To see the other types of publications on this topic, follow the link: Hyperspectral remote sensing.
Dissertations / Theses on the topic 'Hyperspectral remote sensing'
Create a spot-on reference in APA, MLA, Chicago, Harvard, and other styles
Consult the top 50 dissertations / theses for your research on the topic 'Hyperspectral remote sensing.'
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 dissertations / theses on a wide variety of disciplines and organise your bibliography correctly.
1
Kam, F. "Classification techniques for hyperspectral remote sensing." Thesis, Department of Informatics and Sensors, 2011. http://dspace.lib.cranfield.ac.uk/handle/1826/6163.
Full textAbstract:
This study concerns with classification techniques in high dimensional space such as
that of Hyperspectral Imaging (HSI) data sets, with objectives of understanding the
strength and weakness of various classifiers and at the same time to study how their
performances can be assessed particularly when there is an absence of ground truth
target map in the data set. The thesis summaries the work that carried out during the
course of this study and it encompasses a brief survey of machine learning and
classification theories, an outline of the HSI instrumentations, data sets that collected in
the study and classification analysis.
It is found that the supervised classifiers such as the Maximum Likelihood (QD) and the
Mahalanobis Distance (FD) classifiers, especially when they are coupled with
techniques like Regularised Discriminant Analysis (RDA) or leave-one-out covariance
estimations (LOOC), have demonstrated excellent performances comparable to that of
the more complicated and computational costly classifiers like the Support Vector
Machine (SVM). This work has also revealed that separability measures such as the
Total Transformed Divergence (TTD) and Total Jeffries-Matusita Distance (TJM) can be
an invaluable method for assessing the goodness of classification in principle. However,
the present methods for the evaluation of the separability measures are insufficient for
achieving this goal and further work in this area is needed. This study has also
confirmed the effectiveness for using RDA and LOOC techniques for a better estimation
of the covariance when the sample size is small, ie when the sample size per class to
band ratio is less than 100.
Through team work this study has contributed partially a number of publications in the
area of hyperspectral imaging and machine visions.
2
Cisz, Adam. "Performance comparison of hyperspectral target detection algorithms /." Online version of thesis, 2006. https://ritdml.rit.edu/dspace/handle/1850/3020.
Full text
3
Thulin, Susanne Maria, and smthulin@telia com. "Hyperspectral Remote Sensing of Temperate Pasture Quality." RMIT University. Mathematical and Geospatial Sciences, 2009. http://adt.lib.rmit.edu.au/adt/public/adt-VIT20090507.163006.
Full textAbstract:
This thesis describes the research undertaken for the degree of Doctor of Philosophy, testing the hypothesis that spectrometer data can be used to establish usable relationships for prediction of pasture quality attributes. The research data consisted of reflectance measurements of various temperate pasture types recorded at four different times (years 2000 to 2002), recorded by three hyperspectral sensors, the in situ ASD, the airborne HyMap and the satellite-borne Hyperion. Corresponding ground-based pasture samples were analysed for content of chlorophyll, water, crude protein, digestibility, lignin and cellulose at three study sites in rural Victoria, Australia. This context was used to evaluate effects of sensor differences, data processing and enhancement, analytical methods and sample variability on the predictive capacity of derived prediction models. Although hyperspectral data analysis is being applied in many areas very few studies on temperate pastures have been conducted and hardly any encompass the variability and heterogeneity of these southern Australian examples. The research into the relationship between the spectrometer data and pasture quality attribute assays was designed using knowledge gained from assessment of other hyperspectral remote sensing and near-infrared spectroscopy research, including bio-chemical and physical properties of pastures, as well as practical issues of the grazing industries and carbon cycling/modelling. Processing and enhancement of the spectral data followed methods used by other hyperspectral researchers with modifications deemed essential to produce better relationships with pasture assay data. As many different methods are in use for the analysis of hyperspectral data several alternative approaches were investigated and evaluated to determine reliability, robustness and suitability for retrieval of temperate pasture quality attributes. The analyses employed included stepwise multiple linear regression (SMLR) and partial least squares regression (PLSR). The research showed that the spectral research data had a higher potential to be used for prediction of crude protein and digestibility than for the plant fibres lignin and cellulose. Spectral transformation such as continuum removal and derivatives enhanced the results. By using a modified approach based on sample subsets identified by a matrix of subjective bio-physical and ancillary data parameters, the performance of the models were enhanced. Prediction models from PLSR developed on ASD in situ spectral data, HyMap airborne imagery and Hyperion and corresponding pasture assays showed potential for predicting the two important pasture quality attributes crude protein and digestibility in hyperspectral imagery at a few quantised levels corresponding to levels currently used in commercial feed testing. It was concluded that imaging spectrometry has potential to offer synoptic, simultaneous and spatially continuous information valuable to feed based enterprises in temperate Victoria. The thesis provide a significant contribution to the field of hyperspectral remote sensing and good guidance for future hyperspectral researchers embarking on similar tasks. As the research is based on temperate pastures in Victoria, Australia, which are dominated by northern hemisphere species, the findings should be applicable to analysis of temperate pastures elsewhere, for example in Western Australia, New Zealand, South Africa, North America, Europe and northern Asia (China).
4
Gao, Jincheng. "Canopy chlorophyll estimation with hyperspectral remote sensing." Diss., Manhattan, Kan. : Kansas State University, 2006. http://hdl.handle.net/2097/252.
Full text
5
Jia, Xiuping Electrical Engineering Australian Defence Force Academy UNSW. "Classification techniques for hyperspectral remote sensing image data." Awarded by:University of New South Wales - Australian Defence Force Academy. School of Electrical Engineering, 1996. http://handle.unsw.edu.au/1959.4/38713.
Full textAbstract:
Hyperspectral remote sensing image data, such as that recorded by AVIRIS with 224 spectral bands, provides rich information on ground cover types. However, it presents new problems in machine assisted interpretation, mainly in long processing times and the difficulties of class training due to the low ratio of number of training samples to the number of bands. This thesis investigates feasible and efficient feature reduction and image classification techniques which are appropriate for hyperspectral image data. The study is reported in three parts. The first concerns a deterministic approach for hyperspectral data interpretation. Multigroup and multiple threshold spectral coding procedures, and associated techniques for spectral matching and classification, are proposed and tested. By coding on subgroups of bands using one or three thresholds, spectral searching and matching becomes simple, fast and free of the need for radiometric correction. Modifications of existing statistical techniques are proposed in the second part of the investigation A block-based maximum likelihood classification technique is developed. Several subgroups are formed from the complete set of spectral bands in the data, based on the properties of global correlation among the bands. Subgroups which are poorly correlated with each other are treated independently using conventional maximum likelihood classification. Experimental results demonstrate that, when using appropriate subgroup sizes, the new method provides a compromise among classification accuracy, processing time and available training pixels. Furthermore, a segmented, and possibly multi-layer, principal components transformation is proposed as a possible feature reduction technique prior to classification, and for effective colour display. The transformation is performed efficiently on each of the highly correlated subgroups of bands independently. Selected features from each transformed subgroup can be then transformed again to achieve a satisfactory data reduction ratio and to generate the three most significant components for colour display. Classification accuracy is improved and high quality colour image display is achieved in experiments using two AVIRIS data sets.
6
Klonowski, Wojciech Mateusz. "Hyperspectral Remote Sensing Applied to Shallow Coastal Waters." Thesis, Curtin University, 2015. http://hdl.handle.net/20.500.11937/48821.
Full textAbstract:
A shallow water reflectance model was developed for application to optical remote sensing in highly diverse and complex coastal environments. A numerical inversion scheme, based on analytical parameterisation, was applied to airborne hyperspectral imagery collected over two regions of the Western Australian coastline; Jurien Bay and the Ningaloo Marine Park. Detailed maps of water quality, water depth and benthic cover classification were derived with a high degree of accuracy as compared to ground truth data.
7
Bishoff, Josef P. "Target detection using oblique hyperspectral imagery : a domain trade study /." Online version of thesis, 2008. http://hdl.handle.net/1850/7834.
Full text
8
Morman, Christopher Joseph. "Hyperspectral Target Detection Performance Modeling." University of Dayton / OhioLINK, 2015. http://rave.ohiolink.edu/etdc/view?acc_num=dayton1446587051.
Full text
9
Lewis, Ryan H. "Topological & network theoretic approaches in hyperspectral remote sensing /." Online version of thesis, 2010. http://ritdml.rit.edu/handle/1850/12274.
Full text
10
Hay, Lorraine. "Variations in modelled and measured hyperspectral remote sensing reflectance." Thesis, University of Strathclyde, 2006. http://oleg.lib.strath.ac.uk:80/R/?func=dbin-jump-full&object_id=21610.
Full textAbstract:
Measurements of hyperspectral surface reflectance, with a spectral range of 350-800nm and sampling interval of 3.3nm, were made in Scottish coastal waters, the Bristol Channel and the Atlantic Ocean. Analysis of the shape of these spectra by normalisation and differentiation revealed three prominent features: (1) the magnitude of the integral between 400-455nm, (2) the width of a trough occurring between 560-615nm, and (3) the peak to trough height between 660-750nm. The characteristics of these features were not determined by individual seawater constituents, but they proved useful as a tool for water type classification. The sign of the integral between 400-455nm discriminated between open ocean and coastal waters, and coastal sub-types could be distinguished by applying cluster analysis to the other three features. The hyperspectral data were integrated over appropriate bandwidths to generate multi-band surface reflectance values which were used to assess the performance of remote sensing algorithms in coastal water. All the chlorophyll algorithms tested (SeaWiFS OC4V4, MODIS Chlor_a_2 and Chlor_a_3, and MERIS OC4E) overestimated the values measured in situ. The MODIS algorithm for absorption by phytoplankton, αphyto(675), performed poorly, but the MODIS algorithm for the absorption by coloured dissolved organic material, αCDOM(400), provided accurate values of the absorption coefficient (R² = 0.91). Algorithm performance was improved when turbid stations, identified using cluster analysis, were removed. Hyperspectral radiometry was also used to investigate variations in chlorophyll fluorescence line height (FLH) with chlorophyll concentration, solar irradiation and seawater composition. FLH and chlorophyll α concentration were not correlated in the coastal waters sampled and variations in the photosynthetically available radiation (PAR), CDOM and suspended sediment concentrations affected the magnitude of FLH observed. A study of (FLH / Chl) under natural, fluctuating irradiances allowed the onset of adaptive non-photochemical quenching to be monitored in situ.
11
Wood, Peter. "Hyperspectral measurement and modelling of marine remote sensing reflectance." Thesis, University of Strathclyde, 1998. http://ethos.bl.uk/OrderDetails.do?uin=uk.bl.ethos.366770.
Full text
12
Emengini, Ebele Josephine. "Hyperspectral and thermal remote sensing of plant stress responses." Thesis, Lancaster University, 2010. http://ethos.bl.uk/OrderDetails.do?uin=uk.bl.ethos.547950.
Full text
13
Meola, Joseph. "A model-based approach to hyperspectral change detection." The Ohio State University, 2011. http://rave.ohiolink.edu/etdc/view?acc_num=osu1320847592.
Full text
14
Flores, Cordova Africa Ixmucane. "Hyperspectral remote sensing of water quality in Lake Atitlan, Guatemala." Thesis, The University of Alabama in Huntsville, 2014. http://pqdtopen.proquest.com/#viewpdf?dispub=1549067.
Full textAbstract:
Lake Atitlan in Guatemala is a vital source of drinking water. The deteriorating conditions of water quality in this lake threaten human and ecological health as well as the local and national economy. Given the sporadic and limited measurements available, it is impossible to determine the changing conditions of water quality. The goal of this thesis is to use Hyperion satellite images to measure water quality parameters in Lake Atitlan. For this purpose in situ measurements and satellite-derived reflectance data were analyzed to generate an algorithm that estimated Chlorophyll concentrations. This research provides for the first time a quantitative application of hyperspectral satellite remote sensing for water quality monitoring in Guatemala. This approach is readily transferable to other countries in Central America that face similar issues in the management of their water resources.
15
Sani, Yahaya. "Determination and monitoring of vegetation stress using hyperspectral remote sensing." Thesis, University of Nottingham, 2013. http://eprints.nottingham.ac.uk/13740/.
Full textAbstract:
Stress causes crops to grow below their potential and this affects the vitality and physiological functioning of the plants at all levels leading to reduction in yield. Remote sensing of vegetation is regarded as a valuable tool for the detection and discrimination of stress, especially over large or sensitive regions. The main aim of the research carried out is to assess the potential of remote sensing to detect CO2 leakage from CCS repositories. Further to this, the capability of remote sensing to discriminate between stresses with similar mode of action is explored. Two stress factors were selected for study: (1) elevated concentrations of soil CO2 in the plant root zone and; (2) herbicide, applied at sub-lethal levels. To understand the effects of soil CO2 and herbicide stress on vegetation reflectance, field experiments were carried out on maize (2009) and barley (2010) to investigate the effects of elevated soil CO2 concentrations and of different levels of herbicide treatments on vegetation growth and canopy reflectance using hyperspectral remote sensing techniques. The findings from this study shows that the average canopy reflectance response of maize and barley to CO2 and herbicide stress were increased reflectance in the visible and decrease in near infra-red region as well as changes in the position and shape of the red-edge. The red-edge first-derivative for barley treated with CO2 were composed of maximum peaks between 716 and 730nm and smaller peaks at 699 and 759nm, the control had peaks at 727 and 730 nm, with similar smaller peaks. Barley treated with herbicide had early peaks (a day after treatment) at 697, 715 and 717nm with a shoulder at 759nm, as the experiment progressed (16 days after treatment) the stress became apparent and the peak remained stationary at 730nm, the magnitude decreased to 712nm at late treatment period (35 days after treatment). The control had single peak at 726nm. CO2 treated maize had double peaks at 718 and 730nm, with secondary peaks at 707 and 794nm. Maize treated with herbicide had maximum peaks at 716 and 723nm, with the shoulder at 759 nm; the peaks were similar with the control plots but decreased in magnitude. The main differences between the treatments were in the shape and positions of the peaks that identify the red-edge. The canopy reflectances of the plants were further analysed using the blue (400-550nm) and red (550-750nm). In these regions the main feature of concern is chlorophyll content. The analysis showed that the band depths of controls plants were deeper compared to the stressed plants which is dependent on the stress and crop type. Other vegetation indices used in this study were the Chlorophyll Normalized Difference Index (Chl NDI), the Pigment Specific Simple Ratio for chlorophyll a and b (PSSRa and PSSRb) and the Physiological Reflectance Index (PRI). The results show that they were promising indicators of early stress detection, some indices performed better than others depending on the stress type, species and duration of stress. Chl NDI was sensitive to high soil CO2 concentration in maize and barley, sub-lethal herbicide treatment at 10% - 40% level in barley and was insensitive to both low CO2 in the barley and maize as well as 10% herbicide treatment in maize. PSSRa was a good indicator of early CO2 stress in maize and high CO2 in barley as well as 10- 40% herbicide treatments. PSSRb could detect high CO2 level in maize and barley and all levels (5-40%) of herbicide treatments. PRI was insensitive to 5% herbicide treatment in barley but sensitive to high CO2 in maize at early stage of the experiment. This study has demonstrated that remote sensing approach could be deployed for discriminating between different stressors using their red-edge first-derivative peaks, band depths and vegetation indices.
16
Tong, Lei. "Constrained Nonnegative Matrix Factorization for Hyperspectral Unmixing and Its Applications." Thesis, Griffith University, 2016. http://hdl.handle.net/10072/367613.
Full textAbstract:
Hyperspectral remote sensing imagery, containing both spatial and spectral information captured by imaging sensors, has been widely used for ground information extraction. Due to the long distance of the imaging sensors to the targets of monitoring and the intrinsic property of sensors, hyperspectral images normally do not have high spatial resolution, which causes mixed responses of various types of ground objects in the images. Therefore, hyperspectral unmixing has become an important technique to decompose mixed pixels into a collection of spectral signatures, or endmembers, and their corresponding proportions, i.e., abundance. Hyperspectral unmixing methods can be mainly divided into three categories: geometric based, statistics based, and sparse regression based. Among these methods, nonnegative matrix factorization (NMF), as one of the statistical methods, has attracted much attention. It treats unmixing as a blind source separation problem, and decomposes image data into endmember and abundance ma- trices simultaneously. However, the NMF algorithm may fall into local minima because the objective function of NMF is a non-convex function. Adding adequate constraint to NMF has become one solution to solve this problem.
In this thesis, we introduce three different constraints for the NMF based hyperspe tral unmixing method. The first constraint is a partial prior knowledge of endmember constraint. It assumes that some endmembers could be treated as known endmembers before unmixing. The proposed model minimizes the differences between the spectral signatures of endmembers being estimated in the image data and the standard signa- tures of known endmembers extracted from a library or detected from the ground. The benefit of this method is that it not only uses the prior knowledge on the unmixing tasks, but also considers the distribution of the real data in the hyperspectral dataset, so that the discrepancy between the prior knowledge and the data can be compromised. Furthermore, the proposed method is general in nature, and can be easily extended to other NMF based hyperspectral unmixing algorithms.Thesis (PhD Doctorate)
Doctor of Philosophy (PhD)
Griffith School of Engineering
Science, Environment, Engineering and Technology
Full Text
17
Kim, Rhae Sung. "Spectral Matching using Bitmap Indices of Spectral Derivatives for the Analysis of Hyperspectral Imagery." The Ohio State University, 2011. http://rave.ohiolink.edu/etdc/view?acc_num=osu1293667753.
Full text
18
Ientilucci, Emmett J. "Hyperspectral sub-pixel target detection using hybrid algorithms and physics based modeling /." Link to online version, 2005. https://ritdml.rit.edu/dspace/handle/1850/1185.
Full text
19
Lalonde, Mark. "The hyperspectral determination of Sphagnum water content in a bog." Thesis, McGill University, 2014. http://digitool.Library.McGill.CA:80/R/?func=dbin-jump-full&object_id=121269.
Full textAbstract:
Sphagnum's strong water-holding capacities, its dominance in bogs, and the overall importance of water in regulating photosynthesis make it a key ecosystem engineer. Though its effectiveness in this context has rarely been tested, Spectral Vegetation Indices (SVIs) derived from hyperspectral data allow for efficient modeling of Sphagnum gravimetric water content over large scales. This study tests whether a linear model relating a SVI to Sphagnum gravimetric water content (i.e. S. capillifolium, S. magellanicum, S. angustifolium/S. fallax, or all Sphagnum species pooled together) can be applied to the landscape level using airborne hyperspectral imagery taken over Mer Bleue Bog, near Ottawa, Ontario, Canada. The depth of a Sphagnum species sample contributing to the reflectance and the vertical distribution of water across a species sample was also analyzed to test the accuracy of water content measurements. Additionally, image SVI data were compared to field SVI data to test the effectiveness of image spectra. Results indicate that light penetrated 1.5 cm in S. capillifolium samples, 1.0 cm in S. magellanicum samples, and 2.5 cm in S. angustifolium/S. fallax samples. Water variability was highest in samples with elevated water contents for every Sphagnum species analyzed. The Normalized Difference Water Index (NDWI) (dimensionless) was the most effective in estimating Sphagnum gravimetric water content of all SVIs (Root Mean Square Error=161.34%, P= 0.000). Image NDWI values mimicked field NDWI values (Root Mean Square Error= 0.000740, P= 0.0000). The application of the NDWI to areas identified as being favorable for Sphagnum growth in an image resulted in a map of Sphagnum gravimetric water content for a given day in a bog.Le contenu d'eau des sphaignes est important pour le fonctionnement des tourbières ombrotrophes, suite aux fortes capacités de rétention d'eau dans ces espèces, la domination de ses espèces dans les tourbières ombrotrophes, et l'importance d'eau dans la régulation de la photosynthèse. L'efficacité des "Spectral Vegetation Indices" (SVIs) dérivée des données hyperspectrales permet une modélisation efficace du contenu d'eau gravimétrique des sphaignes sur de grands échelons. Cependant, l'efficacité des SVIs dans ce contexte a été rarement examinée. Cette étude examine si un modèle linéaire reliant un SVI au contenu d'eau gravimétrique des sphaignes (i.e. S. capillifolium, S. magellanicum, S. angustifolium/S. fallax, ou tous les espèces Sphagnum jumelées ensemble) peut être appliqué au niveau paysagier en utilisant l'imagerie hyperspectrale aérienne prise au-dessus de la tourbière ombrotrophe Mer Bleue, située à proximité d'Ottawa, en Ontario, au Canada. La profondeur d'un échantillon d'une espèce de Sphagnum qui contribue à la réflectance et la distribution verticale d'eau à travers un échantillon a aussi été analysée pour tester l'exactitude des mesures du contenu d'eau. De plus, les données de SVI générées par les images ont été comparées aux données SVI générées par les mesures prises sur le terrain pour examiner l'efficacité des spectres générées par les images. Les résultats indiquent que la lumière a pénétré à une profondeur de 1.5 cm dans les échantillons de S. capillifolium, 1.0 cm dans les échantillons de S. magellanicum, et 2.5 cm dans les échantillons de S. angustifolium/S. fallax. La variabilité d'eau a été le plus prononcée dans les échantillons avec des contenus d'eaux gravimétriques élevées pour chaque espèce de Sphagnum analysée. La "Normalized Difference Water Index" (NDWI) (sans dimension) a été la plus efficace dans l'estimation du contenu d'eau gravimétrique des espèces de tous les SVIs ("Root Mean Square Error"=161.34%, P= 0.000). Les valeurs de NDWI dérivées du terrain ont été semblabes à celles collectionnées sur le terrain (“Root Mean Square Error”= 0.000740, P= 0.0000). L'application du NDWI aux régions identifiées comme étant favorables pour la croissance des sphaignes dans une image a donné, comme résultat, une carte du contenu d'eau gravimétrique des sphaignes pour une journée dans une tourbière ombrotrophe.
20
Loomis, Michael J. "Depth derivation from the Worldview-2 satellite using hyperspectral imagery." Thesis, Monterey, Calif. : Naval Postgraduate School, 2009. http://edocs.nps.edu/npspubs/scholarly/theses/2009/March/09Mar%5FLoomis.pdf.
Full textAbstract:
Thesis (M.S. in Meteorology and Physical Oceanography)--Naval Postgraduate School, March 2009.Thesis Advisor(s): Durkee, Philip A. ; Olsen, Richard C. "March 2009." Description based on title screen as viewed on April 23, 2009. Author(s) subject terms: Remote sensing, hyperspectral, multispectral, bathymetry, Worldview-2, Quickbird. Includes bibliographical references (p. 51-52). Also available in print.
21
Vivone, Gemine. "Multispectral and hyperspectral pansharpening." Doctoral thesis, Universita degli studi di Salerno, 2014. http://hdl.handle.net/10556/1604.
Full textAbstract:
2012-2013Remote sensing consists in measuring some characteristics of an object from a distance. A key example of remote sensing is the Earth observation from sensors mounted on satellites that is a crucial aspect of space programs. The first satellite used for Earth observation was Explorer VII. It has been followed by thousands of satellites, many of which are still working. Due to the availability of a large number of different sensors and the subsequent huge amount of data collected, the idea of obtaining improved products by means of fusion algorithms is becoming more intriguing. Data fusion is often exploited for indicating the process of integrating multiple data and knowledge related to the same real-world scene into a consistent, accurate, and useful representation. This term is very generic and it includes different levels of fusion. This dissertation is focused on the low level data fusion, which consists in combining several sources of raw data. In this field, one of the most relevant scientific application is surely the Pansharpening. Pansharpening refers to the fusion of a panchromatic image (a single band that covers the visible and near infrared spectrum) and a multispectral/hyperspectral image (tens/hundreds bands) acquired on the same area. [edited by author]
XII ciclo n.s.
22
Wang, Jing. "Hyperspectral Image Classification Based on Deep Learning and Module Inspired by Human Attention Mechanism." Thesis, Griffith University, 2020. http://hdl.handle.net/10072/397634.
Full textAbstract:
Hyperspectral imaging technology acquires image data in a number of continuous narrow bands of the electromagnetic wave. The obtained hyperspectral images contain details of spectral re ectance of targets in addition to spatial information. The ability to characterize abundant spectral details of hyperspectral image makes it particularly suitable for remote sensing image analysis. Hyperspectral remote sensing image classi cation is one of the most important applications in remote sensing, and is the main research problem of this thesis.
Researchers have already proposed a large variety of methods for hyperspectral image classi cation in the last few decades, which can be categorized into traditional methods and deep learning based methods. Recently, with the development of high performance computing and collection of large datasets, deep learning methods have been state of the art in hyperspectral image classi cation. Most of the existing deep learning methods take in the hyperspectral image and learn discriminant features in plain convolutional or fully connected layers. This learning manner treats all raw pixels and extracted features equally. However, human brains do not perform recognition task with equal consideration of every involved element. For recognition or classi cation tasks, it is possible that some parts of inputs or features are more important, while others are useless. Our visual system has the capability of attending to the signi cant aspects and ignoring irrelevant components. This has greatly contributed to our cognition ability and e ciency. Inspired by the attention mechanism of human brain, we design corresponding attention modules in the context of arti cial neural network for hyperspectral image classi cation. In addition, human visual system is a universal feature extractor and classi er in the sense that we can perform classi cation across multiple image styles, modalities and distributions. On the contrary, current deep learning based hyperspectral classi - cation paradigms require an individual model for every data domain. This is expensive and ine cient. Following similar philosophy of attention mechanism, we design domain attention modules for multi-domain hyperspectral image classi cation.
In this thesis, we propose three attention modules for deep learning based hyperspectral image classi cation. In the rst work, we introduce attention based feature weighting networks for improving the classi cation accuracy of current plain neural networks. In a deep network for hyperspectral application, a hierarchy of spectral or spatial features are extracted layer by layer. Each layer contains the same semantic level of features. To model the importance of features in the same level, attention modules are designed by branching from current feature maps. In the attention branch, three steps are executed: summarizing information from current layer, modeling relationship among the features with fully connected or convolution layers, and outputting weighting masks to be multiplied with the original features. We propose feature weighting attention modules for spectral CNN, spatial CNN and spectral-spatial CNN, respectively.
In the second work, we design attention modules speci cally attending to the bands of hyperspectral image. Compared to hidden features extracted in hidden layers of neural networks which have less interpretability and physical meaning, spectral bands of hyperspectral images correspond directly to real wavelength in the physical world. Thus attending to bands has special importance in a couple of aspects. First, it in uences the design and cost of hyperspectral sensor. Second, it is directly related to the dimension of the obtained raw data. Our band attention module can perform both band weighting and band selection. For band weighting, it has the ability to assign sample-wise weights to hyperspectral images and can interfere with the feature learning process in the early stage. For band selection, we carefully design an additional parallel input to the attention module for obtaining xed selected band sets and an activation function for ltering insigni cant bands in the training process.
In the third work, we propose attention mechanisms to address multi-domain hyperspectral image classi cation. Di erent hyperspectral datasets have di erent data modalities, statistical distributions, or spectral dimensionalities. This brings signi cant challenges for a single network to learn all the tasks. The domain shift problem can be alleviated by adjusting the network towards the property of speci c domains. To this end, domain attention modules are designed to attend to the domain of the input data for adapting the network accordingly. Two domain attention modules: hard domain attention and soft domain attention are proposed. For the hard domain attention network, the attention mechanism is implemented by a muxer switch. According to the labels of data domain, a set of small domain speci c adapters are selected and connected to a main backbone network. In this way, the majority of network parameters are shared by all domains with only a small number of domain speci c parameters. For the soft domain attention network, we build the attention mechanism based on squeeze and excitation (SE) block. Several parallel SE blocks are applied as the feature adapters. On top of them, a higher level domain attention SE block is placed to achieve domain assignment.Thesis (PhD Doctorate)
Doctor of Philosophy (PhD)
School of Info & Comm Tech
Science, Environment, Engineering and Technology
Full Text
23
Bellante, Gabriel John. "Hyperspectral remote sensing as a monitoring tool for geologic carbon sequestration." Thesis, Montana State University, 2011. http://etd.lib.montana.edu/etd/2011/bellante/BellanteG1211.pdf.
Full textAbstract:
The contemporary global climate crisis demands mitigation technologies to curb atmospheric greenhouse gas emissions, principally carbon dioxide (COâ‚‚). Geologic carbon sequestration (GCS) is a method by which point source COâ‚‚ emissions are purified and deposited in subsurface geologic formations for long-term storage. Accompanying this technology is the inherent responsibility to monitor these large-scale subsurface reservoirs for COâ‚‚ leaks to ensure safety to local environments and inhabitants, as well as to alleviate global warming. Elevated COâ‚‚ levels in soil are known to cause anoxic conditions in plant roots, thereby interfering with plant respiration and inducing a stress response that could possibly be remotely sensed using aerial imagery. Airborne remote sensing technology has the potential to monitor large land areas at a relatively small cost compared to alternative methods. In 2010, an aerial campaign was conducted during the height of the growing season to obtain an image time series that could be used to identify and characterize COâ‚‚ stress in vegetation from a simulated COâ‚‚ leak. An unsupervised classification was performed to classify COâ‚‚ stressed vegetation as a result of the subsurface injection. Furthermore, a spectral index was derived to amplify the COâ‚‚ stress signal and chart vegetation health trajectories for pixels affected by the COâ‚‚ release. A theoretical framework was developed for analysis strategies that could be implemented to detect a COâ‚‚ leak using aerial hyperspectral imagery with minimal a priori knowledge. Although aerial detection of COâ‚‚ stressed vegetation was possible while no other physiological plant stressors were present, the spectral distinction between vegetation stress agents would have important implications for the appropriate timing that GCS monitoring using remote sensing data could commence. A greenhouse experiment was devised to compare the spectral responses of alfalfa plants to COâ‚‚ and water stress in order to reveal whether COâ‚‚ leak detection is possible when soil water availability is highly variable or during periods of drought. Spectral discernment of a COâ‚‚ leak appears to be possible when soil water is spatially variable and during moderate drought conditions with remote sensing instruments that are sensitive to reflectance in the short wave infrared, where water absorption features related to leaf water content occur.
24
Chen, T. "Hyperspectral imaging for the remote sensing of blood oxygenation and emotions." Thesis, Cranfield University, 2012. http://dspace.lib.cranfield.ac.uk/handle/1826/7502.
Full textAbstract:
This PhD project is a basic research and it concerns with how human’s physiological features, such as tissue oxygen saturation (StO2), can be captured from a stand-off distance and then to understand how this remotely acquired physiological feature can be deployed for biomedical and other applications. This work utilises Hyperspectral Imaging (HSI) within the diffuse optical scattering framework, to assess the StO2 in a contactless remote sensing manner. The assessment involves a detailed investigation about the wavelength dependence of diffuse optical scattering from the skin as well as body tissues, under various forms of optical absorption models. It is concluded that the threechromophore extended Beer Lambert Law model is better suited for assessing the palm and facial tissue oxygenations, especially when spectral data in the wavelengths region of [516-580]nm is used for the analysis. A first attempt of using the facial StO2 to detect and to classify people’s emotional state is initiated in this project. The objective of this work is to understand how strong emotions, such as distress that caused by mental or physical stimulations, can be detected using physiological feature such as StO2. Based on data collected from ~20 participants, it is found that the forehead StO2 is elevated upon the onset of strong emotions that triggered by mental stimulation. The StO2 pattern in the facial region upon strong emotions that are initiated by physical stimulations is quite complicated, and further work is needed for a better understanding of the interplays between bodily physique, individual’s health condition and blood transfusion control mechanism. Most of this work has already been published and future research to follow up when the author returns back to China is highlighted.
25
Villa, Alberto. "Advanced spectral unmixing and classification methods for hyperspectral remote sensing data." Phd thesis, Université de Grenoble, 2011. http://tel.archives-ouvertes.fr/tel-00767250.
Full textAbstract:
La thèse propose des nouvelles techniques pour la classification et le démelange spectraldes images obtenus par télédétection iperspectrale. Les problèmes liées au données (notammenttrès grande dimensionalité, présence de mélanges des pixels) ont été considerés et destechniques innovantes pour résoudre ces problèmes. Nouvelles méthodes de classi_cationavancées basées sur l'utilisation des méthodes traditionnel de réduction des dimension etl'integration de l'information spatiale ont été développés. De plus, les méthodes de démelangespectral ont été utilisés conjointement pour ameliorer la classification obtenu avec lesméthodes traditionnel, donnant la possibilité d'obtenir aussi une amélioration de la résolutionspatial des maps de classification grace à l'utilisation de l'information à niveau sous-pixel.Les travaux ont suivi une progression logique, avec les étapes suivantes:1. Constat de base: pour améliorer la classification d'imagerie hyperspectrale, il fautconsidérer les problèmes liées au données : très grande dimensionalité, presence demélanges des pixels.2. Peut-on développer méthodes de classi_cation avancées basées sur l'utilisation des méthodestraditionnel de réduction des dimension (ICA ou autre)?3. Comment utiliser les differents types d'information contextuel typique des imagés satellitaires?4. Peut-on utiliser l'information données par les méthodes de démelange spectral pourproposer nouvelles chaines de réduction des dimension?5. Est-ce qu'on peut utiliser conjointement les méthodes de démelange spectral pour ameliorerla classification obtenu avec les méthodes traditionnel?6. Peut-on obtenir une amélioration de la résolution spatial des maps de classi_cationgrace à l'utilisation de l'information à niveau sous-pixel?Les différents méthodes proposées ont été testées sur plusieurs jeux de données réelles, montrantresultats comparable ou meilleurs de la plus part des methodes presentés dans la litterature.
26
Zabalza, Jaime. "Feature extraction and data reduction for hyperspectral remote sensing Earth observation." Thesis, University of Strathclyde, 2015. http://oleg.lib.strath.ac.uk:80/R/?func=dbin-jump-full&object_id=26015.
Full textAbstract:
Earth observation and land-cover analysis became a reality in the last 2-3 decades thanks to NASA airborne and spacecrafts such as Landsat. Inclusion of Hyperspectral Imaging (HSI) technology in some of these platforms has made possible acquiring large data sets, with high potential in analytical tasks but at the cost of advanced signal processing. In this thesis, effective/efficient feature extraction methods are proposed. Initially, contributions are introduced for efficient computation of the covariance matrix widely used in data reduction methods such as Principal Component Analysis (PCA). By taking advantage of the cube structure in HSI, onsite and real-time covariance computation is achieved, reducing memory requirements as well. Furthermore, following the PCA algorithm, a novel method called Folded-PCA (Fd-PCA) is proposed for efficiency while extracting both global and local features within the spectral pixels, achieved by folding the spectral samples from vector to matrix arrays. Inspired by Empirical Mode Decomposition (EMD) methods, a recent and promising algorithm, Singular Spectrum Analysis (SSA), is introduced to hyperspectral remote sensing, performing extraction of features in the spectral (1D-SSA) and also the spatial (2D-SSA) domain. By successfully suppressing the noise and enhancing the useful signal, more effective feature extraction and data classification are achieved. Furthermore, a fast implementation of the SSA methods is also proposed, leading to reduction of computational complexity. In addition, combination of both spectral- and spatial-domain exploitation is also included, comprising data reduction. Finally, promising Deep Learning (DL) approaches are evaluated by the analysis of Stacked AutoEncoders (SAEs) for feature extraction and data reduction, introducing a method called Segmented-SAE (S-SAE), working in local regions of the spectral domain. Preliminary results have validated its great potential in this context.
27
Nandi, David Anil. "The use of hyperspectral imaging for remote sensing, and the development of a novel hyperspectral imager." Thesis, Durham University, 2014. http://etheses.dur.ac.uk/11824/.
Full textAbstract:
This thesis determines the potential uses of a novel technology in hyperspectral remote sensing, by testing the capabilities of a prototype imaging spectrometer that was built using microslice technology. These capabilities are compared to those of current hyperspectral remote sensing instruments in the context of the requirements for various remote sensing applications. Due to the wide variety of potential applications for hyperspectral imaging, any unique capability of a new instrument is likely to improve a current application, or even develop a new one. The use of microslice technology allows a 2-dimensional eld of view (FoV) to be imaged simultane ously with a wide spectral range. Modelling of the remote sensing performance of the spectrometer shows that this enables it to achieve a signal to noise ratio (SNR) an order of magnitude higher than conventional hyperspectral instruments. The prototype microslice spectrometer images in the 475-650 nm wavelength range at 7 nm spectral resolution. It also images an instantaneous eld of view (IFoV) of 260 x 52 mrad, at a spatial resolution of 2.6 mrad. Classication techniques are used on ground based laboratory and eld test data from the instrument to demonstrate that it can accurately identify some mineral, vegetation, and water pollutant samples. Various trade-os can theoretically be performed on the prototype specications to develop an instru ment with particular capabilities for a specic application. This novel design means that a greater detector area is required than for conventional designs; but the 2-dimentsional FoV gives greater trade-o exibility, in particular allowing the SNR to enter into the trade-o equation. This unique capability was found to lend itself to two applications in particular: detecting water pollutants in rivers, and detecting hydrocarbons contamination of ecosystems.
28
Khurshid, Khawaja Shahid. "Estimation and mapping of wheat crop chlorophyll content using hyperion hyperspectral data." Thesis, University of Ottawa (Canada), 2004. http://hdl.handle.net/10393/26676.
Full textAbstract:
The estimation of chlorophyll content is an essential biochemical parameter to track the main developmental stages and yield of cereals relevant for precision agriculture. Traditional techniques for chlorophyll content measurements are time consuming, expensive and laborious. Measurements at field level have proven to be a good alternative, but their use is limited due to extensive sampling designs and techniques. Several spectral chlorophyll indices have been developed to estimate chlorophyll content both at the leaf and canopy level using remote sensing data. A methodology of using spectral chlorophyll indices to estimate chlorophyll content from laboratory and satellite hyperspectral data was carried out in this study for wheat crops. The application of this technique under agricultural field conditions has been very limited and not rigorously validated for wheat crops. The main objective of this study is to validate the chlorophyll content estimation using spectral chlorophyll indices, and to examine the potential for chlorophyll content estimation using hyperspectral remote sensing data in the context of precision agriculture. (Abstract shortened by UMI.)
29
Lin, Huang-De Hennessy. "Parametric projection pursuits for dimensionality reduction of hyperspectral signals in target recognition applications." Master's thesis, Mississippi State : Mississippi State University, 2004. http://library.msstate.edu/etd/show.asp?etd=etd-12162003-202048.
Full text
30
Kamalesh, Vidhya Lakshmi. "Vegetation parameter retrieval from hyperspectral, multiple view angle PROBA/CHRIS data." Thesis, Swansea University, 2011. http://ethos.bl.uk/OrderDetails.do?uin=uk.bl.ethos.678514.
Full text
31
Yeu, Yeon. "FEATURE EXTRACTION FROM HYPERSPECTRAL IMAGERY FOR OBJECT RECOGNITION." The Ohio State University, 2011. http://rave.ohiolink.edu/etdc/view?acc_num=osu1306848130.
Full text
32
Pinnel, Nicole. "A method for mapping submerged macrophytes in lakes using hyperspectral remote sensing." [S.l.] : [s.n.], 2007. http://mediatum2.ub.tum.de/doc/604557/document.pdf.
Full text
33
Yetkin, Erdem. "Alteration Identification By Hyperspectral Remote Sensing In Sisorta Gold Prospect (sivas-turkey)." Phd thesis, METU, 2009. http://etd.lib.metu.edu.tr/upload/12611049/index.pdf.
Full textAbstract:
Imaging spectrometry data or hyperspectral imagery acquired using airborne systems have been used in the geologic community since the early 1980&rsquos and represent a mature technology. The solar spectral range 0.4&ndash
2.5 &
#956
m provides abundant information about hydroxyl-bearing minerals, sulfates and carbonates common to many geologic units and hydrothermal alteration assemblages. Satellite based Hyperion image data is used to implement and test hyperspectral processing techniques to identify alteration minerals and associate the results with the geological setting. Sisorta gold prospect is characterized by porphyry related epithermal and mesothermal alteration zones that are mapped through field studies. Image specific corrections are applied to obtain error free image data. Extensive field mapping and spectroscopic survey are used to identify nine endmembers from the image. Partial unmixing techniques are applied and used to assess the endmembers. Finally the spectral correlation mapper is used to map the endmembers which are kaolinite, dickite, halloysite, illite, montmorillonite and alunite as clay group and hematite, goethite and jarosite as the iron oxide group. The clays and iron oxides are mapped with approximately eighty percent accuracy. The study introduces an image specific algorithm for alteration minerals identification and discusses the outcomes within the geological perspective.
34
Yan, Lin. "REGION-BASED GEOMETRIC ACTIVE CONTOUR FOR CLASSIFICATION USING HYPERSPECTRAL REMOTE SENSING IMAGES." The Ohio State University, 2011. http://rave.ohiolink.edu/etdc/view?acc_num=osu1315344636.
Full text
35
Aqdus, Syed Ali. "Airborne multispectral and hyperspectral remote sensing techniques in archaeology a comparative study /." Thesis, Thesis restricted. Connect to e-thesis to view abstract, 2009. http://theses.gla.ac.uk/812/.
Full textAbstract:
Thesis (Ph.D.) - University of Glasgow, 2009.Ph.D. thesis submitted to the Faculty of Physical Sciences, Department of Geographical and Earth Sciences and the Faculty of Arts, Department of Archaeology, University of Glasgow, 2009. Includes bibliographical references. Print version also available.
36
Alam, Fahim Irfan. "Deep Feature Learning for Spectral-Spatial Classification of Hyperspectral Remote Sensing Images." Thesis, Griffith University, 2019. http://hdl.handle.net/10072/386535.
Full textAbstract:
The recent advances in aerial- and satellite-based hyperspectral imaging sensor technologies
have led to an increased availability of Earth's images with high spatial and
spectral resolution, which opened the door to a large range of important applications.
Hyperspectral imaging records detailed spectrum of the received light in each spatial position
in the image, in which each pixel contains a highly detailed representation of the
reflectance of the materials present on the ground, and a better characterization in terms
of geometrical details. Since different substances exhibit different spectral signatures,
the abundance of informative content conveyed in the hyperspectral images permits an
improved characterization of different land coverage. Therefore, hyperspectral imaging
emerged as a well-suited technology for accurate image classi fication in remote sensing.
In spite of that, a signi ficantly increased complexity of the analysis introduces a series of
challenges that need to be addressed on a serious note. In order to fully exploit the potential
offered by these sensors, there is a need to develop accurate and effective models
for spectral-spatial analysis of the recorded data.
This thesis aims at presenting novel strategies for the analysis and classifi cation of hyperspectral
remote sensing images, placing the focal point on the investigation on deep
networks for the extraction and integration of spectral and spatial information. Deep
learning has demonstrated cutting-edge performances in computer vision, particularly in
object recognition and classi cation. It has also been successfully adopted in hyperspectral
remote sensing domain as well. However, it is a very challenging task to fully utilize
the massive potential of deep models in hyperspectral remote sensing applications since
the number of training samples is limited which limits the representation capability of a
deep model. Furthermore, the existing architectures of deep models need to be further
investigated and modifi ed accordingly to better complement the joint use of spectral
and spatial contents of hyperspectral images. In this thesis, we propose three different
deep learning-based models to effectively represent spectral-spatial characteristics of
hyperspectral data in the interest of classifi cation of remote sensing images.
Our first proposed model focuses on integrating CRF and CNN into an end-to-end
learning framework for classifying images. Our main contribution in this model is the
introduction of a deep CRF in which the CRF parameters are computed using CNN and
further optimized by adopting piecewise training. Furthermore, we address the problem
of over fitting by employing data augmentation techniques and increased the size of the
training samples for training deep networks. Our proposed 3DCNN-CRF model can
be trained to fully exploit the usefulness of CRF in the context of classi fication by
integrating it completely inside of a deep model.
Considering that the separation of constituent materials and their abundances provide
detailed analysis of the data, our second algorithm investigates the potential of using
unmixing results in deep models to classify images. We extend an existing region based
structure preserving non-negative matrix factorization method to estimate groups of
spectral bands with the goal to capture subtle spectral-spatial distribution from the
image. We subsequently use these important unmixing results as input to generate
superpixels, which are further represented by kernel density estimated probability distribution
function. Finally, these abundance information-guided superpixels are directly
supplied into a deep model in which the inference is implicitly formulated as a recurrent
neural network to perform the eventual classifi cation.
Finally, we perform a detailed investigation on the possibilities of adopting generative
adversarial models into hyperspectral image classifi cation. We present a GAN-based
spectral-spatial method that primarily focuses on signifi cantly improving the multiclass
classi cation ability of the discriminator of GAN models. In this context, we propose to
adopt the triplet constraint property and extend it to build a useful feature embedding
for remote sensing images for use in classi cation. Furthermore, our proposed Triplet-
3D-GAN model also includes feedback from discriminator's intermediate features to
improve the quality of the generator's sample generation process.Thesis (PhD Doctorate)
Doctor of Philosophy (PhD)
School of Info & Comm Tech
Science, Environment, Engineering and Technology
Full Text
37
Fountanas, Leonidas. "Principal components based techniques for hyperspectral image data." Thesis, Monterey, Calif. : Springfield, Va. : Naval Postgraduate School ; Available from National Technical Information Service, 2004. http://library.nps.navy.mil/uhtbin/hyperion/04Dec%5FFountanas.pdf.
Full text
38
MAKKI, IHAB. "Hyperspectral Imaging for Landmine Detection." Doctoral thesis, Politecnico di Torino, 2017. http://hdl.handle.net/11583/2700516.
Full textAbstract:
This PhD thesis aims at investigating the possibility to detect landmines using hyperspectral imaging. Using this technology, we are able to acquire at each pixel of the image spectral data in hundreds of wavelengths. So, at each pixel we obtain a reflectance spectrum that is used as fingerprint to identify the materials in each pixel, and mainly in our project help us to detect the presence of landmines.
The proposed process works as follows: a preconfigured drone (hexarotor or octorotor) will carry the hyperspectral camera. This programmed drone is responsible of flying over the contaminated area in order to take images from a safe distance. Various image processing techniques will be used to treat the image in order to isolate the landmine from the surrounding. Once the presence of a mine or explosives is suspected, an alarm signal is sent to the base station giving information about the type of the mine, its location and the clear path that could be taken by the mine removal team in order to disarm the mine.
This technology has advantages over the actually used techniques:
• It is safer because it limits the need of humans in the searching process and gives the opportunity to the demining team to detect the mines while they are in a safe region.
• It is faster. A larger area could be cleared in a single day by comparison with demining techniques
• This technique can be used to detect at the same time objects other than mines such oil or minerals.
First, a presentation of the problem of landmines that is expanding worldwide referring to some statistics from the UN organizations is provided. In addition, a brief presentation of different types of landmines is shown. Unfortunately, new landmines are well camouflaged and are mainly made of plastic in order to make their detection using metal detectors harder. A summary of all landmine detection techniques is shown to give an idea about the advantages and disadvantages of each technique.
In this work, we give an overview of different projects that worked on the detection of landmines using hyperspectral imaging. We will show the main results achieved in this field and future work to be done in order to make this technology effective.
Moreover, we worked on different target detection algorithms in order to achieve high probability of detection with low false alarm rate. We tested different statistical and linear unmixing based methods. In addition, we introduced the use of radial basis function neural networks in order to detect landmines at subpixel level. A comparative study between different detection methods will be shown in the thesis.
A study of the effect of dimensionality reduction using principal component analysis prior to classification is also provided. The study shows the dependency between the two steps (feature extraction and target detection). The selection of target detection algorithm will define if feature extraction in previous phase is necessary.
A field experiment has been done in order to study how the spectral signature of landmine will change depending on the environment in which the mine is planted. For this, we acquired the spectral signature of 6 types of landmines in different conditions: in Lab where specific source of light is used; in field where mines are covered by grass; and when mines are buried in soil. The results of this experiment are very interesting. The signature of two types of landmines are used in the simulations. They are a database necessary for supervised detection of landmines. Also we extracted some spectral characteristics of landmines that would help us to distinguish mines from background.
39
Arkun, Sedat. "Hyperspectral remote sensing and the urban environment : a study of automated urban feature extraction using a CASI image of high spatial and spectral resolution." Title page, contents, research aims and abstract only, 1999. http://web4.library.adelaide.edu.au/theses/09ARM/09arma721.pdf.
Full text
40
Ling, Bohua. "Estimates of canopy nitrogen content in heterogeneous grasslands of Konza Prairie by hyperspectral remote sensing." Thesis, Kansas State University, 2013. http://hdl.handle.net/2097/15616.
Full textAbstract:
Master of ScienceDepartment of Geography
Douglas Goodin
Hyperspectral data has been widely used for estimates of canopy biochemical content over the past decades. Most of these studies were conducted in forests or crops with relatively uniform canopies. Feasibility of the use of hyperspectral analysis in heterogeneous canopies with diverse plant species and canopy structures remains uncertain. Spectral data at the canopy level, with mixed background noise, canopy biochemical and biophysical properties create more problems in spectral analysis than that at the leaf level. Complications of heterogeneous canopies make biochemical retrieval through remote sensing even more difficult due to more uneven spatial distribution of biochemical constituents. The objective of my research was to map canopy nitrogen content in tallgrass prairie with mixed canopies by means of hyperspectral data from in-situ and airborne measurements. Research efforts were divided into three steps: (1) the green leaf area index (LAI) retrieval, given LAI is an important parameter in scaling nitrogen content from leaves to canopies; (2) canopy nitrogen modeling from analysis of in-situ hyperspectral data; and (3) canopy nitrogen mapping based on aerial hyperspectral imagery. Research results revealed that a fine chlorophyll absorption feature in the green-yellow region at wavelengths of 562 – 600 nm was sensitive to canopy nitrogen status. Specific spectral features from the normalized spectral data by the first derivative or continuum removal in this narrow spectral region could be selected by multivariate regression for nitrogen modeling. The optimal nitrogen models with high predictive accuracy measured as low values of root-mean-square error (RMSE) were applied to the aerial hyperspectral imagery for canopy nitrogen mapping during the growth seasons from May to September. These maps would be of great value in studies on the interactions between canopy vegetation quality and grazing patterns of large herbivores in tallgrass prairie.
41
Meerdink, Susan Kay. "Remote Sensing of Plant Species Using Airborne Hyperspectral Visible-Shortwave Infrared and Thermal Infrared Imagery." Thesis, University of California, Santa Barbara, 2019. http://pqdtopen.proquest.com/#viewpdf?dispub=13420575.
Full textAbstract:
In California, natural vegetation is experiencing an increasing amount of stress due to prolonged droughts, wildfires, insect infestation, and disease. Remote sensing technologies provide a means for monitoring plant species presence and function temporally across landscapes. In this his dissertation, I used hyperspectral visible shortwave infrared (VSWIR), hyperspectral thermal (TIR), and hyperspectral VSWIR + broadband TIR imagery to derive key observations of plant species across a gradient of environmental conditions and time frames. In Chapter 2, I classified plant species using hyperspectral VSWIR imagery from 2013–2015 spring, summer, and fall. Plant species maps had the highest classification accuracy using spectra from a single date (mean kappa 0.80–0.86). The inclusion of spectra from other dates decreased accuracy (mean kappa 0.78–0.83). Leave-one-out analysis emphasized the need to have spectra from the image date in the classification training, otherwise classification accuracy dropped significantly (mean kappa 0.31–0.73). In Chapter 3, I used hyperspectral TIR imagery to determine the extent that high precision spectral emissivity and canopy temperature can be exploited for vegetation research at the canopy level. I found that plant species show distinct spectral separation at the leaf level, but separability among species is lost at the canopy level. However, species’ canopy temperatures exhibited different distributions among dates and species. Variability in canopy temperatures was largely explained by LiDAR derived canopy structural attributes (e.g. canopy density) and the surrounding environment (e.g. presence of pavement). In Chapter 4, I used combined hyperspectral VSWIR and broadband TIR imagery to monitor plant stress during California’s 2013–2015 severe drought. The temperature condition index (TCI) was calculated to measure plant stress by using plant species’ surface minus air temperature distributions across dates. Plant stress was not evenly distributed across the landscape or time with lower elevation open shrub/meadows, showing the largest amount of stress in June 2014, and August 2015 imagery. Plant stress spatial variability across the study area was related to a slope’s aspect with highly stressed plants located on south or south-southwest facing slopes. Overall, this dissertation quantifies the ability to temporally study plant species using hyperspectral VSWIR, hyperspectral TIR, and combined VSWIR+TIR imagery. This analysis supports a range of current and planned missions including Surface Biology and Geology (SBG), Environmental Mapping and Analysis Program (EnMAP), National Ecological Observatory Network (NEON), Hyperspectral Thermal Emission Spectrometer (HyTES), and ECOsystem Spaceborne Thermal Radiometer Experiment on Space Station (ECOSTRESS).
42
Balashova, Natalia. "Remote Sensing for Organic and Conventional Corn Assessment." Bowling Green State University / OhioLINK, 2015. http://rave.ohiolink.edu/etdc/view?acc_num=bgsu1446803968.
Full text
43
Zhou, Bo. "Application of hyperspectral remote sensing in detecting and mapping Sericea lespedeza in Missouri." Diss., Columbia, Mo. : University of Missouri-Columbia, 2007. http://hdl.handle.net/10355/5051.
Full textAbstract:
Thesis (M.A.)--University of Missouri-Columbia, 2007.The entire dissertation/thesis text is included in the research.pdf file; the official abstract appears in the short.pdf file (which also appears in the research.pdf); a non-technical general description, or public abstract, appears in the public.pdf file. Title from title screen of research.pdf file (viewed on November 9, 2007) Includes bibliographical references.
44
White, Davina Cherie. "Hyperspectral remote sensing of canopy scale vegetation stress associated with buried gas pipelines." Thesis, University of Newcastle Upon Tyne, 2007. http://ethos.bl.uk/OrderDetails.do?uin=uk.bl.ethos.484812.
Full textAbstract:
This applied study investigates the capability of field and airborne hyperspectral remote sensing to detect the spectral and spatial characteristics of sub-surface soil disturbance from associated overlying subtle canopy scale vegetation stress features. A 9 km stretch of buried gas pipeline, in Aberdeenshire, was used as a real world case study. Hyperspectral techniques, in particular derivative analysis, have a number of advantages over existing broadband approaches under operational conditions, reducing changes in illumination or background reflectance and aiding in suppressing the continuum caused by other leaf biochemicals. Various peaks in red-edge derivative spectra are also well correlated with plant pigment concentrations thus being able to detect more subtle stress features than conventional broadband reflectance approaches. The capability of these hyperspectral techniques to detect a generic stress response to gas induced soil oxygen depletion, which could also result from soil compaction and water logging due to sub-surface soil disturbance, has shown great potential under controlled conditions. However, their transferability to heterogeneous, canopy scale, field conditions for operational applications has not been published yet. This thesis aims to develop a rigorous method for spatially intensive field spectroradiometry acquisition to identify the full spatial and spectral (VIS-NIR) characteristics of subtle, canopy scale, surface vegetation stress features that maybe indicative of sub-surface soil disturbance to inform operational sensor specifications. In order to achieve this aim field spectroscopy data of barley, wheat, oilseed rape and grassland were acquired at selected transects perpendicular to a buried gas pipeline. The application of derivative analysis, vegetation band ratios and in particular the Smith et al (2004) 725:702 nm ratio, Lagrangian red-edge and continuum removal are evaluated to identify the optimal hyperspectral analytical technique for detecting vegetation stress associated sub-surface soil disturbance under operational conditions. Moreover, the ability of operational airborne hyperspectral sensors to detect the same stress features in the field data is investigated through field spectroradiometry data simulating sensor spectral Gaussian point spread functions and acquired CASI-2 imagery of the study area. First derivative analysis coupled with the 723:700 and 725:702 nm ratios was the most effective hyperspectral approach for detecting vegetation stress associated with pipeline soil disturbance. The 725:702 nm ratio of Smith et at. (2004) and a 723:700 nm ratio performed consistently well detecting stress for all sites over two consecutive field seasons under different cropping regimes, barley being particularly stress sensitive. The ratios exhibited a parabolic trend of decreasing ratio values with proximity to the pipeline, whilst being insensitive to soil background effects, intimating their transferability to heterogeneous field conditions. Field spectroradiometry data simulating the default channel settings for CASI-2, AVIRIS and programmable 725:702 and 723:700 nm wavelengths for CASI-2, AVIRIS and Eagle hyperspectral airborne sensors revealed that channel centre wavelength positions influenced the sensors ability to detect stress. The CASI-2 713:703 and 743:703 nm default channels were able to distinguish vegetation stress associated with pipeline earthworks to the same degree as the 725:702 and 723:700 nm wavelengths. Under operational conditions the CASI-2 743:703 nm ratio was also capable of detecting barley field stress identified by the original field spectra 723:700 and 725:702 nm ratios.
45
Arellano, Mora Paul-Nelson. "Hyperspectral Remote Sensing for Detecting Vegetation Affected by Hydrocarbons in the Amazon Forest." Thesis, University of Leicester, 2014. http://hdl.handle.net/2381/40508.
Full textAbstract:
This thesis seeks to understand the effects of hydrocarbons on the vegetation of tropical forests. It explores hyperspectral methods to detect changes in biophysical and biochemical parameters of vegetation affected by hydrocarbons in the Amazon rainforest of Ecuador. The literature review revealed that experiments in the laboratory, showed that in specific species hydrocarbons caused a reduced level of chlorophyll content, which is an indicator of stress. However, it was unclear whether the same effect would be observed in tropical forests. Fieldwork was conducted in several sites of the Amazon forest of Ecuador to establish whether this was the case. Foliage samples were collected in sites located within oil spills and also from pristine forest in the Yasuni National Park. More than 1,100 leaves from three different levels of the vertical canopy profile (upper, medium and understory) were analysed for biophysical, biochemical and spectral properties. A second-order polynomial chlorophyll content model was estimated based on several published calibrations models which use a portable chlorophyll meter. Modelled chlorophyll content showed high correlations with methods using reflectance indices (0.76) and the inversion process of the PROSPECT radiative transfer model (0.71). The analysis of biophysical and biochemical parameters at the three canopy levels of vegetation growing near hydrocarbons leakages showed decreasing levels of foliar chlorophyll content which suggest a reduced photosynthetic activity, higher levels of water content, which may explain the thicker leaves in the upper canopy, and thinner leaves in the understory. Based on these results, hyperspectral Hyperion and CHRIS-Proba satellite images were used to explore the potential of several vegetation indices to detect the symptoms of vegetation affected by hydrocarbons. The results indicated that a combination of an index sensitive to chlorophyll content at canopy level (Sum Green) with the NDVI index (Normalized Difference Vegetation Index) are suitable to detect vegetation affected by hydrocarbons. Those indices accurately identified vegetation growing near sites polluted by the petroleum industry and also when applied to an area affected by hydrocarbons from natural macro-seepages, and areas where hydrocarbons may be near the surface. Two new vegetation indices are proposed to identify vegetation affected by hydrocarbon pollution. Those indices showed sensitivity to differentiate secondary forest polluted and non-polluted. Chlorophyll content maps were computed based on an approach which uses the MTCI (MERIS Terrestrial Chlorophyll Index) at leaf level and scaled up to canopy level. The results of this research contribute to knowledge of regarding forest degradation. The approach could be used to detect hydrocarbon seepages as indicators of petroleum reservoirs, as well as significant pollution from oil spills in forest ecosystems. Moreover, the parameters for hydrocarbon stressed vegetation could be employed in a carbon cycle model to explore the impacts of hydrocarbon pollution on the carbon dioxide and water fluxes from tropical forests which are crucial for the carbon and water cycles.
46
CELESTI, MARCO. "Development of novel methods to evaluate vegetation status from multi-source remote sensing data." Doctoral thesis, Università degli Studi di Milano-Bicocca, 2018. http://hdl.handle.net/10281/199119.
Full textAbstract:
Negli ultimi decenni il telerilevamento iperspettrale è stato utilizzato per stimare con successo informazioni sulle proprietà delle piante a diverse scale di indagine. Questa tesi si focalizza sull’uso di dati telerilevati iperspettrali per stimare lo stato di salute della vegetazione a livello di canopy. Recentemente, il telerilevamento passivo della fluorescenza della clorofilla indotta dal Sole (F) è emerso come un ambito scientifico di grande interesse per la studio del comportamento dinamico della fotosintesi. F è un prodotto dell’assorbimento della luce da parte della clorofilla a, emesso come radiazione elettromagnetica nel rosso e nel vicino infrarosso (≈ 640 nm to 850 nm), ed è relazionato allo stato energetico dei fotosistemi. Oltre che dalla fisiologia F è influenzata anche dalla struttura della foglia e della canopy, dalla concentrazione dei pigmenti e dalle condizioni di illuminazione/meteorologiche, e la sua interpretazione univoca è tuttora complessa. Questo guida l’interesse nell’esplorare metriche in grado di isolare l’informazione legata alla fisiologia presente nel dato di F. In questa tesi ho analizzato due casi studio: i) una foresta di pino loblolly (Parker Tract, North Carolina, U.S.A.) dove ho investigato la variazione di F e delle metriche derivate al variare dei processi fisiologici legati all’età delle piante; e ii) un esperimento di stress indotto dove è stata inibita la fotosintesi di un tappeto erboso.
Nel primo caso di studio, utilizzato dati iperspettrali acquisiti con il sensore aereo HyPlant per caratterizzare la F emessa da 18 stand coetanei di pino di èta compresa tra 3 e 46 anni, e per calcolare la radiazione fotosinteticamente attiva assorbita (APAR). Ho calcolato gli yield di F normalizzando F per la APAR. I risultati mostrano che nel loblolly: i) gli yield di F nel rosso e nel vicino infrarosso cambiano al variare dell’età dello stand, i giovani pini dissipano più F nel rosso rispetto a quelli più maturi (fino al 60% in più) e il declino dello yield di F nel rosso con l’età dello stand è più pronunciato di quello di F; ii) il declino dello yield di F nel rosso può essere relazionato alla limitazione idraulica che si manifesta nel loblolly durante la crescita.
Nel secondo caso di studio tre plot di 9m x 12m di un tappeto erboso sono stati trattati con diverse dosi di Chlortoluron. Questo erbicida inibisce la fotosintesi bloccando la catena di trasporto degli elettroni. Ho utilizzato dati acquisiti a terra con spettroradiometri ad alta risoluzione , immagini aeree con i sensori HyPlant e TASI-600, e misure di scambi gassosi, per studiare le dinamiche a breve termine dell’efficienza fotosintetica della vegetazione, indotte dallo stress. I risultati mostrano che subito dopo il trattamento si è verificato un rapido incremento di F, dello yield di F, della temperatura della canopy e del Photosynthetic Reflectance Index (PRI). Successivamente è stato osservato un decremento dose-specifico di F e del PRI, assieme a una riduzione del contenuto di clorofilla a e di altri indici di vegetazione legati ai pigmenti.
A partire dai dati spettrali acquisiti a terra sul tappeto erboso, ho invertito numericamente una versione semplificata del modello SCOPE, per stimare per la prima contemporaneamente lo spettro completo di F, lo yield di F, assieme ai principali parametri della vegetazione che controllano l’assorbimento e il riassorbimento della luce. L’effetto del contenuto di pigmenti, delle proprietà strutturali della foglia e della canopy, e la fisiologia sono state discriminate con successo. La loro osservazione combinata nel tempo ha portato al riconoscimento di pattern dinamici di risposta e di adattamento allo stress.Hyperspectral Remote Sensing (RS) data have been exploited in the last decades to successfully retrieve information about plant properties at different scales. This thesis focuses on the use of state of the art hyperspectral RS data to retrieve vegetation status at canopy level, using both experimental and modeled data. In the last years, RS of Sun-induced chlorophyll fluorescence (F) emerged as a novel and promising scientific field for studying the dynamic behavior of photosynthesis. F is a physical side product of chlorophyll a light absorption that is emitted as an electromagnetic radiation in the red and far-red spectral regions (≈ 640 nm to 850 nm), and it is related to the energetic status of the photosystems. Nevertheless, apart from physiology F is concurrently influenced by leaf and canopy structure, pigment concentration and weather/illumination conditions, and its unambiguous interpretation is still challenging. This drives the interest in exploring F-derived metrics able to disentangle the physiological information from the remotely sensed F signal. In this thesis I analyzed data from two case studies: i) a managed loblolly pine forest (Parker Tract forest, North Carolina, U.S.A.) where I investigated how F and F-derived metrics vary with age-related changes in plant physiology; and ii) an experiment of induced stress where the photosynthesis of a homogeneous lawn was inhibited with a herbicide treatment. In the first case-study, I used hyperspectral data acquired with the HyPlant airborne sensor to characterize the F emission of 18 evenly aged stands in a range from 3 to 46 years old, and to calculate the Absorbed Photosynthetically Active Radiation (APAR). I computed the F yields in the red and far-red regions normalizing the corresponding F data for the APAR. Results show that in loblolly pine: i) red F and red F yield change with stand age, younger loblolly pines dissipate more red F than older one (up to 60% more) and the decline of F yield with stand age is more pronounced than that for red F; ii) the decline of red F yield can be related to the increase in water limitation occurring as loblolly trees grow in age and height. In the second case study three 9 m × 12 m plots of a homogeneous lawn were treated with different doses of a commercial formulation of Chlortoluron. This herbicide inhibits photosynthesis by blocking the electron transport chain in the photosynthetic apparatus. I exploited data collected on the ground with very high resolution spectroradiometers, airborne images collected with the HyPlant and the TASI-600 sensors, as well as canopy-level gas exchange measurements collected with closed chambers, in order to detect short-term dynamics of photosynthetic efficiency in vegetation, induced by stress. Results show that immediately after the application there was a rapid increase of F, F yields and the photosynthetic reflectance index (PRI). Canopy temperature also increased after the application of Dicuran. A later decrease of fluorescence and PRI was observed together with a reduction of chlorophyll a content and a drop of pigment-related vegetation indices. Moreover, the dosage of Chlortoluron had an impact in the dynamics of F. Starting from the ground level hyperspectral measurements over the lawn, I inverted numerically a simplified version of the SCOPE model to concurrently retrieve F, F yield and several biochemical and biophysical parameters of the vegetation from apparent reflectance data. For the first time the full spectrum of canopy F, the fluorescence yield, as well as the main vegetation parameters that control light absorption and reabsorption, were retrieved concurrently using canopy-level high resolution apparent reflectance measurements. The effects of pigment content, leaf/canopy structural properties and physiology were effectively discriminated. Their combined observation over time led to the recognition of dynamic patterns of stress adaptation and stress recovery.
47
Cole, Elizabeth. "High resolution remote sensing for landscape scale restoration of peatland." Thesis, University of Manchester, 2013. https://www.research.manchester.ac.uk/portal/en/theses/high-resolution-remote-sensing-for-landscape-scale-restoration-of-peatland(a3777efd-0f95-4fdc-bac3-13d5020d4105).html.
Full textAbstract:
Upland peatlands provide vital ecosystem services, especially carbon storage and biodiversity. However, large areas of peatland are heavily degraded in the UK. When peat becomes exposed the potential for it to actively sequester carbon is greatly reduced and carbon stores are rapidly lost through erosion. Peatland restoration is a tool that addresses the government public service agreement targets for biodiversity, and soil and water protection in uplands. Blanket bogs are a UK Biodiversity Action Plan priority habitat. Many areas fall under designations for sites of protection under the EU habitats directive which is aimed at bringing the areas into ‘favourable condition’.The Moors for the Future Partnership is restoring large areas of badly eroded peat in the Peak District National Park to stabilise the surface and re-establish ecosystem functions. Monitoring is of pivotal importance to judge the success of the restoration work. This project assesses the suitability of high resolution remote sensing as an alternative monitoring tool to traditional field based plot surveys which are both time consuming and expensive. Remote sensing has been seen as a potential tool for mapping and monitoring peatlands, but to date the application of high spatial and spectral resolution remote sensing to monitoring peatland restoration has not been fully investigated. A floristic restoration trajectory has been established using a statistical classification (TWINSPAN) of vegetation cover data combined with expert knowledge of previous restoration, and autecology of the moorland species. Hyperspectral classification techniques were applied, including: Spectral Angle Mapping (SAM); Support Vector Machines (SVM); and maximum likelihood classification using both Minimum Noise Fraction (MNF), and narrow band vegetation indices. A successful classification of the restoration succession has been achieved. A predictive model for vegetation cover of plant functional types has been produced using a Partial Least Squares Regression and applied to the whole restoration site at the landscape-scale. RMSEs of between 10 and 16% indicate that the models can be used as a useful operational tool. A spectral library of key moorland species and their phenological response has been established using field spectroscopy in parallel to the image analysis. This has enabled the suggestion that the species are most separable from one another in July and it is recommended that this is the optimal month for remote sensing monitoring. This has facilitated the development of a set of recommendations for the most appropriate vegetation indices to use throughout the year depending species to be differentiated. High spatial and spectral resolution remote sensing data is needed to successfully characterise the vegetation response to restoration management in the upland peatland environment.
48
Snirer, Eva. "Hyperspectral remote sensing of individual gravesites - exploring the effects of cadaver decomposition on vegetation and soil spectra." Thesis, McGill University, 2014. http://digitool.Library.McGill.CA:80/R/?func=dbin-jump-full&object_id=121458.
Full textAbstract:
The detection of clandestine graves is an emerging tool in hyperspectral remote sensing. Though previous studies have demonstrated that it is possible to use hyperspectral remote sensing techniques in detection of mass graves, there is a lack of studies demonstrating the feasibility to utilize this same technology for the detection of individual burial sites. This thesis summarizes the first year of a multi-year study to ascertain the detectable changes to vegetation and soil spectra caused by the chemicals released from a single decomposing body. Eighteen pig (Sus scrofa) carcasses were buried in a temperate environment in Ottawa, ON. Three scenarios were examined; surface body deposition, 30 cm, and 90 cm soil cover. A Twin Otter aircraft with hyperspectral sensors covering the visible to shortwave infrared range was used to collect the imagery. In addition to the airborne sensor, a portable spectroradiometer was used to collect plant and soil spectra in the lab (the soil and plant samples were collected coincidentally with the airborne imagery). Through chemical analysis of the soil collected both before site set up and coincidentally with the airborne imagery, I was able to determine the changes in chemistry and spectra caused by decomposing cadavers rather than just soil disturbance. Statistical analysis of the Chlorophyll and Carotenoids extraction demonstrates separability of vegetation into three categories: 1) background, 2) disturbed soil, shallow and deep graves, and 3) surface burials. Statistical analysis of the vegetation spectra corresponded to the chemical analysis in differentiating between background, disturbed soil, shallow and deep graves, and surface burials, as well analysis of the soil spectra allowed for separation into disturbed soil, shallow and deep graves, and surface burials.La détection des fosses clandestines (tombes) est un domain d'étude récent (un nouvel outil) dans la télédétection hyperspectrale. Bien que des études antérieures ont démontrés qu'il est possible d'utiliser des techniques de télédétection hyperspectrale pour la localisation des fosses communes, il y a un manque d'études démontrant la faisabilité d'utiliser cette même technologie pour la détection des tombes individuelles. Cette thèse se porte sur la première année d'une étude a long terme, elle constate que des changements sont détectables au niveau de la réponse spectrale de la végétation et de du sol. Ces changements sont causés par les produits chimiques libérées par un corps en décomposition. Dix-huit carcasses de porc (Sus scrofa) ont été enterrées dans un environnement tempéré à Ottawa, ON. Trois scénarios ont été examinés: la décomposition d'un corps déposé en surface, un corps enterré à 30 cm dans le sol, et un corps enterré à 90 cm dans le sol. Un avion Twin Otter avec des capteurs hyperspectrales couvrant les ondes visible à l'infrarouge du spectre électromagnétique ont été utilisés pour recueillir des images aériennes du site. En plus, un spectroradiomètre portable a été utilisé pour recueillir des signatures spectrales des plantes et du sol en laboratoire (les échantillons ont été collectés en même temps que l'imagerie aérienne). Grâce à l'analyse chimique du sol faite avant et après l'établissement du site, ainsi qu'en même temps que l'imagerie aérienne, j'ai déterminer que certains changements chimiques ainsi que des changements dans la réflectance sont causés par la décomposition des cadavres plutôt que par la perturbation du sol. L'analyse statistique des niveaux de chlorophylle et des caroténoïdes démontre une séparabilité de la végétation en trois catégories: 1) le fond, 2) les sols perturbés, les tombes peu profondes et les tombes profondes, et 3) les corps déposé en surface. L'analyse statistique des signatures spectrales de la végétation confirme à l'analyse chimique pour différencier entre le fond, le sol perturbé, les tombes peu profondes et profondes, et les corps décomposant en surface. L'analyse des signatures spectres de sol a aussi permis de séparer entre un sol perturbé, une tombe peu profonde ou profonde, ou un « enterrement » de surface.
49
Romanko, Matthew. "Remote Sensing in Precision Agriculture: Monitoring Plant Chlorophyll, and Soil Ammonia, Nitrate, and Phosphate in Corn and Soybean Fields." Bowling Green State University / OhioLINK, 2017. http://rave.ohiolink.edu/etdc/view?acc_num=bgsu1490964339514842.
Full text
50
Hilker, Thomas. "Estimation of photosynthetic light-use efficience from automated multi-angular spectroradiometer measurements of coastal Douglas-fir." Thesis, University of British Columbia, 2008. http://hdl.handle.net/2429/2685.
Full textAbstract:
Global modeling of gross primary production (GPP) is a critical component of climate change research. On local scales, GPP can be assessed from measuring CO₂ exchange above the plant canopy using tower-based eddy covariance (EC) systems. The limited footprint inherent to this method however, restricts observations to relatively few discrete areas making continuous predictions of global CO₂ fluxes difficult. Recently, the advent of high resolution optical remote sensing devices has offered new possibilities to address some of the scaling issues related to GPP using remote sensing. One key component for inferring GPP spectrally is the efficiency (ε) with which plants can use absorbed photosynthetically active radiation to produce biomass. While recent years have seen progress in measuring ε using the photochemical reflectance index (PRI), little is known about the temporal and spatial requirements for up-scaling these findings continuously throughout the landscape. Satellite observations of canopy reflectance are subject to view and illumination effects induced by the bi-directional reflectance distribution function(BRDF) which can confound the desired PRI signal. Further uncertainties include dependencies of PRI on canopy structure, understorey, species composition and leaf pigment concentration. The objective of this research was to investigate the effects of these factors on PRI to facilitate the modeling of GPP in a continuous fashion. Canopy spectra were sampled over a one-year period using an automated tower-based, multi-angular spectroradiometer platform (AMSPEC), designed to sample high spectral resolution data. The wide range of illumination and viewing geometries seen by the instrument permitted comprehensive modeling of the BRDF. Isolation of physiologically induced changes in PRI yielded a high correlation (r²=0.82, p<0.05) to EC-measured ε, thereby demonstrating the capability of PRI to model ε throughout the year. The results were extrapolated to the landscape scale using airborne laser-scanning (light detection and ranging, LiDAR) and high correlations were found between remotely-sensed and EC-measured GPP (r²>0.79, p<0.05). Permanently established tower-based canopy reflectance measurements are helpful for ongoing research aimed at up-scaling ε to landscape and global scales and facilitate a better understanding of physiological cycles of vegetation and serve as a calibration tool for broader band satellite observations.