Abstract: A methodology for detecting image anomalies in a target area for classifying objects therein, in which at least two images of the target area are obtained from a sensor representing different polarization components. The methodology can be used to classify and/or discriminate manmade objects from natural objects in a target area, for example. A data cube is constructed from the at least two images with the at least two images being aligned, such as on a pixel-wise basis. A processor computes the global covariance of the data cube and thereafter locates a test window over a portion of the data cube. The local covariance of the contents of the test window is computed and objects are classified within the test window when an image anomaly is detected in the test window. For example, an image anomaly may be determined when a matrix determinant ratio of the local covariance and the global covariance exceeds a probability ratio threshold. The window can then be moved, e.g.

Abstract: A methodology for detecting image anomalies in a target area for classifying objects therein, in which at least two images of the target area are obtained from a sensor representing different polarization components. The methodology can be used to classify and/or discriminate manmade objects from natural objects in a target area, for example. A data cube is constructed from the at least two images with the at least two images being aligned, such as on a pixel-wise basis. A processor computes the global covariance of the data cube and thereafter locates a test window over a portion of the data cube. The local covariance of the contents of the test window is computed and objects are classified within the test window when an image anomaly is detected in the test window. For example, an image anomaly may be determined when a matrix determinant ratio of the local covariance and the global covariance exceeds a probability ratio threshold. The window can then be moved, e.g.

Abstract: A computer-implemented method for tracking a small sample size user-identified object comprising extracting a plurality of blocks of pixels from a first frame of a plurality of frames of a scene detected by a hyperspectral (HS) sensor, comparing a reference sample of the object with the plurality of blocks to generate a first attribute set corresponding to contrasting HS response values of the reference sample and HS response values of each block of the plurality of blocks, comparing a test sample of a portion of the first frame to each block of the plurality of blocks to generate a second attribute set corresponding to contrasting HS response values of the test samples and HS response values of each block of the plurality of blocks and determining if the object exists in two or more of the frames by comparing the first HS attribute set with the second HS attribute set.

Abstract: A computer-implemented method for tracking a small sample size user-identified object comprising extracting a plurality of blocks of pixels from a first frame of a plurality of frames of a scene detected by a hyperspectral (HS) sensor, comparing a reference sample of the object with the plurality of blocks to generate a first attribute set corresponding to contrasting HS response values of the reference sample and HS response values of each block of the plurality of blocks, comparing a test sample of a portion of the first frame to each block of the plurality of blocks to generate a second attribute set corresponding to contrasting HS response values of the test samples and HS response values of each block of the plurality of blocks and determining if the object exists in two or more of the frames by comparing the first HS attribute set with the second HS attribute set.

Abstract: A method of detecting an image anomaly (target) within a scene represented by image data comprises obtaining test data from a test window within the scene, combining the test data with reference data to generate combined data, then comparing the combined data with either the test data or the reference data. An improved image analyzer includes a computational unit configured to execute this method. In one example, image data is generated by a hyperspectral imager.