Evaluation methods of similarity between simulated SAR images and real SAR images
Chunxia Nie, Yingying Kong, HENRY K. LEUNG, Si Yan · 2016
Evaluation of similarity between simulated synthetic aperture radar (SAR) images and real SAR images is crucial to improve SAR simulating algorithms and it still lacks quantitative evaluation of simulated SAR images. Thus, this paper proposes four methods (i.e., PCA similarity, SIFT matching rate, change detection rate and difference-value of target rate field of view) to evaluate the similarity between simulated SAR images and real SAR images. Experimental results indicate that PCA similarity, SIFT matching rate and change detection rate can be applied to evaluate images of river, plain terrain, mountainous terrain and man-made target, while difference-value of target rate field of view is only applicable for images of man-made target.