The invention discloses an SAR (Synthetic Aperture Radar) image change detection method based on a residual network, and mainly solves the problems of low detection accuracy and poor noise inhibitioneffect of SAR image change detection in an existing method. The method comprises the following specific steps that: (1) constructing a training tag set; (2) constructing a training data set; (3) constructing a residual change detection network; (4) training the residual change detection network; (5) generating a detection data set; (6) obtaining a feature vector set; and (7) outputting a change detection result map. By use of the method, a seven-layer residual change detection network is constructed, SAR image pixel neighbourhood features are extracted, and the method has the advantages that SAR image change detection noise is lowered and SAR image change detection accuracy is improved.