Feature Selection for Edge Detection in PolSAR Images

Edge detection is one of the most critical operations for moving from data to information. Finding edges between objects relevant image understanding, classification, segmentation, and change detection, among other applications. The Gambini Algorithm a good choice finding evidence edges. It finds point at which function difference properties maximized. This algorithm very general accepts many types objective functions. We use an built with likelihoods. Imaging active microwave sensors has revolutionary role in remote sensing. technology potential provide high-resolution images regardless Sun’s illumination almost independently atmospheric conditions. Images PolSAR are sensitive target’s dielectric structures several polarization states electromagnetic waves. polarimetric synthetic-aperture radar (PolSAR) imagery challenging because low signal-to-noise ratio format (complex matrices). There known marginal models stemming complex Wishart model full format. Each these renders different likelihood. work generalizes previous studies by incorporating intensities as edge detection. discuss solutions often problem parameter estimation. propose technique rejects estimates thin evidence. Using this idea discarding potentially irrelevant evidence, we fusing pieces channels that only incorporate those likely contribute positively. approach both single- multilook three sensors.