Retrieve seismic damages from remote sensing images by change detection algorithm

This paper addresses the problem of seismic damage based on the remote sensing image change detection. Comparison technique has long been recognized as an accurate model to describe a variety of image characteristics. Here, we develop algorithms, subtraction and ratio based on a disjointed pixel, correlativity and comparability based on neighboring pixels, respectively. By choosing the suitable threshold to segment differences in resulting images, we perform the post-earthquake damage assessment. Finally, the high resolution images acquired pre-and post-earthquake are presented for illustration and performance evaluation, and satisfied result are gained.