A method of PolSAR imagery segmentation based on region statistical characteristic is proposed in this paper to deal with the shortage of recent reasearch in speckle noise and segmentation efficiency. First

watershed using gradient to segment imagery is used to get the initial segmentation. The ROA gradient is introduced because to the differential gradient is not showing a constant false alarm rate which probably provides false edges. Meanwhile

a method of gradient construction based on morphological operations is given to eliminate the local minimum of the gradient

which would induce massive over-segmentation after the watershed transformation and suppresses over-segmentation. After that the maximum likelihood estimation of the region coherence matrix is estimated by the arithmetical mean of pixels in region. Then

an object function effectively weigh dissimilarity between regions is deduced in combination with the Wishart distribution of the coherence matrix and a hypothesis test

Additionally

the RAG is build based on an object function and the final segmentation is retrieved after merging the hierarchical regions. Simulated data

L-band airborne SAR data of oberpfaffenhofen test site

provided by the DLR and X-band PolSAR data acquired in Hainan Lingshui County

are used to validate the method of this paper. The initial segmentation proved that gradient reconstruction would not undermine the original gradient construction and can effectively suppress over-segmentation. The qualitative and quantitative analysis of the final segmentation show that object function can provide a good segmentation result and has a high segmentation efficiency

information maintenance and segmentation precision. These results validate that the proposed method can effectively reduce the speckle noise and improve efficiency

thereby it can provides a more accurate segmentation.