Research on remote sensing images change detection based on improved IR-MAD algorithm
Coastal zones show drastic changes with the rapid development of satellite remote sensing technology in China,conditions are in place to carry out high-frequency coastal zone change detection,in order to improve the capacity of high-frequency monitoring of the coastal zone region and to help in the efficient supervision of coastal zones and the maintenance of coastal zone rights and interests.In this paper,a high-frequency coastal zone change detection method combining image clustering and IR-MAD algorithm is proposed for the domestically produced high resolution images.The method firstly utilizes the water edge line for spatial constraints,then uses MeanShift algorithm for image clustering,IR-MAD algorithm to get the chi-square diagram to extract the change patches.Finally,the pseudo-changes of the waves,shadows,ships and other pseudo-changes are eliminated,and the final change patches are obtained.Comparative experimental results show that this method performs well in completeness,accuracy and precision of coastal zone change map extraction,and can be effectively applied to the high-frequency operationalization of coastal zone remote sensing,which is of great significance to the supervision and management of coastal zones.
万方数据
维普
