The authors describe a computerized system using pattern-recognition and artificial intelligence techniques for SAR (synthetic aperture radar) imageprocessing and interpretation. Such images are characterized by cons...
详细信息
The authors describe a computerized system using pattern-recognition and artificial intelligence techniques for SAR (synthetic aperture radar) imageprocessing and interpretation. Such images are characterized by considerable speckle noise, which gives rise to serious problems in early processing. To overcome this drawback, the authors propose to consider not only intensity images but also texture images (obtained by the fractal approach), since texture has proved an effective feature for SAR data classification. In particular, a combined intensity-texture image segmentation is proposed. The interpretation module is based on production rules and frame networks, as well as on a heterarchical control structure. Some preliminary results for a Seasat SAR image are presented and discussed.
We are developing a system designed around an IBM PC-AT to perform automatic diagnosis of diseases from images of the retina. The system includes hardware for color image capture and display. We are developing softwar...
详细信息
An algorithm is presented in which binary object pixels are addressed and processed in order of increasing distance from the background. The distances are defined as path lengths. The metric can be chosen to obtain ar...
详细信息
An algorithm is presented in which binary object pixels are addressed and processed in order of increasing distance from the background. The distances are defined as path lengths. The metric can be chosen to obtain arbitrarily good approximations of the Euclidean metric. The algorithm incorporates an efficient propagation method in which extensive use is made of directional information. It is applied to the Hilditch skeleton and proves to be as fast as the best of the other Hilditch skeletonization algorithms available in software.< >
Computer-aided interpretation is described of forest radar images registered with multispectral thematic mapper images by computer-aided image enhancement, classification, and autocorrelation analysis. Synthetic apert...
详细信息
Computer-aided interpretation is described of forest radar images registered with multispectral thematic mapper images by computer-aided image enhancement, classification, and autocorrelation analysis. Synthetic aperture radar images may be used to discriminate between forests of various spatial structures as well as various types (predominant species). The spatial distribution of trees within radar resolution pixels, especially the gaps between trees, influences both the average backscatter from a forest stand and the texture of radar images, depending on the relative scale of these gaps with respect to the resolution, and the angle of incidence of the radar signal.< >
An approach to pattern classification based on relative constraints in a discrete relaxation framework is described. Classical pattern classification techniques partition feature spaces into disjoint decision regions ...
详细信息
An approach to pattern classification based on relative constraints in a discrete relaxation framework is described. Classical pattern classification techniques partition feature spaces into disjoint decision regions where thresholds are absolute, i.e. fixed numerical quantities. The approach defines pattern classes relative to one another and so results in decision boundaries that depend on the data being classified. Such a formulation leads to a classification scheme based on finding unambiguous labelings using a discrete relaxation-labeling algorithm. Classes are defined exclusively in relative terms, using fairly weak constraints. As a result, there are not many locally incompatible hypotheses to eliminate by Waltz filtering. A ranking scheme is developed which orders hypotheses so that unambiguous labelings can be quickly found through depth-first search. When an unambiguous labeling does not exist, classes can be assigned by picking the most compatible hypotheses. Results of work in progress in classifying Landsat multispectral imagery are presented.< >
A set of multispectral image context classification techniques are discussed which are based on a recursive algorithm for optimal estimation of the state of a two-dimensional discrete Markov random field. The three re...
详细信息
A set of multispectral image context classification techniques are discussed which are based on a recursive algorithm for optimal estimation of the state of a two-dimensional discrete Markov random field. The three recursive algorithms are forms of dynamic programming. Because the estimation equations of the recursive algorithm are quite simple, the computation complexity of the approach is low. It is shown that recursive contextual classification can improve classification performance, as compared to noncontextual classification. In addition, this algorithm has the advantage over other techniques in that it handles multispectral data naturally and simultaneously.< >
From the Publisher: Presents the first unified theory of image segmentation, written by the winners of the 1985 patternrecognition Society medal. Until now, imageprocessing algorithms have always been beset by uncer...
ISBN:
(纸本)0471918261
From the Publisher: Presents the first unified theory of image segmentation, written by the winners of the 1985 patternrecognition Society medal. Until now, imageprocessing algorithms have always been beset by uncertainties, no one method proving completely satisfactory. Wilson and Spann tackle the problem of uncertainty head-on. They describe a new class of algorithms (based, in part, on quadtrees) and demonstrate their applications, including grey level and texture segmentation. These algorithms produce excellent results in a wide range of synthetic and natural data. Provides many examples of applications from medicine to remotesensing.
Speckle in images is often modeled as multiplicative noise, but it also affects the spatial frequency content and the contrast of the image. In order to extract the maximum information from images degraded by speckle,...
详细信息
Speckle in images is often modeled as multiplicative noise, but it also affects the spatial frequency content and the contrast of the image. In order to extract the maximum information from images degraded by speckle, all these factors must be taken into account. It is shown in what sense speckle may be defined as multiplicative noise. The fundamental limitations on information content caused by speckle are pointed out, and various techniques for information extraction are discussed and compared. Some speckle-related problems of interest are image classificaton, image restoration or enhancement, patternrecognition, image compression, and speckle removal. Some of the more promising techniques involve the use of locally adaptive nonlinear filters.
暂无评论