A change detection algorithm based on object feature for SAR image
In the detection of change objects of SAR images,regard to the images containing high intensity noise,traditional methods will produce a considerable false alarm. This paper presents a new method of change detection based on directional information.Direction measure is proposed to suppress noises and brighten object contours. Then region blocks which contain changing objects are extracted according to the difference map of direction measure.Finally,changed objects can be obtained using the OTSU threshold method for each region block. The experimental results on SAR images demonstrate the efficiency and accuracy of the proposed method.
Change detection direction measure region blocks multi-threshold and object
Xiaohua Zhang Hongfeng Li Licheng Jiao
Institution of Intelligent Information Processing and Key Laboratory of Intelligent Perception and Image Understanding of Ministry of Education,Xidian University,Xian 710071,China
国际会议
2009 2nd Asian-Pacific Conference on Synthetic Aperture Radar(第二届亚太合成孔径雷达会议)
西安
英文
693-696
2009-10-26（万方平台首次上网日期，不代表论文的发表时间）