Research on SAR image change detection algorithm based on hybrid genetic FCM and image registration

Abstract

At present, image change detection technology as an important image processing technology has been widely used, and a novel synthetic aperture radar (SAR) image change detection algorithm based on hybrid genetic FCM and image registration is proposed in this paper. First of all, the algorithm performs registration with two photographs taken in the same region images at different time using Harris operator and Sift operator, then, the ratio method and logarithmic method is combined to extract the initial differences image, and then the Principal Component Analysis (PCA) method is used to reduce the dimension of the difference image. Finally the hybrid genetic fuzzy C-means (FCM) algorithm is used to determine the classification of feature vector space, and the classification results are compared with the reference image, to obtain the change information. The purpose of using hybrid genetic FCM is to divide the initial difference image clustering into the changed type and the unchanged type, so as to get the final segmentation result. The FCM algorithm improved by genetic algorithm can effectively avoid that the FCM algorithm will fall into local minimum when the initial cluster center selection is not appropriate. As the genetic algorithm is a global optimization search algorithm, it can improve the segmentation effect of FCM. The experimental results show that the proposed algorithm has the highest global correct rate of 98.10% and 99.74%.