Filtering Approach Based on Voter Model and Spatial-Contextual Information to the Binary Change Map in SAR Images

摘要

If a binary change map (BCM) is generated without considering spatial-contextual information, it can produce a number of false and missed alarms because of the significant speckle noise in synthetic aperture radar (SAR) images. Therefore, this paper first proposes the local continuity assumption of the changed and unchanged areas. Then, a new filtering approach to BCM of SAR images is proposed based on the voter model and spatial-contextual information. Experimental results indicate that the proposed approach can effectively decrease the detection noise, improve the detection accuracy of BCM and perform better than a median filter, mathematical morphological filters and a filter based on a Markov random field.