The automatic judgment for SAR change detection

An automatic judgment algorithm based on the autocorrelation coefficient is proposed to solve the problem of targets´ arrival or departure after change detection. By building probability distribution model of the polarization vectors in the two SAR images, estimating the autocorrelation coefficient of the current and neighboring sliding window in the extent around change part respectively. Comparing the coherence property, we can distinguish the arrival or departure of the change. Comparing to current algorithm based on the covariance, the algorithm in this paper exploits the autocorrelation coefficient to indicate the coherence property. This algorithm improves the accuracy and robustness of distinguish the arrival or departure. At last, by simulating, we proved the validity of this algorithm.