A Machine Learning-based Change Detection Scheme for the Remote Sensing Images using Gaussian and Lee filters

This article presents an updated change detection algorithm specifically for remote sensing synthetic aperture radar (SAR) images and optical images. This method uses Lee and Gaussian filtering techniques to overcome challenges such as speckle noise and illumination changes in SAR images. First, apply the Lee filter to the SAR image and use the Gaussian filter to the optical image to remove noise from the image. The Support Vector Machine (SVM) algorithm is then used to identify changes in the two images. Compared with existing methods, the proposed method can provide better PCC values and lower total error (PTE) values. Experiments on different SAR and optical images have confirmed the effectiveness of this method.