Pub Date : 2003-10-27DOI: 10.1109/WARSD.2003.1295170
J.P. Kerckes
The quantitative forecasting of spectral imaging system performance is an important capability at every stage of system development including system requirement definition, system design, and even sensor operation. However, due to the complexity of the end-to-end remote sensing system involved, the analyses are often performed piecemeal by various groups, and then merged together. The ability to understand system sensitivities also supports the best use of an operational system and is thus desirable. It was with this perspective and goal to better perform end-to-end remote sensing system analyses that work was undertaken in the late 1980s to develop models that can be efficiently used as part of the system design and operation. Both simulation and analytical models were developed. The simulation approach has the advantage of creating an actual image, which can include non-linear effects or specified instrument artifacts, while the analytical approach has the benefit of being much simpler computationally and amenable to large numbers of comprehensive trade studies. In the mid 1990s, the analytical approach was extended to the case of unresolved object detection. By taking advantage of the spectral information, objects and materials that are not spatially resolved in the imagery can still be detected and identified. Subsequently, this model, which was developed for the reflective solar part of the optical spectrum, was extended to the thermal infrared. Here, surfaces are characterized not only by, their spectral emissivity means and covariances, but also their physical temperature mean and standard deviation. The model has also been extended to explore linear unmixing applications through the implementation of multiple classes in the target class. This has allowed the exploration of the role of class variability in unmixing abundance estimation. This paper provides an overview of this model development activity as well as show examples of how it can be used in the various applications. Examples include the impact of system parameters sub-pixel object detection and abundance estimation applications. Key capabilities as well as limitations of this analytical modeling approach are identified. System understanding developed through the use of the model is highlighted and the future enhancements are discussed.
{"title":"Spectral imaging system performance forecasting","authors":"J.P. Kerckes","doi":"10.1109/WARSD.2003.1295170","DOIUrl":"https://doi.org/10.1109/WARSD.2003.1295170","url":null,"abstract":"The quantitative forecasting of spectral imaging system performance is an important capability at every stage of system development including system requirement definition, system design, and even sensor operation. However, due to the complexity of the end-to-end remote sensing system involved, the analyses are often performed piecemeal by various groups, and then merged together. The ability to understand system sensitivities also supports the best use of an operational system and is thus desirable. It was with this perspective and goal to better perform end-to-end remote sensing system analyses that work was undertaken in the late 1980s to develop models that can be efficiently used as part of the system design and operation. Both simulation and analytical models were developed. The simulation approach has the advantage of creating an actual image, which can include non-linear effects or specified instrument artifacts, while the analytical approach has the benefit of being much simpler computationally and amenable to large numbers of comprehensive trade studies. In the mid 1990s, the analytical approach was extended to the case of unresolved object detection. By taking advantage of the spectral information, objects and materials that are not spatially resolved in the imagery can still be detected and identified. Subsequently, this model, which was developed for the reflective solar part of the optical spectrum, was extended to the thermal infrared. Here, surfaces are characterized not only by, their spectral emissivity means and covariances, but also their physical temperature mean and standard deviation. The model has also been extended to explore linear unmixing applications through the implementation of multiple classes in the target class. This has allowed the exploration of the role of class variability in unmixing abundance estimation. This paper provides an overview of this model development activity as well as show examples of how it can be used in the various applications. Examples include the impact of system parameters sub-pixel object detection and abundance estimation applications. Key capabilities as well as limitations of this analytical modeling approach are identified. System understanding developed through the use of the model is highlighted and the future enhancements are discussed.","PeriodicalId":395735,"journal":{"name":"IEEE Workshop on Advances in Techniques for Analysis of Remotely Sensed Data, 2003","volume":"28 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2003-10-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122574368","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2003-10-27DOI: 10.1109/WARSD.2003.1295181
J. Cipar, T. Cooley, R. Lockwood
We use AVIRIS data collected at Fort A. P. Hill, Virginia, to evaluate how well airborne hyperspectral imagery can be used to distinguish vegetation land covers. Fort A. P. Hill is located in east-central Virginia and is heavily forested with a mix of deciduous and coniferous species native to the mid-Atlantic region. The location and extent of the forest species is documented in a land cover database compiled by the Fort for planning and resource protection purposes. The AVIRIS data set consists of several low-altitude (3.7-m GSD) flight lines on two dates: November 1999 and September 2001. Our goal is to characterize the both the natural variability of vegetation land covers using mathematical and biophysical metrics and to assess differences between land covers for classification purposes.
我们使用在弗吉尼亚州A. P. Hill堡收集的AVIRIS数据来评估航空高光谱图像用于区分植被土地覆盖的效果。a . P. Hill堡位于弗吉尼亚州中东部，森林茂密，既有大西洋中部地区的落叶物种，也有针叶物种。森林物种的位置和范围记录在由堡垒为规划和资源保护目的编制的土地覆盖数据库中。AVIRIS数据集包括1999年11月和2001年9月两个日期的几条低空(3.7米GSD)航线。我们的目标是利用数学和生物物理指标来描述植被土地覆盖的自然变异性，并评估土地覆盖之间的差异以进行分类。
{"title":"Distinguishing vegetation land covers using hyperspectral imagery","authors":"J. Cipar, T. Cooley, R. Lockwood","doi":"10.1109/WARSD.2003.1295181","DOIUrl":"https://doi.org/10.1109/WARSD.2003.1295181","url":null,"abstract":"We use AVIRIS data collected at Fort A. P. Hill, Virginia, to evaluate how well airborne hyperspectral imagery can be used to distinguish vegetation land covers. Fort A. P. Hill is located in east-central Virginia and is heavily forested with a mix of deciduous and coniferous species native to the mid-Atlantic region. The location and extent of the forest species is documented in a land cover database compiled by the Fort for planning and resource protection purposes. The AVIRIS data set consists of several low-altitude (3.7-m GSD) flight lines on two dates: November 1999 and September 2001. Our goal is to characterize the both the natural variability of vegetation land covers using mathematical and biophysical metrics and to assess differences between land covers for classification purposes.","PeriodicalId":395735,"journal":{"name":"IEEE Workshop on Advances in Techniques for Analysis of Remotely Sensed Data, 2003","volume":"81 1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2003-10-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128042955","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2003-10-27DOI: 10.1109/WARSD.2003.1295216
Chein-I. Chang
Hyperspectral imagery offers a means of uncovering enormous spectral information that can be used for various applications in data exploitation. How effectively such information is used affects the way image analysis algorithms are designed. In this paper, we take up this issue and focus on algorithms designed and developed for target detection and classification in hyperspectral imagery. In order to effectively characterize the information available before and after the data are processed, the a priori information and a posteriori information are used in accordance with how the information is obtained. A piece of information is referred to as a priori information if it is provided by known knowledge before data are processed. On the other hand, a piece of information is referred to as a posteriori information if it is unknown a priori, but can be obtained directly from the data in an unsupervised fashion during the course of data processing. Since a priori information is known beforehand, it can be further decomposed into two types of information, desired and undesired a priori information. The desired a priori information is the knowledge that will assist, improve and enhance data analysis, whereas the undesired a priori information is the knowledge that hinders, interferes or destructs analysis during data processing. This paper investigates how these three types of information play their roles in design and development of several hyperspectral target detection and classification algorithms. Experiments are also conducted to validate their utility.
{"title":"How to effectively utilize information to design hyperspectral target detection and classification algorithms","authors":"Chein-I. Chang","doi":"10.1109/WARSD.2003.1295216","DOIUrl":"https://doi.org/10.1109/WARSD.2003.1295216","url":null,"abstract":"Hyperspectral imagery offers a means of uncovering enormous spectral information that can be used for various applications in data exploitation. How effectively such information is used affects the way image analysis algorithms are designed. In this paper, we take up this issue and focus on algorithms designed and developed for target detection and classification in hyperspectral imagery. In order to effectively characterize the information available before and after the data are processed, the a priori information and a posteriori information are used in accordance with how the information is obtained. A piece of information is referred to as a priori information if it is provided by known knowledge before data are processed. On the other hand, a piece of information is referred to as a posteriori information if it is unknown a priori, but can be obtained directly from the data in an unsupervised fashion during the course of data processing. Since a priori information is known beforehand, it can be further decomposed into two types of information, desired and undesired a priori information. The desired a priori information is the knowledge that will assist, improve and enhance data analysis, whereas the undesired a priori information is the knowledge that hinders, interferes or destructs analysis during data processing. This paper investigates how these three types of information play their roles in design and development of several hyperspectral target detection and classification algorithms. Experiments are also conducted to validate their utility.","PeriodicalId":395735,"journal":{"name":"IEEE Workshop on Advances in Techniques for Analysis of Remotely Sensed Data, 2003","volume":"56 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2003-10-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126249010","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2003-10-27DOI: 10.1109/WARSD.2003.1295175
Bor-Chen Kuo, L. Ko, Jinn-Min Yang, Chia-Hao Pai
In this paper, a new sequential feature extraction and classification algorithm is proposed for improving the classification accuracy of reject region data.
为了提高拒绝区域数据的分类精度，提出了一种新的序列特征提取与分类算法。
{"title":"Combining feature extractions and classifiers for multispectral data classification","authors":"Bor-Chen Kuo, L. Ko, Jinn-Min Yang, Chia-Hao Pai","doi":"10.1109/WARSD.2003.1295175","DOIUrl":"https://doi.org/10.1109/WARSD.2003.1295175","url":null,"abstract":"In this paper, a new sequential feature extraction and classification algorithm is proposed for improving the classification accuracy of reject region data.","PeriodicalId":395735,"journal":{"name":"IEEE Workshop on Advances in Techniques for Analysis of Remotely Sensed Data, 2003","volume":"23 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2003-10-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128493926","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2003-10-27DOI: 10.1109/WARSD.2003.1295217
Q. Du
This paper addresses unsupervised band selection for hyperspectral image analysis. The proposed approach is based on high-order moments. Such moments indicate the deviation of probability distribution function of an image from the Gaussian distribution, so the selected bands have higher chances to contain important target information. Since the bands with close moment values can be very similar, a band similarity measurement is incorporated into the band selection technique to further select most distinct bands using the criterion of divergence. The number of bands to be selected is pre-estimated using a Neyman-Pearson detection theory-based eigen-thresholding approach. The performance of such a band selection technique is evaluated by the detection and classification performance using the selected bands, i.e., the capability of preserving the target information in the original image data.
{"title":"Band selection and its impact on target detection and classification in hyperspectral image analysis","authors":"Q. Du","doi":"10.1109/WARSD.2003.1295217","DOIUrl":"https://doi.org/10.1109/WARSD.2003.1295217","url":null,"abstract":"This paper addresses unsupervised band selection for hyperspectral image analysis. The proposed approach is based on high-order moments. Such moments indicate the deviation of probability distribution function of an image from the Gaussian distribution, so the selected bands have higher chances to contain important target information. Since the bands with close moment values can be very similar, a band similarity measurement is incorporated into the band selection technique to further select most distinct bands using the criterion of divergence. The number of bands to be selected is pre-estimated using a Neyman-Pearson detection theory-based eigen-thresholding approach. The performance of such a band selection technique is evaluated by the detection and classification performance using the selected bands, i.e., the capability of preserving the target information in the original image data.","PeriodicalId":395735,"journal":{"name":"IEEE Workshop on Advances in Techniques for Analysis of Remotely Sensed Data, 2003","volume":"37 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2003-10-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131592939","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2003-10-27DOI: 10.1109/WARSD.2003.1295189
M. Kalacska, G. Sánchez-Azofeifa, B. Rivard, J. Calvo-Alvarado
A non-linear transition function (Lorentzian cumulative function) best represented the relationship between leaf area index (LAI) and spectral vegetation indices (SVI) calculated from a Landsat ETM+ image from a tropical moist forest. The three parameters of the function (transition height, center and half-width) describe the sensitivity of the index to a range of LAI values. From the SVIs tested, the Modified Single Ratio (MSR) had best sensitivity to LAI in this environment being sensitive to changes in LAI from 0.0-4.7.
{"title":"Estimation of transition function parameters to evaluate the sensitivity of vegetation indices to leaf area index in a tropical moist forest","authors":"M. Kalacska, G. Sánchez-Azofeifa, B. Rivard, J. Calvo-Alvarado","doi":"10.1109/WARSD.2003.1295189","DOIUrl":"https://doi.org/10.1109/WARSD.2003.1295189","url":null,"abstract":"A non-linear transition function (Lorentzian cumulative function) best represented the relationship between leaf area index (LAI) and spectral vegetation indices (SVI) calculated from a Landsat ETM+ image from a tropical moist forest. The three parameters of the function (transition height, center and half-width) describe the sensitivity of the index to a range of LAI values. From the SVIs tested, the Modified Single Ratio (MSR) had best sensitivity to LAI in this environment being sensitive to changes in LAI from 0.0-4.7.","PeriodicalId":395735,"journal":{"name":"IEEE Workshop on Advances in Techniques for Analysis of Remotely Sensed Data, 2003","volume":"44 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2003-10-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124005882","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2003-10-27DOI: 10.1109/WARSD.2003.1295166
D. Landgrebe
This paper begins with some brief historical comments to set the stage for a discussion of the long term potential for land remote sensing technology. This is followed by comments about what is needed to accelerate the achievement of this potential. The discussion is concluded with what concomitant development is needed with regard to a hyperspectral data analysis system.
{"title":"Multispectral land sensing: where from, where to?","authors":"D. Landgrebe","doi":"10.1109/WARSD.2003.1295166","DOIUrl":"https://doi.org/10.1109/WARSD.2003.1295166","url":null,"abstract":"This paper begins with some brief historical comments to set the stage for a discussion of the long term potential for land remote sensing technology. This is followed by comments about what is needed to accelerate the achievement of this potential. The discussion is concluded with what concomitant development is needed with regard to a hyperspectral data analysis system.","PeriodicalId":395735,"journal":{"name":"IEEE Workshop on Advances in Techniques for Analysis of Remotely Sensed Data, 2003","volume":"30 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2003-10-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126844053","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2003-10-27DOI: 10.1109/WARSD.2003.1295223
A. Katartzis, I. Vanhamel, H. Sahli
We propose a new multispectral image classification method, based on a Markovian model, defined on the hierarchy of a multiscale region adjacency graph. The paper describes the main principles of our method and illustrates classification results on a set of artificial and remote sensing images, together with qualitative and quantitative comparisons with a variety of multi-and single-resolution Bayesian classification approaches.
{"title":"A hierarchical Markovian model for multiscale region-based classification of multispectral images","authors":"A. Katartzis, I. Vanhamel, H. Sahli","doi":"10.1109/WARSD.2003.1295223","DOIUrl":"https://doi.org/10.1109/WARSD.2003.1295223","url":null,"abstract":"We propose a new multispectral image classification method, based on a Markovian model, defined on the hierarchy of a multiscale region adjacency graph. The paper describes the main principles of our method and illustrates classification results on a set of artificial and remote sensing images, together with qualitative and quantitative comparisons with a variety of multi-and single-resolution Bayesian classification approaches.","PeriodicalId":395735,"journal":{"name":"IEEE Workshop on Advances in Techniques for Analysis of Remotely Sensed Data, 2003","volume":"14 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2003-10-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124389453","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2003-10-27DOI: 10.1109/WARSD.2003.1295211
M. M. Dundar, D. Landgrebe
In this study a supervised classifier based on the kernel implementation of the Bayes rule is introduced. The proposed technique first suggests an implicit nonlinear transformation of the data into a feature space and then seeks to fit normal distributions having a common covariance matrix onto the mapped data. The use of kernel concept in this process gives us the flexibility required to model complex data structures that originate from a wide-range of class conditional distributions. Although the decision boundaries in the new feature space are piece-wise linear, these corresponds to powerful nonlinear boundaries in the original input space. For the data we considered we have obtained some encouraging results.
{"title":"A kernel-based supervised classifier for the analysis of hyperspectral data","authors":"M. M. Dundar, D. Landgrebe","doi":"10.1109/WARSD.2003.1295211","DOIUrl":"https://doi.org/10.1109/WARSD.2003.1295211","url":null,"abstract":"In this study a supervised classifier based on the kernel implementation of the Bayes rule is introduced. The proposed technique first suggests an implicit nonlinear transformation of the data into a feature space and then seeks to fit normal distributions having a common covariance matrix onto the mapped data. The use of kernel concept in this process gives us the flexibility required to model complex data structures that originate from a wide-range of class conditional distributions. Although the decision boundaries in the new feature space are piece-wise linear, these corresponds to powerful nonlinear boundaries in the original input space. For the data we considered we have obtained some encouraging results.","PeriodicalId":395735,"journal":{"name":"IEEE Workshop on Advances in Techniques for Analysis of Remotely Sensed Data, 2003","volume":"86 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2003-10-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130802907","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2003-10-27DOI: 10.1109/WARSD.2003.1295167
Larry Ryan
Thousands of students worldwide use MultiSpec for land cover analysis.
全世界成千上万的学生使用MultiSpec进行土地覆盖分析。
{"title":"MultiSpec in the classroom","authors":"Larry Ryan","doi":"10.1109/WARSD.2003.1295167","DOIUrl":"https://doi.org/10.1109/WARSD.2003.1295167","url":null,"abstract":"Thousands of students worldwide use MultiSpec for land cover analysis.","PeriodicalId":395735,"journal":{"name":"IEEE Workshop on Advances in Techniques for Analysis of Remotely Sensed Data, 2003","volume":"5 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2003-10-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132335692","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：