Applied Imagery Pattern Recognition Workshop, 2002. Proceedings.最新文献

{"title":"Using optical imagery to enhance radar tracking performance","authors":"T. Cushman, J. VanDamme, J. Perretta, R. Bonneau, Mark D. Barnell","doi":"10.1109/AIPR.2002.1182275","DOIUrl":"https://doi.org/10.1109/AIPR.2002.1182275","url":null,"abstract":"A difficult problem with ground moving target indicator (GMTI) radar detection is how consistently to track targets moving through non-homogeneous regions of clutter such as forest and urban boundaries. Although attempts have been made to mitigate this detection problem using terrain mapping data, such data does not give current clutter information due to changes in vegetation, roads, buildings, and seasonal variations. We propose to use electro-optical imagery to enhance the detection performance of GMTI radar We use a multiresolution Markov model to represent both target and background clutter. This multiresolution structure allows us to match GMTI clutter accurately with the geographically registered electro-optical imagery for consistent moving target detection through clutter boundary areas.","PeriodicalId":379110,"journal":{"name":"Applied Imagery Pattern Recognition Workshop, 2002. Proceedings.","volume":"5 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2002-10-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115551203","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}

{"title":"Video-rate visible to LWIR hyperspectral imaging and image exploitation","authors":"M. Dombrowski, Jagmohan Bajaj, P. Willson","doi":"10.1109/AIPR.2002.1182273","DOIUrl":"https://doi.org/10.1109/AIPR.2002.1182273","url":null,"abstract":"Hyperspectral imaging provides the potential to extract information about objects in a scene that is unavailable to panchromatic imagers. This increased utility, however, comes at the cost of tremendously increased data. To have the broadest range of applications, extraction of the spectral information must occur in real-time. Attempting to produce and exploit complete cubes of hyperspectral imagery at video rates, however, presents unique problems, since data rates are scaled by the number of spectral planes in the cube. MIDIS (multi-band identification and discrimination imaging spectroradiometer) allows both real-time collection and processing of hyperspectral imagery over the range of 0.4 /spl mu/m to 12 /spl mu/m. We present the major design innovations associated with producing high-speed, high-sensitivity hyperspectral imagers operating in the VIS/NIR SWIR/MWIR and LWIR and of the electronics able to handle data rates up to 160 megapixels per second, continuously. Details of two realtime spectral imaging techniques used in MIDIS, dispersive and Fourier transform, are presented. Key to development of MIDIS are high-speed, high sensitivity arrays operating in the stated bands. Real-time algorithms able to exploit the spectral dimension of the imagery are also discussed. Beyond design and performance issues, the paper also discusses applications of real-time hyperspectral imaging technology, including problems such as mine detection, countering CC&D (camouflage, concealment, and deception), and counter terrorism applications.","PeriodicalId":379110,"journal":{"name":"Applied Imagery Pattern Recognition Workshop, 2002. Proceedings.","volume":"48 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2002-10-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"117270748","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}

{"title":"Using hyperspectral reflectance data for discrimination between healthy and diseased plants, and determination of damage-level in diseased plants","authors":"H. Muhammed","doi":"10.1109/AIPR.2002.1182254","DOIUrl":"https://doi.org/10.1109/AIPR.2002.1182254","url":null,"abstract":"It has been found, through many research works, that hyperspectral reflectance data can be used for studying the pathological conditions of crops. The influence of the pathological status of a crop on its spectral characteristics can be visible or detectable in the visible and/or the near-infrared regions of the electromagnetic spectrum, depending on the spectral effects of the pathological conditions of the crop. Differences in the spectral characteristics between normal (i.e. healthy) crops and others suffering from physiological stress or disease, can be revealed and/or magnified by simply normalising the data properly. Such effects can be achieved by normalising the hyperspectral reflectance data into zero-mean and unit variance vectors (i.e. whitening the data). Spectral-wise and/or band-wise normalisation can be performed here. In the experimental part of this work we used a reference data set consisting of hyperspectral reflectance data vectors and the corresponding field measurements of leaf-damage level in the plants. Then, after normalising the new hyperspectral reflectance data; a nearest neighbour classifier is used to classify our new data against the reference data. The correlation coefficient and the sum of squared differences are used as distance measures (between two vectors) in the nearest neighbour classifier. High correlation is obtained between the classification results and the corresponding field leaf-damage measurements, confirming the usefulness and efficiency of this method for this type of analysis.","PeriodicalId":379110,"journal":{"name":"Applied Imagery Pattern Recognition Workshop, 2002. Proceedings.","volume":"4 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2002-10-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121756792","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}

{"title":"A non-invasive spectral reflectance method for mapping blood oxygen saturation in wounds","authors":"L. Martínez","doi":"10.1109/AIPR.2002.1182263","DOIUrl":"https://doi.org/10.1109/AIPR.2002.1182263","url":null,"abstract":"Reflectance spectroscopy is a common technique used to analyze blood oxygen saturation in biological tissues. This study describes a non-invasive method for mapping blood oxygen saturation in wounds using reflectance measurements obtained from a hyperspectral imaging system. The maps illustrate changes in blood oxygen saturation with high spatial resolution. This allows observation into local blood oxygen supplies of tissue as opposed to common point measurement techniques such as pulse oximetry and optical fiber probing. To demonstrate this, an algorithm designed to calculate the blood oxygen saturation from narrow band reflectance images at 760 and 800 nm was used. The algorithm incorporates the contributions of absorption by oxy-(HbO/sub 2/) and deoxy-hemoglobin (Hb) at the oxygen (O/sub 2/) sensitive and O/sub 2/ insensitive wavelength. Results of the algorithms performance are presented to demonstrate that reflectance measurements can be used to determine the amount of blood O/sub 2/ saturation of a wound.","PeriodicalId":379110,"journal":{"name":"Applied Imagery Pattern Recognition Workshop, 2002. Proceedings.","volume":"7 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2002-10-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132266187","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}

{"title":"Spectral imaging and biomedicine: new devices, new approaches","authors":"R. Levenson, P. J. Cronin, N. Harvey","doi":"10.1109/AIPR.2002.1182262","DOIUrl":"https://doi.org/10.1109/AIPR.2002.1182262","url":null,"abstract":"The advent of molecular medicine and new demands on pathologists to deliver prognostic and therapy-shaping analyses has created a need for enhanced imaging tools. Spectral imaging coupled with microscopy is a relatively novel and largely unexplored technology that holds out promise of satisfying, at least in part, such a need. New optical methods for spectral discrimination are being combined with powerful software approaches, often originally developed in different fields, to explore and exploit a wealth of information beyond the capabilities of conventional color-based imaging approaches. Some of the new devices and software tools are described and illustrated here. While the results are indeed promising, it must be stressed that this field is in its infancy, and the optimal uses of this technology in the clinical arena still await definition.","PeriodicalId":379110,"journal":{"name":"Applied Imagery Pattern Recognition Workshop, 2002. Proceedings.","volume":"159 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2002-10-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130076688","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}

{"title":"ICA mixture model based unsupervised classification of hyperspectral imagery","authors":"C. A. Shah, M. Arora, S. Robila, P. Varshney","doi":"10.1109/AIPR.2002.1182251","DOIUrl":"https://doi.org/10.1109/AIPR.2002.1182251","url":null,"abstract":"Conventional remote sensing classification techniques that model the data in each class with a multivariate Gaussian distribution are inefficient, as this assumption is generally not valid in practice. We present a novel, independent component analysis (ICA) based approach for unsupervised classification of hyperspectral imagery. ICA, employed for a mixture model, estimates the data density in each class and models class distributions with nonGaussian structure, formulating the ICA mixture model (ICAMM). We apply the ICAMM for unsupervised classification of a test image from the AVIRIS sensor. Four feature extraction techniques namely principal component analysis, segmented principal component analysis, orthogonal subspace projection and projection pursuit have been considered as preprocessing steps for reducing the data dimensionality. The results demonstrate that the ICAMM significantly outperforms the K-means algorithm for land cover classification of hyperspectral imagery implemented on reduced data sets. Moreover, datasets extracted using segmented principal component analysis produce the highest classification accuracy.","PeriodicalId":379110,"journal":{"name":"Applied Imagery Pattern Recognition Workshop, 2002. Proceedings.","volume":"14 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2002-10-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132153772","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}

{"title":"Real-time embedded hyperspectral image compression for tactical military platforms","authors":"D. Lorts","doi":"10.1109/AIPR.2002.1182267","DOIUrl":"https://doi.org/10.1109/AIPR.2002.1182267","url":null,"abstract":"Summary form only given. This paper presents the current on-going research efforts in which a real-time hyperspectral data compression system developed and demonstrated for a military customer is being ported to an embedded platform fit for deployment onto a tactical platform such as an unmanned aerial vehicle (UAV). The original system consists of a PC host containing multiple PCI boards with SHARC processors interfaced to a state-of-the-art hyperspectral image (HSI) sensor. The resulting embedded implementation will leverage a scalable multiprocessing architecture. Processing nodes based on PowerPC processors with AltiVec technology provide the compute power, while the scalable standard RACEway fabric (ANSI/VITA 5-1994) handles the large interprocessor communication bandwidth. The motivation for this effort is derived from the increased interest in fielding hyperspectral sensors in the intelligence, surveillance, and reconnaissance missions of the military. Historically, there has been significant work performed to develop various data link systems. Data transmission requirements have grown quickly to whatever capacity was available in the data link. With hyperspectral data, this problem becomes even more significant. Sensors such as the EO/IR packages generate large two-dimensional (2-D) data sets. There are many standards developed to compress 2-D data sets, including the ubiquitous JPEG family of routines. With hyperspectral data, there is now a third dimension contained in the collection, that being the spectral components associated with each spatial pixel element. No longer do 2-D approaches apply efficiently. The \"data cube\" produced by an HSI sensor has correlation components in spatial, temporal, and spectral dimension. The principle component transformation algorithm is one such routine that can work within the data cube environment. The results of this port to a deployable, embedded system architecture will be a scalable product that can be integrated into a larger system that may provide actual data exploitation either on the unmanned platform or on the ground element. Performance characteristics between the two implementations are compared. An attempt to \"generalize\" the parallelism to increase the scalability to any number of available processing elements is a critical objective to increase the utility of this approach. The final product of this work will be the creation of a commercial off-the-shelf (COTS) subsystem that can be leveraged by system developers.","PeriodicalId":379110,"journal":{"name":"Applied Imagery Pattern Recognition Workshop, 2002. Proceedings.","volume":"47 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2002-10-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121049738","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}

{"title":"The PNNL quantitative IR database for infrared remote sensing and hyperspectral imaging","authors":"S. Sharpe, R. Sams, T. Johnson","doi":"10.1109/AIPR.2002.1182253","DOIUrl":"https://doi.org/10.1109/AIPR.2002.1182253","url":null,"abstract":"Pacific Northwest National Laboratory (PNNL) is presently compiling a quantitative, high spectral resolution (0.10 cm/sup -1/) set of infrared reference data that are specifically designed for atmospheric monitoring, remote sensing, and hyperspectral imaging. The final list of target compounds will contain nearly 500 gas-phase species, whereby each species is reported as a composite reference spectrum at 25/spl deg/C, with most species also having reference data for 5/spl deg/C and 50/spl deg/C. Each composite spectrum is a quantitative Beer's law fit to typically 10 or more individual infrared measurements, whereby each individual spectrum corresponds to a different partial pressure of the sample that has been pressurized to 760 torr with N/sub 2/ so as to emulate atmospheric pressure broadening. Details of the data acquisition protocol and spectral analysis are discussed.","PeriodicalId":379110,"journal":{"name":"Applied Imagery Pattern Recognition Workshop, 2002. Proceedings.","volume":"24 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2002-10-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116197902","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}

{"title":"Fingerprint enhancement by spectral analysis techniques","authors":"Teddy Ko","doi":"10.1109/AIPR.2002.1182266","DOIUrl":"https://doi.org/10.1109/AIPR.2002.1182266","url":null,"abstract":"This paper presents techniques from spectral analysis that can be used for the enhancement and restoration of deteriorated latent fingerprints, which do not have adequate quality for their positive identification. The use of fingerprints as evidence of crime has been one of the most important utilities in forensics since the late 19/sup th/ century. Following the acquisition of a latent fingerprint, there are two operations to be performed: latent fingerprint enhancement and matching. To facilitate more accurate extraction of prominent features in the fingerprint, enhancement has to be performed in order to eliminate the noise and variety of background patterns. Matching is performed by comparing the enhanced latent fingerprint with one of the fingerprints already in a database. The latent fingerprints are often blurred, incomplete, degraded, and their spatial definition is not clear. These features make their classification and comparison very difficult if not impossible. For the solution of this problem, we analyzed some spatial non-linear filters and frequency domain filters using adaptive fast Fourier transform. As a result of the analysis we found that applying selective filters to the Fourier spectra, the resulting image showed prominent features in fingerprints that could not be extracted by other enhancement techniques.","PeriodicalId":379110,"journal":{"name":"Applied Imagery Pattern Recognition Workshop, 2002. Proceedings.","volume":"25 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2002-10-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116744810","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}

{"title":"Spectral reflectance technique for retinal blood oxygen evaluation in humans","authors":"J. Beach","doi":"10.1109/AIPR.2002.1182264","DOIUrl":"https://doi.org/10.1109/AIPR.2002.1182264","url":null,"abstract":"Retinal diseases such as diabetic retinopathy, glaucoma, vein occlusion and optic nerve atrophy are associated with abnormal oxygen tension and bloodflow in surface vessels of the retina. Visible reflected light spectroscopy has in the past been applied to hemoglobin oxygen saturation measurements from the body surface, as with measurements from skin and wounds, from intact brain vessels during surgical intervention, and as well, from the retinal vessels and other ocular structures inside the body. All of these measurements are complicated by lack of visible light penetration, interference from secondary chromophores, specular return of light and wavelength-dependent penetration to different layers in different structures. We present spectral reflectance curves obtained with a prism-grating-prism (PGP) spectrographic camera from structures producing hemoglobin signatures, including the retinal artery and vein, the pigmented retina and the optic disk, as well as from the macular area which is free of this signature. Oxygen-dependent changes in the hemoglobin signature are determined from vessels and tissue surround.","PeriodicalId":379110,"journal":{"name":"Applied Imagery Pattern Recognition Workshop, 2002. Proceedings.","volume":"14 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2002-10-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124923936","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}