Academic literature on the topic 'Robust Event-based Object Classification'
Create a spot-on reference in APA, MLA, Chicago, Harvard, and other styles
Consult the lists of relevant articles, books, theses, conference reports, and other scholarly sources on the topic 'Robust Event-based Object Classification.'
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.
Journal articles on the topic "Robust Event-based Object Classification"
1
Liu, Changyu, Bin Lu, and Huiling Li. "Secure Access Control and Large Scale Robust Representation for Online Multimedia Event Detection." Scientific World Journal 2014 (2014): 1–12. http://dx.doi.org/10.1155/2014/219732.
Full textAbstract:
We developed an online multimedia event detection (MED) system. However, there are a secure access control issue and a large scale robust representation issue when we want to integrate traditional event detection algorithms into the online environment. For the first issue, we proposed a tree proxy-based and service-oriented access control (TPSAC) model based on the traditional role based access control model. Verification experiments were conducted on the CloudSim simulation platform, and the results showed that the TPSAC model is suitable for the access control of dynamic online environments. For the second issue, inspired by the object-bank scene descriptor, we proposed a 1000-object-bank (1000OBK) event descriptor. Feature vectors of the 1000OBK were extracted from response pyramids of 1000 generic object detectors which were trained on standard annotated image datasets, such as the ImageNet dataset. A spatial bag of words tiling approach was then adopted to encode these feature vectors for bridging the gap between the objects and events. Furthermore, we performed experiments in the context of event classification on the challenging TRECVID MED 2012 dataset, and the results showed that the robust 1000OBK event descriptor outperforms the state-of-the-art approaches.
2
Venkateswara Rao, N., G. Anil Kumar, and B. Harish. "HOG based object detection and classification." International Journal of Engineering & Technology 7, no. 3.3 (2018): 151. http://dx.doi.org/10.14419/ijet.v7i3.3.15585.
Full textAbstract:
The intension of the project is to classify objects in real world and to tracks them throughout their life spans. Object detection algorithms use feature extraction and learning algorithms to classification of an object category. Our algorithm uses a combination of “histogram of oriented gradient” (HOG) and “support vector machine” (SVM) classifier to classify of objects. Results have shown this to be a robust method in both classifying the objects along with tracking them in real time world.
3
Saikrishnan, Venkatesan, and Mani Karthikeyan. "Mayfly Optimization with Deep Learning-based Robust Object Detection and Classification on Surveillance Videos." Engineering, Technology & Applied Science Research 13, no. 5 (2023): 11747–52. http://dx.doi.org/10.48084/etasr.6231.
Full textAbstract:
Surveillance videos are recordings captured by video recording devices for monitoring and securing an area or property. These videos are frequently used in applications, involving law enforcement, security systems, retail analytics, and traffic monitoring. Surveillance videos can provide valuable visual information for analyzing patterns, identifying individuals or objects of interest, and detecting and investigating incidents. Object detection and classification on video surveillance involves the usage of computer vision techniques to identify and categorize objects within the video footage. Object detection algorithms are employed to locate and identify objects within each frame. These algorithms use various techniques, namely bounding box regression, Convolutional Neural Networks (CNNs), and feature extraction to detect objects of interest. This study presents the Mayfly Optimization with Deep Learning-based Robust Object Detection and Classification (MFODL-RODC) method on surveillance videos. The main aim of the MFODL-RODC technique lies in the accurate classification and recognition of objects in surveillance videos. To accomplish this, the MFODL-RODC method follows a two-step process, consisting of object detection and object classification. The MFODL-RODC method uses the EfficientDet object detector for the object detection process. Besides, the classification of detected objects takes place using the Variational Autoencoder (VAE) model. The MFO algorithm is employed to enrich the performance of the VAE model. The simulation examination of the MFODL-RODC technique is performed on benchmark datasets. The extensive results accentuated the improved performance of the MFODL-RODC method over other existing algorithms with an output of 98.89%.
4
El Shair, Zaid, and Samir A. Rawashdeh. "High-Temporal-Resolution Object Detection and Tracking Using Images and Events." Journal of Imaging 8, no. 8 (2022): 210. http://dx.doi.org/10.3390/jimaging8080210.
Full textAbstract:
Event-based vision is an emerging field of computer vision that offers unique properties, such as asynchronous visual output, high temporal resolutions, and dependence on brightness changes, to generate data. These properties can enable robust high-temporal-resolution object detection and tracking when combined with frame-based vision. In this paper, we present a hybrid, high-temporal-resolution object detection and tracking approach that combines learned and classical methods using synchronized images and event data. Off-the-shelf frame-based object detectors are used for initial object detection and classification. Then, event masks, generated per detection, are used to enable inter-frame tracking at varying temporal resolutions using the event data. Detections are associated across time using a simple, low-cost association metric. Moreover, we collect and label a traffic dataset using the hybrid sensor DAVIS 240c. This dataset is utilized for quantitative evaluation using state-of-the-art detection and tracking metrics. We provide ground truth bounding boxes and object IDs for each vehicle annotation. Further, we generate high-temporal-resolution ground truth data to analyze tracking performance at different temporal rates. Our approach shows promising results, with minimal performance deterioration at higher temporal resolutions (48–384 Hz) when compared with the baseline frame-based performance at 24 Hz.
5
Song, Hanbin, Sanghyeop Yeo, Youngwan Jin, et al. "Short-Wave Infrared (SWIR) Imaging for Robust Material Classification: Overcoming Limitations of Visible Spectrum Data." Applied Sciences 14, no. 23 (2024): 11049. http://dx.doi.org/10.3390/app142311049.
Full textAbstract:
This paper presents a novel approach to material classification using short-wave infrared (SWIR) imaging, aimed at applications where differentiating visually similar objects based on material properties is essential, such as in autonomous driving. Traditional vision systems, relying on visible spectrum imaging, struggle to distinguish between objects with similar appearances but different material compositions. Our method leverages SWIR’s distinct reflectance characteristics, particularly for materials containing moisture, and demonstrates a significant improvement in accuracy. Specifically, SWIR data achieved near-perfect classification results with an accuracy of 99% for distinguishing real from artificial objects, compared to 77% with visible spectrum data. In object detection tasks, our SWIR-based model achieved a mean average precision (mAP) of 0.98 for human detection and up to 1.00 for other objects, demonstrating its robustness in reducing false detections. This study underscores SWIR’s potential to enhance object recognition and reduce ambiguity in complex environments, offering a valuable contribution to material-based object recognition in autonomous driving, manufacturing, and beyond.
6
Guan, Yurong, Muhammad Aamir, Zhihua Hu, et al. "A Region-Based Efficient Network for Accurate Object Detection." Traitement du Signal 38, no. 2 (2021): 481–94. http://dx.doi.org/10.18280/ts.380228.
Full textAbstract:
Object detection in images is an important task in image processing and computer vision. Many approaches are available for object detection. For example, there are numerous algorithms for object positioning and classification in images. However, the current methods perform poorly and lack experimental verification. Thus, it is a fascinating and challenging issue to position and classify image objects. Drawing on the recent advances in image object detection, this paper develops a region-baed efficient network for accurate object detection in images. To improve the overall detection performance, image object detection was treated as a twofold problem, involving object proposal generation and object classification. First, a framework was designed to generate high-quality, class-independent, accurate proposals. Then, these proposals, together with their input images, were imported to our network to learn convolutional features. To boost detection efficiency, the number of proposals was reduced by a network refinement module, leaving only a few eligible candidate proposals. After that, the refined candidate proposals were loaded into the detection module to classify the objects. The proposed model was tested on the test set of the famous PASCAL Visual Object Classes Challenge 2007 (VOC2007). The results clearly demonstrate that our model achieved robust overall detection efficiency over existing approaches using fewer or more proposals, in terms of recall, mean average best overlap (MABO), and mean average precision (mAP).
7
Guan, Yurong, Muhammad Aamir, Zhihua Hu, et al. "An Object Detection Framework Based on Deep Features and High-Quality Object Locations." Traitement du Signal 38, no. 3 (2021): 719–30. http://dx.doi.org/10.18280/ts.380319.
Full textAbstract:
Objection detection has long been a fundamental issue in computer vision. Despite being widely studied, it remains a challenging task in the current body of knowledge. Many researchers are eager to develop a more robust and efficient mechanism for object detection. In the extant literature, promising results are achieved by many novel approaches of object detection and classification. However, there is ample room to further enhance the detection efficiency. Therefore, this paper proposes an image object detection and classification, using a deep neural network (DNN) for based on high-quality object locations. The proposed method firstly derives high-quality class-independent object proposals (locations) through computing multiple hierarchical segments with super pixels. Next, the proposals were ranked by region score, i.e., several contours wholly enclosed in the proposed region. After that, the top-ranking object proposal was adopted for post-classification by the DNN. During the post-classification, the network extracts the eigenvectors from the proposals, and then maps the features with the softmax classifier, thereby determining the class of each object. The proposed method was found superior to traditional approaches through an evaluation on Pascal VOC 2007 Dataset.
8
Ghadi, Yazeed Yasin, Adnan Ahmed Rafique, Tamara al Shloul, Suliman A. Alsuhibany, Ahmad Jalal, and Jeongmin Park. "Robust Object Categorization and Scene Classification over Remote Sensing Images via Features Fusion and Fully Convolutional Network." Remote Sensing 14, no. 7 (2022): 1550. http://dx.doi.org/10.3390/rs14071550.
Full textAbstract:
The latest visionary technologies have made an evident impact on remote sensing scene classification. Scene classification is one of the most challenging yet important tasks in understanding high-resolution aerial and remote sensing scenes. In this discipline, deep learning models, particularly convolutional neural networks (CNNs), have made outstanding accomplishments. Deep feature extraction from a CNN model is a frequently utilized technique in these approaches. Although CNN-based techniques have achieved considerable success, there is indeed ample space for improvement in terms of their classification accuracies. Certainly, fusion with other features has the potential to extensively improve the performance of distant imaging scene classification. This paper, thus, offers an effective hybrid model that is based on the concept of feature-level fusion. We use the fuzzy C-means segmentation technique to appropriately classify various objects in the remote sensing images. The segmented regions of the image are then labeled using a Markov random field (MRF). After the segmentation and labeling of the objects, classical and CNN features are extracted and combined to classify the objects. After categorizing the objects, object-to-object relations are studied. Finally, these objects are transmitted to a fully convolutional network (FCN) for scene classification along with their relationship triplets. The experimental evaluation of three publicly available standard datasets reveals the phenomenal performance of the proposed system.
9
Lang, Graham K., and Peter Seitz. "Robust classification of arbitrary object classes based on hierarchical spatial feature-matching." Machine Vision and Applications 10, no. 3 (1997): 123–35. http://dx.doi.org/10.1007/s001380050065.
Full text
10
Konstantinov, A. V., S. R. Kirpichenko, and L. V. Utkin. "Generating Survival Interpretable Trajectories and Data." Doklady Mathematics 110, S1 (2024): S75—S86. https://doi.org/10.1134/s1064562424601999.
Full textAbstract:
Abstract A new model for generating survival trajectories and data based on applying an autoencoder of a specific structure is proposed. It solves three tasks. First, it provides predictions in the form of the expected event time and the survival function for a new feature vector based on the Beran estimator. Second, the model generates additional data based on a given training set that would supplement the original dataset. Third, the most important, it generates a prototype time-dependent trajectory for an object, which characterizes how features of the object could be changed to achieve a different time to an event. The trajectory can be viewed as a type of the counterfactual explanation. The proposed model is robust during training and inference due to a specific weighting scheme incorporated into the variational autoencoder. The model also determines the censored indicators of new generated data by solving a classification task. The paper demonstrates the efficiency and properties of the proposed model using numerical experiments on synthetic and real datasets. The code of the algorithm implementing the proposed model is publicly available.
Dissertations / Theses on the topic "Robust Event-based Object Classification"
1
CENG, YUN-FENG, and 曾雲楓. "Deep-Learning-Based Object Classification and Grasping Point Determination Based on RGB-D Image for Robot Arm Operation." Thesis, 2019. http://ndltd.ncl.edu.tw/handle/6epvc5.
Full textBook chapters on the topic "Robust Event-based Object Classification"
1
Kalina, Jan, and Patrik Janáček. "Robustness Aspects of Optimized Centroids." In Studies in Classification, Data Analysis, and Knowledge Organization. Springer International Publishing, 2023. http://dx.doi.org/10.1007/978-3-031-09034-9_22.
Full textAbstract:
AbstractCentroids are often used for object localization tasks, supervised segmentation in medical image analysis, or classification in other specific tasks. This paper starts by contributing to the theory of centroids by evaluating the effect of modified illumination on the weighted correlation coefficient. Further, robustness of various centroid-based tools is investigated in experiments related to mouth localization in non-standardized facial images or classification of high-dimensional data in a matched pairs design. The most robust results are obtained if the sparse centroid-based method for supervised learning is accompanied with an intrinsic variable selection. Robustness, sparsity, and energy-efficient computation turn out not to contradict the requirement on the optimal performance of the centroids.
2
Ben Itzhak, Sagi, Nahum Kiryati, Orith Portnoy, and Arnaldo Mayer. "Localization-Guided Supervision for Robust Medical Image Classification by Vision Transformers." In Lecture Notes in Computer Science. Springer Nature Switzerland, 2025. https://doi.org/10.1007/978-3-031-92648-8_8.
Full textAbstract:
Abstract A major challenge in developing data-driven algorithms for medical imaging is the limited size of available datasets. Furthermore, these datasets often suffer from inter-site heterogeneity caused by the use of different scanners and scanning protocols. These factors may contribute to overfitting, which undermines the generalization ability and robustness of deep learning classification models in the medical domain, leading to inadequate performance in real-world applications. To address these challenges and mitigate overfitting, we propose a framework which incorporates explanation supervision during training of Vision Transformer (ViT) models for image classification. Our approach leverages foreground masks of the class object during training to regularize attribution maps extracted from ViT, encouraging the model to focus on relevant image regions and make predictions based on pertinent features. We introduce a new method for generating explanatory attribution maps from ViT-based models and construct a dual-loss function that combines a conventional classification loss with a term that regularizes attribution maps. Our approach demonstrates superior performance over existing methods on two challenging medical imaging datasets, highlighting its effectiveness in the medical domain and its potential for application in other fields. Source code is available at: https://github.com/sagibe/LGMViT.
3
Zaidi, Syed Farhan Alam, Rahat Hussain, Muhammad Sibtain Abbas, Jaehun Yang, Doyeop Lee, and Chansik Park. "iSafe Welding System: Computer Vision-Based Monitoring System for Safe Welding Work." In CONVR 2023 - Proceedings of the 23rd International Conference on Construction Applications of Virtual Reality. Firenze University Press, 2023. http://dx.doi.org/10.36253/979-12-215-0289-3.66.
Full textAbstract:
The construction industry faces significant challenges, including a high prevalence of occupational incidents, often involving fires, explosions, and burn-related accidents due to worker non-compliance with safety protocols. Adherence to safety guidelines and proper utilization of safety equipment are critical to preventing such incidents and safeguarding workers in hazardous work environments. Consequently, a monitoring system tailored for construction safety during welding operations becomes imperative to mitigate the risk of fire accidents. This paper conducts a brief analysis of OSHA rules pertaining to welding work and introduces the iSafe Welding system, an advanced real-time safety monitoring and compliance enforcement solution designed specifically for construction site welding operations. Harnessing the real-time object detection algorithm YOLOv7 in conjunction with rule-based scene classification, the system excels in identifying potential safety violations. Rigorous evaluation, encompassing precision, recall, mean Average Precision (mAP), accuracy, and the F1-Score, sheds light on its strengths and areas for improvement. The system showcases robust performance in rule-based scene classification, achieving high accuracy, precision, and recall rates. Notably, the iSafe Welding system demonstrates a formidable potential for enhancing construction site safety and regulatory compliance. Ongoing enhancements, including dataset expansion and model refinement, underscore its commitment to real-world deployment and its strength in ensuring worker safety
4
Zaidi, Syed Farhan Alam, Rahat Hussain, Muhammad Sibtain Abbas, Jaehun Yang, Doyeop Lee, and Chansik Park. "iSafe Welding System: Computer Vision-Based Monitoring System for Safe Welding Work." In CONVR 2023 - Proceedings of the 23rd International Conference on Construction Applications of Virtual Reality. Firenze University Press, 2023. http://dx.doi.org/10.36253/10.36253/979-12-215-0289-3.66.
Full textAbstract:
The construction industry faces significant challenges, including a high prevalence of occupational incidents, often involving fires, explosions, and burn-related accidents due to worker non-compliance with safety protocols. Adherence to safety guidelines and proper utilization of safety equipment are critical to preventing such incidents and safeguarding workers in hazardous work environments. Consequently, a monitoring system tailored for construction safety during welding operations becomes imperative to mitigate the risk of fire accidents. This paper conducts a brief analysis of OSHA rules pertaining to welding work and introduces the iSafe Welding system, an advanced real-time safety monitoring and compliance enforcement solution designed specifically for construction site welding operations. Harnessing the real-time object detection algorithm YOLOv7 in conjunction with rule-based scene classification, the system excels in identifying potential safety violations. Rigorous evaluation, encompassing precision, recall, mean Average Precision (mAP), accuracy, and the F1-Score, sheds light on its strengths and areas for improvement. The system showcases robust performance in rule-based scene classification, achieving high accuracy, precision, and recall rates. Notably, the iSafe Welding system demonstrates a formidable potential for enhancing construction site safety and regulatory compliance. Ongoing enhancements, including dataset expansion and model refinement, underscore its commitment to real-world deployment and its strength in ensuring worker safety
5
Mikriukov, Georgii, Gesina Schwalbe, Christian Hellert, and Korinna Bade. "Evaluating the Stability of Semantic Concept Representations in CNNs for Robust Explainability." In Communications in Computer and Information Science. Springer Nature Switzerland, 2023. http://dx.doi.org/10.1007/978-3-031-44067-0_26.
Full textAbstract:
AbstractAnalysis of how semantic concepts are represented within Convolutional Neural Networks (CNNs) is a widely used approach in Explainable Artificial Intelligence (XAI) for interpreting CNNs. A motivation is the need for transparency in safety-critical AI-based systems, as mandated in various domains like automated driving. However, to use the concept representations for safety-relevant purposes, like inspection or error retrieval, these must be of high quality and, in particular, stable. This paper focuses on two stability goals when working with concept representations in computer vision CNNs: stability of concept retrieval and of concept attribution. The guiding use-case is a post-hoc explainability framework for object detection (OD) CNNs, towards which existing concept analysis (CA) methods are successfully adapted. To address concept retrieval stability, we propose a novel metric that considers both concept separation and consistency, and is agnostic to layer and concept representation dimensionality. We then investigate impacts of concept abstraction level, number of concept training samples, CNN size, and concept representation dimensionality on stability. For concept attribution stability we explore the effect of gradient instability on gradient-based explainability methods. The results on various CNNs for classification and object detection yield the main findings that (1) the stability of concept retrieval can be enhanced through dimensionality reduction via data aggregation, and (2) in shallow layers where gradient instability is more pronounced, gradient smoothing techniques are advised. Finally, our approach provides valuable insights into selecting the appropriate layer and concept representation dimensionality, paving the way towards CA in safety-critical XAI applications.
6
Yao, Wei, and Jianwei Wu. "Airborne LiDAR for Detection and Characterization of Urban Objects and Traffic Dynamics." In Urban Informatics. Springer Singapore, 2021. http://dx.doi.org/10.1007/978-981-15-8983-6_22.
Full textAbstract:
AbstractIn this chapter, we present an advanced machine learning strategy to detect objects and characterize traffic dynamics in complex urban areas by airborne LiDAR. Both static and dynamical properties of large-scale urban areas can be characterized in a highly automatic way. First, LiDAR point clouds are colorized by co-registration with images if available. After that, all data points are grid-fitted into the raster format in order to facilitate acquiring spatial context information per-pixel or per-point. Then, various spatial-statistical and spectral features can be extracted using a cuboid volumetric neighborhood. The most important features highlighted by the feature-relevance assessment, such as LiDAR intensity, NDVI, and planarity or covariance-based features, are selected to span the feature space for the AdaBoost classifier. Classification results as labeled points or pixels are acquired based on pre-selected training data for the objects of building, tree, vehicle, and natural ground. Based on the urban classification results, traffic-related vehicle motion can further be indicated and determined by analyzing and inverting the motion artifact model pertinent to airborne LiDAR. The performance of the developed strategy towards detecting various urban objects is extensively evaluated using both public ISPRS benchmarks and peculiar experimental datasets, which were acquired across European and Canadian downtown areas. Both semantic and geometric criteria are used to assess the experimental results at both per-pixel and per-object levels. In the datasets of typical city areas requiring co-registration of imagery and LiDAR point clouds a priori, the AdaBoost classifier achieves a detection accuracy of up to 90% for buildings, up to 72% for trees, and up to 80% for natural ground, while a low and robust false-positive rate is observed for all the test sites regardless of object class to be evaluated. Both theoretical and simulated studies for performance analysis show that the velocity estimation of fast-moving vehicles is promising and accurate, whereas slow-moving ones are hard to distinguish and yet estimated with acceptable velocity accuracy. Moreover, the point density of ALS data tends to be related to system performance. The velocity can be estimated with high accuracy for nearly all possible observation geometries except for those vehicles moving in or (quasi-)along the track. By comparative performance analysis of the test sites, the performance and consistent reliability of the developed strategy for the detection and characterization of urban objects and traffic dynamics from airborne LiDAR data based on selected features was validated and achieved.
7
Khellat-Kihel, Souad, Zhenan Sun, and Massimo Tistarelli. "An Hybrid Attention-Based System for the Prediction of Facial Attributes." In Lecture Notes in Computer Science. Springer International Publishing, 2021. http://dx.doi.org/10.1007/978-3-030-82427-3_9.
Full textAbstract:
AbstractRecent research on face analysis has demonstrated the richness of information embedded in feature vectors extracted from a deep convolutional neural network. Even though deep learning achieved a very high performance on several challenging visual tasks, such as determining the identity, age, gender and race, it still lacks a well grounded theory which allows to properly understand the processes taking place inside the network layers. Therefore, most of the underlying processes are unknown and not easy to control. On the other hand, the human visual system follows a well understood process in analyzing a scene or an object, such as a face. The direction of the eye gaze is repeatedly directed, through purposively planned saccadic movements, towards salient regions to capture several details. In this paper we propose to capitalize on the knowledge of the saccadic human visual processes to design a system to predict facial attributes embedding a biologically-inspired network architecture, the HMAX. The architecture is tailored to predict attributes with different textural information and conveying different semantic meaning, such as attributes related and unrelated to the subject’s identity. Salient points on the face are extracted from the outputs of the S2 layer of the HMAX architecture and fed to a local texture characterization module based on LBP (Local Binary Pattern). The resulting feature vector is used to perform a binary classification on a set of pre-defined visual attributes. The devised system allows to distill a very informative, yet robust, representation of the imaged faces, allowing to obtain high performance but with a much simpler architecture as compared to a deep convolutional neural network. Several experiments performed on publicly available, challenging, large datasets demonstrate the validity of the proposed approach.
8
Schott, Moritz, Adina Zell, Sven Lautenbach, et al. "Analyzing and Improving the Quality and Fitness for Purpose of OpenStreetMap as Labels in Remote Sensing Applications." In Volunteered Geographic Information. Springer Nature Switzerland, 2023. http://dx.doi.org/10.1007/978-3-031-35374-1_2.
Full textAbstract:
AbstractOpenStreetMap (OSM) is a well-known example of volunteered geographic information. It has evolved to one of the most used geographic databases. As data quality of OSM is heterogeneous both in space and across different thematic domains, data quality assessment is of high importance for potential users of OSM data. As use cases differ with respect to their requirements, it is not data quality per se that is of interest for the user but fitness for purpose. We investigate the fitness for purpose of OSM to derive land-use and land-cover labels for remote sensing-based classification models. Therefore, we evaluated OSM land-use and land-cover information by two approaches: (1) assessment of OSM fitness for purpose for samples in relation to intrinsic data quality indicators at the scale of individual OSM objects and (2) assessment of OSM-derived multi-labels at the scale of remote sensing patches ($$1.22 \times 1.22$$ 1.22 × 1.22 km) in combination with deep learning approaches. The first approach was applied to 1000 randomly selected relevant OSM objects. The quality score for each OSM object in the samples was combined with a large set of intrinsic quality indicators (such as the experience of the mapper, the number of mappers in a region, and the number of edits made to the object) and auxiliary information about the location of the OSM object (such as the continent or the ecozone). Intrinsic indicators were derived by a newly developed tool based on the OSHDB (OpenStreetMap History DataBase). Afterward, supervised and unsupervised shallow learning approaches were used to identify relationships between the indicators and the quality score. Overall, investigated OSM land-use objects were of high quality: both geometry and attribute information were mostly accurate. However, areas without any land-use information in OSM existed even in well-mapped areas such as Germany. The regression analysis at the level of the individual OSM objects revealed associations between intrinsic indicators, but also a strong variability. Even if more experienced mappers tend to produce higher quality and objects which underwent multiple edits tend to be of higher quality, an inexperienced mapper might map a perfect land-use polygon. This result indicates that it is hard to predict data quality of individual land-use objects purely on intrinsic data quality indicators. The second approach employed a label-noise robust deep learning method on remote sensing data with OSM labels. As the quality of the OSM labels was manually assessed beforehand, it was possible to control the amount of noise in the dataset during the experiment. The addition of artificial noise allowed for an even more fine-grained analysis on the effect of noise on prediction quality. The noise-tolerant deep learning method was capable to identify correct multi-labels even for situations with significant levels of noise added. The method was also used to identify areas where input labels were likely wrong. Thereby, it is possible to provide feedback to the OSM community as areas of concern can be flagged.
9
Wang, Xiaochun, Xiali Wang, and Don Mitchell Wilkes. "Supervised Learning for Data Classification Based Object Recognition." In Machine Learning-based Natural Scene Recognition for Mobile Robot Localization in An Unknown Environment. Springer Singapore, 2019. http://dx.doi.org/10.1007/978-981-13-9217-7_9.
Full text
10
Vimal kumar, V., S. Priya, M. Shanmugapriya, and Aparna George. "Deep Learning-Based Bluetooth-Controlled Robot for Automated Object Classification." In Lecture Notes in Networks and Systems. Springer Singapore, 2021. http://dx.doi.org/10.1007/978-981-33-4355-9_45.
Full textConference papers on the topic "Robust Event-based Object Classification"
1
Yang, Panpan, Ziming Wang, Huajin Tang, and Rui Yan. "Multi-scale Harmonic Mean Time Surfaces for Event-based Object Classification." In 2024 International Joint Conference on Neural Networks (IJCNN). IEEE, 2024. http://dx.doi.org/10.1109/ijcnn60899.2024.10650679.
Full text
2
Hasssan, Ahmed, Jian Meng, and Jae-Sun Seo. "Spiking Neural Network with Learnable Threshold for Event-based Classification and Object Detection." In 2024 International Joint Conference on Neural Networks (IJCNN). IEEE, 2024. http://dx.doi.org/10.1109/ijcnn60899.2024.10650320.
Full text
3
Sironi, Amos, Manuele Brambilla, Nicolas Bourdis, Xavier Lagorce, and Ryad Benosman. "HATS: Histograms of Averaged Time Surfaces for Robust Event-Based Object Classification." In 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR). IEEE, 2018. http://dx.doi.org/10.1109/cvpr.2018.00186.
Full text
4
Schumann, Arne, Lars Sommer, Johannes Klatte, Tobias Schuchert, and Jurgen Beyerer. "Deep cross-domain flying object classification for robust UAV detection." In 2017 14th IEEE International Conference on Advanced Video and Signal Based Surveillance (AVSS). IEEE, 2017. http://dx.doi.org/10.1109/avss.2017.8078558.
Full text
5
Kumar Mandal, Diptesh, Kazunori Kaede, and Keiichi Watanuki. "Advancing Vision-based Adaptive Gripping Technology with Machine Learning: Leveraging Pre-trained Models for Enhanced Object Classification." In 2024 AHFE International Conference on Human Factors in Design, Engineering, and Computing (AHFE 2024 Hawaii Edition). AHFE International, 2024. http://dx.doi.org/10.54941/ahfe1005719.
Full textAbstract:
The rapid advancements in robotics and automation have highlighted the need for robotic systems capable of adapting to the diverse physical properties of objects. Traditional grippers often lack the versatility to handle both hard and soft objects without extensive reprogramming or hardware adjustments. Previous studies have explored various approaches to this challenge, including tactile sensors and force feedback mechanisms to distinguish object properties. For instance, research by Calandra et al. (2018) utilized deep learning with tactile data to enable robotic hands to identify objects and adjust grip accordingly. Similarly, the paper by Li et al. (2020) “Design and performance characterization of a soft robot hand with fingertip haptic feedback for teleoperation”, focuses on designing and characterizing a soft robotic hand with fingertip haptic feedback for teleoperation emphasizing real-time tactile sensing and feedback mechanisms. However, these studies primarily focus on tactile feedback or specialized hardware, limiting their applicability in scenarios where such systems are not available or practical.This study introduces a novel machine learning-based approach, focusing on the use of visual data alone to classify and adapt to the hardness or softness of objects. By leveraging the CIFAR-100 dataset, we trained a deep learning model based on the ResNet50 architecture, achieving significant results in binary classification of hard and soft objects. The CIFAR-100 dataset, consisting of 100 diverse object categories/classes, was reorganized into two classes: hard (39 categories) and soft (61 categories). The ResNet50 model, pre-trained on ImageNet, was fine-tuned specifically for this task, with modifications to the last 210 layers to enhance its adaptability.Data augmentation techniques, including rotations, translations, shearing, zooming, and horizontal flipping, were applied to simulate real-world variations, ensuring robust learning. The model was further refined with additional fully connected layers, dropout, and batch normalization to prevent overfitting. Optimized using the AdamW optimizer, the model achieved a training accuracy of 83.31% and a validation accuracy of 80.25%, with a test accuracy of 80%. The precision, recall, and F1-scores were 0.82, 0.86, and 0.84 for the soft object class, and 0.76, 0.71, and 0.73 for the hard object class, demonstrating the model’s effectiveness in distinguishing between hard and soft objects without the need for specialized sensors. It is also observed that the accuracy of the model in relation to hard objects is significantly lesser as compared to soft objects. It is understood that the CIFAR-100 dataset (comprising of 100 classes) is inadequate for model training, so we are exploring the ILSVRC (ImageNet subset) dataset for model training in future. The research is useful in a variety of fields where the focus lies on object handling. In everyday life, we encounter a wide array of objects with varying degrees of hardness, requiring different levels of care and precision during handling. By relying on visual data and machine learning algorithms, as demonstrated in this research, robotic systems can become more autonomous and versatile, reducing the burden on human operators and improving overall efficiency. This approach can lead to cost savings by reducing the need for specialized hardware, such as tactile sensors, which are often expensive and difficult to integrate.
6
Jones, Darren, Pavankumar Gangadhar, Randall McGrail, Sudipta Pati, Erik Antonsson, and Ravi Patel. "Process Improvements for Determining Fault Tolerant Time Intervals." In WCX SAE World Congress Experience. SAE International, 2024. http://dx.doi.org/10.4271/2024-01-2791.
Full textAbstract:
<div class="section abstract"><div class="htmlview paragraph">ISO 26262-1:2018 defines the fault tolerant time interval (FTTI) as the minimum time span from the occurrence of a fault within an electrical / electronic system to a possible occurrence of a hazardous event. FTTI provides a time limit within which compliant vehicle safety mechanisms must detect and react to faults capable of posing risk of harm to persons. This makes FTTI a vital safety characteristic for system design. Common automotive industry practice accommodates recording fault times of occurrence definitively. However, current practice for defining the time of hazardous event onset relies upon subjective judgements.</div><div class="htmlview paragraph">This paper presents a novel method to define hazardous event onset more objectively. The method introduces the Streetscope Collision Hazard Measure (SHM<sup>TM</sup>) and a refined approach to hazardous event classification. SHM inputs kinematic factors such as proximity, relative speed, and acceleration as well as environmental characteristics like traffic patterns, visibility, and road conditions. SHM utilizes these inputs to calculate a time-stamped, 0-to-100 normalized, hazard metric for the subject, or ego, vehicle. SAE J2980 exemplifies the industry standard practice for hazard and operability analysis (HAZOP) and hazard analysis and risk assessment (HARA). This paper adds an extensive operational situations (OpSit) catalog and hazard effect descriptors to further objectify definition of applicable, vehicle-level hazardous events. The OpSit catalog describes numerous driving scenarios that span the road vehicle operational design domain (ODD). Hazard effect descriptors like side collision, pedestrian impact, and strike stationary object support refined onset determinations.</div><div class="htmlview paragraph">This method allows stakeholders to assign a SHM threshold for hazardous event onset for every applicable combination of malfunction, hazard, operational situation, and hazard effect. Test vehicle dash cameras and simulation data sets demonstrate robust measurement of the time interval between fault injection and exceeding the SHM threshold. The minimum time intervals identified for each hazard becomes its FTTI. Incorporating novel SHM, OpSit catalog, and hazard effect descriptors into industry standard recommended practices improves FTTI determinations.</div></div>
7
Laroca, Rayson, and David Menotti. "Automatic License Plate Recognition: An Efficient and Layout-Independent Approach Based on the YOLO Detector." In Concurso de Teses e Dissertações da SBC. Sociedade Brasileira de Computação - SBC, 2020. http://dx.doi.org/10.5753/ctd.2020.11372.
Full textAbstract:
Automatic License Plate Recognition (ALPR) has been a frequent topic of research due to many practical applications, such as border control and traffic law enforcement. This work presents an efficient, robust and layout-independent ALPR system based on the YOLO object detector that contains a unified approach for license plate detection and layout classification and leverages post-processing rules in the recognition stage to eliminate a major shortcoming of existing ALPR systems (being layout dependent). We also introduce a publicly available dataset for ALPR that has become very popular, having been downloaded more than 550 times by researchers from 76 different countries in the last year alone. The proposed system, which performs in real time even when there are 4 vehicles in the scene, outperformed both previous works and commercial systems on four public datasets widely used in the literature.
8
Laroca, Rayson, and David Menotti. "Automatic License Plate Recognition: An Efficient and Layout-Independent System Based on the YOLO Detector." In Conference on Graphics, Patterns and Images. Sociedade Brasileira de Computação, 2020. http://dx.doi.org/10.5753/sibgrapi.est.2020.12978.
Full textAbstract:
Automatic License Plate Recognition (ALPR) has been a frequent topic of research due to many practical applications, such as border control and traffic law enforcement. This work presents an efficient, robust and layout-independent ALPR system based on the YOLO object detector that contains a unified approach for license plate detection and layout classification and that leverages post-processing rules in the recognition stage to eliminate a major shortcoming of existing ALPR systems (being layout dependent). We also introduce a publicly available dataset for ALPR, called UFPR-ALPR, that has become very popular, having been downloaded more than 650 times by researchers from 80 different countries over the past two years. The proposed system, which performs in real time even when there are 4 vehicles in the scene, outperformed both previous works and commercial systems on four public datasets widely used in the literature. The entire ALPR system (i.e., the architectures and weights), along with all annotations made by us are publicly available at https://web.inf.ufpr.br/vri/publications/layout-independent-alpr/.
9
Da Silva, Diego Alves, Aline Geovanna Soares, Antonio Lundgren, Estanislau Lima, and Byron Leite Dantas Bezerra. "NAO-Read: Empowering the Humanoid Robot NAO to Recognize Texts in Objects in Natural Scenes." In Conference on Graphics, Patterns and Images. Sociedade Brasileira de Computação, 2020. http://dx.doi.org/10.5753/sibgrapi.est.2020.12999.
Full textAbstract:
Robotics is a field of research that has undergone several changes in recent years. Currently, robot applications are commonly used for many applications, such as pump deactivation, mobile robotic manipulation, etc. However, most robots today are programmed to follow a predefined path. This is sufficient when the robot is working in a settled environment. Nonetheless, for many tasks, autonomous robots are needed. In this way, NAO humanoid robots constitute the new active research platform within the robotics community. In this article, we present a vision system that connects to the NAO robot, allowing robots to detect and recognize the visible text present in objects in images of natural scenes and use that knowledge to interpret the content of a given scene. The proposed vision system is based on deep learning methods and was designed to be used by NAO robots and consists of five stages: 1) capturing the image; 2) after capturing the image, the YOLOv3 algorithm is used for object detection and classification; 3) selection of the objects of interest; 4) text detection and recognition stage, based on the OctShuffleMLT approach; and 5) synthesis of the text. The choice of these models was due to the better results obtained in the COCO databases, in the list of objects, and in the ICDAR 2015, in the text list, these bases are very similar to those found with the NAO robot. Experimental results show that the rate of detecting and recognizing text from the images obtained through the NAO robot camera in the wild are similar to those presented in models pre-trained with natural scenes databases.
10
Gao, Shan, Guangqian Guo, and C. L. Philip Chen. "Event-Based Incremental Broad Learning System for Object Classification." In 2019 IEEE/CVF International Conference on Computer Vision Workshop (ICCVW). IEEE, 2019. http://dx.doi.org/10.1109/iccvw.2019.00361.
Full textReports on the topic "Robust Event-based Object Classification"
1
Asari, Vijayan, Paheding Sidike, Binu Nair, Saibabu Arigela, Varun Santhaseelan, and Chen Cui. PR-433-133700-R01 Pipeline Right-of-Way Automated Threat Detection by Advanced Image Analysis. Pipeline Research Council International, Inc. (PRCI), 2015. http://dx.doi.org/10.55274/r0010891.
Full textAbstract:
A novel algorithmic framework for the robust detection and classification of machinery threats and other potentially harmful objects intruding onto a pipeline right-of-way (ROW) is designed from three perspectives: visibility improvement, context-based segmentation, and object recognition/classification. In the first part of the framework, an adaptive image enhancement algorithm is utilized to improve the visibility of aerial imagery to aid in threat detection. In this technique, a nonlinear transfer function is developed to enhance the processing of aerial imagery with extremely non-uniform lighting conditions. In the second part of the framework, the context-based segmentation is developed to eliminate regions from imagery that are not considered to be a threat to the pipeline. Context based segmentation makes use of a cascade of pre-trained classifiers to search for regions that are not threats. The context based segmentation algorithm accelerates threat identification and improves object detection rates. The last phase of the framework is an efficient object detection model. Efficient object detection �follows a three-stage approach which includes extraction of the local phase in the image and the use of local phase characteristics to locate machinery threats. The local phase is an image feature extraction technique which partially removes the lighting variance and preserves the edge information of the object. Multiple orientations of the same object are matched and the correct orientation is selected using feature matching by histogram of local phase in a multi-scale framework. The classifier outputs locations of threats to pipeline.�The advanced automatic image analysis system is intended to be capable of detecting construction equipment along the ROW of pipelines with a very high degree of accuracy in comparison with manual threat identification by a human analyst. �
2
Wilson, D., Steven Peckham, Max Krackow, Sora Haley, Sophia Bragdon, and Jay Clausen. Discriminating buried munitions based on physical models for their thermal response. Engineer Research and Development Center (U.S.), 2025. https://doi.org/10.21079/11681/49749.
Full textAbstract:
Munitions and other objects buried near the Earth’s surface can often be recognized in infrared imagery because their thermal and radiative properties differ from the surrounding undisturbed soil. However, the evolution of the thermal signature over time is subject to many complex interacting processes, including incident solar radiation, heat conduction in the ground, longwave radiation from the surface, and sensible and latent heat exchanges with the atmosphere. This complexity makes development of robust classification algorithms particularly challenging. Machine-learning algorithms, although increasingly popular, often require large training datasets including all environments to which they will be applied. Algorithms incorporating an understanding of the physical processes underlying the thermal signature potentially provide improved performance and mitigate the need for large training datasets. To that end, this report formulates a simplified model for the energy exchange near the ground and describes how it can be incorporated into maximum-likelihood ratio and Bayesian classifiers capable of distinguishing buried objects from their surroundings. In particular, a version of the Bayesian classifier is formulated that leverages the differing amplitude and phase response of a buried object over a 24-hour period. These algorithms will be tested on experimental data in a future study.