefits of including GhostConv layers in YOLO models, making cataract diagnosis more accurate, efficient, and scalable for clinical applications.
39
Chen, Haosong, Fujie Zhang, Chaofan Guo, Junjie Yi, and Xiangkai Ma. "SA-SRYOLOv8: A Research on Star Anise Variety Recognition Based on a Lightweight Cascaded Neural Network and Diversified Fusion Dataset." Agronomy 14, no. 10 (2024): 2211. http://dx.doi.org/10.3390/agronomy14102211.
Full textAbstract:
Star anise, a widely popular spice, benefits from classification that enhances its economic value. In response to the low identification efficiency and accuracy of star anise varieties in the market, as well as the scarcity of related research, this study proposes an efficient identification method based on non-similarity augmentation and a lightweight cascaded neural network. Specifically, this approach utilizes a Siamese enhanced data network and a front-end SRGAN network to address sample imbalance and the challenge of identifying blurred images. The YOLOv8 model is further lightweight to reduce memory usage and increase detection speed, followed by optimization of the weight parameters through an extended training strategy. Additionally, a diversified fusion dataset of star anise, incorporating open data, was constructed to further validate the feasibility and effectiveness of this method. Testing showed that the SA-SRYOLOv8 detection model achieved an average detection precision (mAP) of 96.37%, with a detection speed of 146 FPS. Ablation experiment results showed that compared to the original YOLOv8 and the improved YOLOv8, the cascade model’s mAP increased by 0.09 to 0.81 percentage points. Additionally, when compared to mainstream detection models such as SSD, Fast R-CNN, YOLOv3, YOLOv5, YOLOX, and YOLOv7, the cascade model’s mAP increased by 1.81 to 19.7 percentage points. Furthermore, the model was significantly lighter, at only about 7.4% of the weight of YOLOv3, and operated at twice the speed of YOLOv7. Visualization results demonstrated that the cascade model accurately detected multiple star anise varieties across different scenarios, achieving high-precision detection targets. The model proposed in this study can provide new theoretical frameworks and ideas for constructing real-time star anise detection systems, offering new technological applications for smart agriculture.
40
Castro-Bello, Mirna, Dominic Brian Roman-Padilla, Cornelio Morales-Morales, et al. "Convolutional Neural Network Models in Municipal Solid Waste Classification: Towards Sustainable Management." Sustainability 17, no. 8 (2025): 3523. https://doi.org/10.3390/su17083523.
Full textAbstract:
Municipal Solid Waste (MSW) management presents a significant challenge for traditional separation practices, due to a considerable increase in quantity, diversity, complexity of types of solid waste, and a high demand for accuracy in classification. Image recognition and classification of waste using computer vision techniques allow for optimizing administration and collection processes with high precision, achieving intelligent management in separation and final disposal, mitigating environmental impact, and contributing to sustainable development objectives. This research consisted of evaluating and comparing the effectiveness of four Convolutional Neural Network models for MSW detection, using a Raspberry Pi 4 Model B. To this end, the models YOLOv4-tiny, YOLOv7-tiny, YOLOv8-nano, and YOLOv9-tiny were trained, and their performance was compared in terms of precision, inference speed, and resource usage in an embedded system with a custom dataset of 1883 organic and inorganic waste images, labeled with Roboflow by delimiting the area of interest for each object. Image preprocessing was applied, with resizing to 640 × 640 pixels and contrast auto-adjustments. Training considered 85% of images and testing considered 15%. Each training stage was conducted over 100 epochs, adjusting configuration parameters such as learning rate, weight decay, image rotation, and mosaics. The precision results obtained were as follows: YOLOv4-tiny, 91.71%; YOLOv7-tiny, 91.34%; YOLOv8-nano, 93%; and YOLOv9-tiny, 92%. Each model was applied in an embedded system with an HQ camera, achieving an average of 86% CPU usage and an inference time of 1900 ms. This suggests that the models are feasible for application in an intelligent container for classifying organic and inorganic waste, ensuring effective management and promoting a culture of environmental care in society.
41
Mien, Trinh Luong, Nguyen Dinh Tu, and Nguyen Van Lam. "Deploying YOLOv8 for Real-Time Road Crack Detection on Smart Road Length Measurement Devices." Journal of Future Artificial Intelligence and Technologies 2, no. 1 (2025): 135–44. https://doi.org/10.62411/faith.3048-3719-102.
Full textAbstract:
Nowadays, the construction, monitoring, quality control, and maintenance of roads always require high-precision, easy-to-use measuring devices in the field. This study develops an embedded computer program using the YOLOv8 model integrated into a smart road length measuring device to detect road surface cracks. First, the study analyzes and clarifies the outstanding points of the YOLOv8 model, including anchor-free: Eliminates the use of traditional anchor boxes, helping to simplify the training process, increase accuracy and reduce computational costs; C2f (Cross-Stage Partial with feature fusion); Replacing CSSPLayer in YOLOv5, this module improves feature extraction and maintains good computational performance; SPFF (Spatial Pyramid Pooling - Fast): Helps expand the receptive field without increasing computational costs, supporting object recognition at multiple scales; PAN++: Enhances the transfer of features from lower layers to the output, helping to detect small objects such as cracks well. Then, this work evaluated the performance of the proposed model by applying YOLOv8 on the surface crack dataset taken from the open-source Roboflow Universe. The results showed that the developed YOLOv8 model achieved good results with the recall indexes of 57.3%, mAP50 59.3%, precision of 64.4% mAP50-95 48.9% for YOLOv8n model, and can improve the accuracy further, reaching 79.8% using the YOLOv8m model, while still meeting the real-time processing speed of the device. These experimental results using the YOLOv8 model for detecting cracks on DongAnh road, Hanoi, Vietnam, confirm the possibility of applying the integrated road crack detection model on smart road length measuring devices in practice, achieved a precision of 43.3%, supporting the survey, inspection, supervision, construction and management of traffic infrastructure more scientifically and effectively
42
Sahithi, Kothagundla, Kalwakolu Ganesh, ,Mahankali Uday Kiran, and Padma Rajani. "Automatic Detection and Predictive Geo-location of Foreign Object Debris on Airport Runway." INTERNATIONAL JOURNAL OF SCIENTIFIC RESEARCH IN ENGINEERING AND MANAGEMENT 09, no. 06 (2025): 1–7. https://doi.org/10.55041/ijsrem.ncft004.
Full textAbstract:
Abstract—This paper presents an advanced Foreign Object Debris (FOD) detection system for airport runways using YOLOv8 and a geolocation prediction model based on machine learning regression. By integrating a self- attention mechanism with CNNs, the system achieves high detection accuracy, especially for small objects under challenging weather conditions. Ablation studies show improved Mean Average Precision (mAP), and comparisons with YOLOv5, YOLOX, and YOLOv7 confirm YOLOv8's superior performance. The geolocation model further enhances practical deployment for real- world FOD detection and removal.
43
Yang, Shixiong, Jingfa Yao, and Guifa Teng. "Corn Leaf Spot Disease Recognition Based on Improved YOLOv8." Agriculture 14, no. 5 (2024): 666. http://dx.doi.org/10.3390/agriculture14050666.
Full textAbstract:
Leaf spot disease is an extremely common disease in the growth process of maize in Northern China and its degree of harm is quite significant. Therefore, the rapid and accurate identification of maize leaf spot disease is crucial for reducing economic losses in maize. In complex field environments, traditional identification methods are susceptible to subjective interference and cannot quickly and accurately identify leaf spot disease through color or shape features. We present an advanced disease identification method utilizing YOLOv8. This method utilizes actual field images of diseased corn leaves to construct a dataset and accurately labels the diseased leaves in these images, thereby achieving rapid and accurate identification of target diseases in complex field environments. We have improved the model based on YOLOv8 by adding Slim-neck modules and GAM attention modules and introducing them to enhance the model’s ability to identify maize leaf spot disease. The enhanced YOLOv8 model achieved a precision (P) of 95.18%, a recall (R) of 89.11%, an average recognition accuracy (mAP50) of 94.65%, and an mAP50-95 of 71.62%, respectively. Compared to the original YOLOv8 model, the enhanced model showcased enhancements of 3.79%, 4.65%, 3.56%, and 7.3% in precision (P), recall (R), average recognition accuracy (mAP50), and mAP50-95, respectively. The model can effectively identify leaf spot disease and accurately calibrate its location. Under the same experimental conditions, we compared the improved model with the YOLOv3, YOLOv5, YOLOv6, Faster R-CNN, and SSD models. The results show that the improved model not only enhances performance, but also reduces parameter complexity and simplifies the network structure. The results indicated that the improved model enhanced performance, while reducing experimental time. Hence, the enhanced method proposed in this study, based on YOLOv8, exhibits the capability to identify maize leaf spot disease in intricate field environments, offering robust technical support for agricultural production.
44
Chen, Aoxiang. "Comparative Analysis of YOLO Variants Based on Performance Evaluation for Object Detection." ITM Web of Conferences 70 (2025): 03008. https://doi.org/10.1051/itmconf/20257003008.
Full textAbstract:
This study focuses on analysing and exploring the You Only Look Once (YOLO) algorithm. Specifically, this article analyses the evolution and performance of three versions (YOLOv1, YOLOv5, and YOLOv8) in object detection. The research begins by detailing the fundamental concepts of object detection and the datasets commonly used in this field. It then delves into the specific architectures and experimental outcomes associated with each YOLO version. The analysis reveals that while YOLOv8 introduces advanced features and improvements, earlier versions like YOLOv5 may offer superior stability and performance under certain conditions, particularly in specific tasks such as car detection. The discussion emphasizes the significant impact of factors such as batch size on model performance, suggesting that fine-tuning these parameters can optimize the algorithm for particular applications. The study concludes that the future of YOLO development lies in exploring and refining different variants, particularly those of YOLOv8, to better meet diverse requirements. By focusing on five distinct YOLOv8 variants, the research aims to enhance the adaptability and effectiveness of the YOLO framework across a wide range of object detection challenges, thereby contributing valuable insights into the ongoing advancement of this technology.
45
Yang, Renxu, Debao Yuan, Maochen Zhao, et al. "Camellia oleifera Tree Detection and Counting Based on UAV RGB Image and YOLOv8." Agriculture 14, no. 10 (2024): 1789. http://dx.doi.org/10.3390/agriculture14101789.
Full textAbstract:
The detection and counting of Camellia oleifera trees are important parts of the yield estimation of Camellia oleifera. The ability to identify and count Camellia oleifera trees quickly has always been important in the context of research on the yield estimation of Camellia oleifera. Because of their specific growing environment, it is a difficult task to identify and count Camellia oleifera trees with high efficiency. In this paper, based on a UAV RGB image, three different types of datasets, i.e., a DOM dataset, an original image dataset, and a cropped original image dataset, were designed. Combined with the YOLOv8 model, the detection and counting of Camellia oleifera trees were carried out. By comparing YOLOv9 and YOLOv10 in four evaluation indexes, including precision, recall, mAP, and F1 score, Camellia oleifera trees in two areas were selected for prediction and compared with the real values. The experimental results show that the cropped original image dataset was better for the recognition and counting of Camellia oleifera, and the mAP values were 8% and 11% higher than those of the DOM dataset and the original image dataset, respectively. Compared to YOLOv5, YOLOv7, YOLOv9, and YOLOv10, YOLOv8 performed better in terms of the accuracy and recall rate, and the mAP improved by 3–8%, reaching 0.82. Regression analysis was performed on the predicted and measured values, and the average R2 reached 0.94. This research shows that a UAV RGB image combined with YOLOv8 provides an effective solution for the detection and counting of Camellia oleifera trees, which is of great significance for Camellia oleifera yield estimation and orchard management.
46
Xiong, Chenqin, Tarek Zayed, Xingyu Jiang, Ghasan Alfalah, and Eslam Mohammed Abelkader. "A Novel Model for Instance Segmentation and Quantification of Bridge Surface Cracks—The YOLOv8-AFPN-MPD-IoU." Sensors 24, no. 13 (2024): 4288. http://dx.doi.org/10.3390/s24134288.
Full textAbstract:
Surface cracks are alluded to as one of the early signs of potential damage to infrastructures. In the same vein, their detection is an imperative task to preserve the structural health and safety of bridges. Human-based visual inspection is acknowledged as the most prevalent means of assessing infrastructures’ performance conditions. Nonetheless, it is unreliable, tedious, hazardous, and labor-intensive. This state of affairs calls for the development of a novel YOLOv8-AFPN-MPD-IoU model for instance segmentation and quantification of bridge surface cracks. Firstly, YOLOv8s-Seg is selected as the backbone network to carry out instance segmentation. In addition, an asymptotic feature pyramid network (AFPN) is incorporated to ameliorate feature fusion and overall performance. Thirdly, the minimum point distance (MPD) is introduced as a loss function as a way to better explore the geometric features of surface cracks. Finally, the middle aisle transformation is amalgamated with Euclidean distance to compute the length and width of segmented cracks. Analytical comparisons reveal that this developed deep learning network surpasses several contemporary models, including YOLOv8n, YOLOv8s, YOLOv8m, YOLOv8l, and Mask-RCNN. The YOLOv8s + AFPN + MPDIoU model attains a precision rate of 90.7%, a recall of 70.4%, an F1-score of 79.27%, mAP50 of 75.3%, and mAP75 of 74.80%. In contrast to alternative models, our proposed approach exhibits enhancements across performance metrics, with the F1-score, mAP50, and mAP75 increasing by a minimum of 0.46%, 1.3%, and 1.4%, respectively. The margin of error in the measurement model calculations is maintained at or below 5%. Therefore, the developed model can serve as a useful tool for the accurate characterization and quantification of different types of bridge surface cracks.
47
Hwang, Byeong Hyeon, and Mi Jin Noh. "Comparative Analysis of Toxic Marine Organism Detection Performance Across YOLO Models and Exploration of Applications in Smart Aquaculture Technology." Korean Institute of Smart Media 13, no. 11 (2024): 22–29. https://doi.org/10.30693/smj.2024.13.11.22.
Full textAbstract:
The rise in sea temperatures due to global warming has accelerated the migration of marine species, leading to the frequent discovery of toxic marine organisms in domestic waters. The blue-ringed octopus in particular is very dangerous because it contains a deadly poison called tetrodotoxin. Therefore, early detection of these toxic species and minimizing the risk to human life is crucial. This study evaluates the effectiveness of using the latest object detection technology, the YOLO model, to detect toxic marine species, aiming to provide valuable information for the development of a smart fisheries system. The analysis results showed that YOLOv8 achieved the highest precision at 0.989, followed by YOLOv7 at 0.775 and YOLOv5m at 0.318. In terms of recall, YOLOv8 scored 0.969, YOLOv5l scored 0.845, and YOLOv7 scored 0.783. For mAP50 and mAP50-95 metrics, YOLOv8 also performed the best with scores of 0.978 and 0.834, respectively. Overall, YOLOv8 demonstrated the highest performance, indicating its strong suitability for real-time detection of toxic marine organisms. On the other hand, the YOLOv5 series showed lower performance, revealing limitations in detection under complex conditions. These findings suggest that the use of the latest YOLO model is essential for establishing an early warning system for toxic marine species.
48
Özcan, Büşra, and Halit Bakır. "YAPAY ZEKA DESTEKLİ BEYİN GÖRÜNTÜLERİ ÜZERİNDE TÜMÖR TESPİTİ." International Conference on Pioneer and Innovative Studies 1 (June 13, 2023): 297–306. http://dx.doi.org/10.59287/icpis.847.
Full textAbstract:
Günümüzde yapay zeka uygulamaları sağlık sektöründe büyük bir ivme kazanmıştır. Hem hız hemdoğruluk açısından tercih edilmektedir. Bu çalışma kapsamında insan hayatı için kritik rol oynayan beyintümörleri tespiti için en güncel obje tespit algoritmaları uygulanmış ve en iyi sonucu veren model tespitedilmiştir. Popüler ve güncel YOLO modellerinden olan YOLOv5, YOLOv7 ve YOLOv8 uygulanmıştır.Model eğitilmeden önce beyin tümörleri tespitinde daha iyi sonuç vermesi adına görüntüler üzerinde renkdönüşümleri, histogram eşitleme, renk kanalları ve bu kanallar üzerinde filtre işlemleri gerçekleştirilmiş veen belirgin olan sonuç tüm görsellere uygulanmıştır. Bu işlemler sonucunda en iyi sonucu en güncel YOLOmodeli olan YOLOv8 %87 mAP değeri ile vermiştir.
49
Kumar, G. Prasanth, Pamula Sona, Konjarla Bharat Kumar, and B. Koushik Venkata Ramakrishna. "Unmanned Aerial Vehicle-Based Road Damage Detection Using Convolutional Neural Networks." INTERANTIONAL JOURNAL OF SCIENTIFIC RESEARCH IN ENGINEERING AND MANAGEMENT 09, no. 04 (2025): 1–9. https://doi.org/10.55041/ijsrem43714.
Full textAbstract:
Using images taken by unmanned aerial vehicles (UAVs) and deep learning algorithms, this study introduces a novel method for detecting road damage and cracks. Maintaining sturdy roadways depends on regularly inspecting and repairing street foundations, but manual data collection is often hazardous and time-consuming. To address this, we leverage UAV technology and artificial intelligence (AI) to improve roadway hazard identification. Our approach utilizes YOLOv4, YOLOv5, and YOLOv7, which are advanced object detection models for processing UAV imagery. Experimental results on Chinese and Spanish datasets demonstrate that YOLOv7 achieves the highest accuracy in identifying road damage. Furthermore, we introduce YOLOv8, an improved method that enhances prediction precision, surpassing prior models when trained on road damage and crack detection datasets. This study paves the way for future advancements by showcasing the potential of UAV-based deep learning in automating and improving road condition assessments. Keywords:Deep learning,Road damage,YOLO
50
Do, Van-Dinh, Van-Hung Le, Huu-Son Do, Van-Nam Phan, and Trung-Hieu Te. "TQU-HG dataset and comparative study for hand gesture recognition of RGB-based images using deep learning." Indonesian Journal of Electrical Engineering and Computer Science 34, no. 3 (2024): 1603. http://dx.doi.org/10.11591/ijeecs.v34.i3.pp1603-1617.
Full textAbstract:
Hand gesture recognition has great applications in human-computer interaction (HCI), human-robot interaction (HRI), and supporting the deaf and mute. To build a hand gesture recognition model using deep learning (DL) with high results then needs to be trained on many data and in many different conditions and contexts. In this paper, we publish the TQU-HG dataset of large RGB images with low resolution (640×480) pixels, low light conditions, and fast speed (16 fps). TQU-HG dataset includes 60,000 images collected from 20 people (10 male, 10 female) with 15 gestures of both left and right hands. A comparative study with two branches: i) based on Mediapipe TML and ii) Based on convolutional neural networks (CNNs) (you only look once (YOLO); YOLOv5, YOLOv6, YOLOv7, YOLOv8, YOLO-Nas, single shot multiBox detector (SSD) VGG16, residual network (ResNet)18, ResNext50, ResNet152, ResNext50, MobileNet V3 small, and MobileNet V3 large), the architecture and operation of CNNs models are also introduced in detail. We especially fine-tune the model and evaluate it on TQU-HG and HaGRID datasets. The quantitative results of the training and testing are presented (F1-score of YOLOv8, YOLO-Nas, MobileNet V3 small, ResNet50 is 98.99%, 98.98%, 99.27%, 99.36%, respectively on the TQU-HG dataset and is 99.21%, 99.37%, 99.36%, 86.4%, 98.3%, respectively on the HaGRID dataset). The computation time of YOLOv8 is 6.19 fps on the CPU and 18.28 fps on the GPU.