Aiming at the defect that BP neural network classification model taking a long time for network training, combine with the condition that wavelet network model’s shortness of direction information depicting, put forward a method for SAR image classification, which uses the energy and phase feature of Brushlet as texture feature of SAR image, in addition, inputs the feature vector which describes energy and phase information to adaptive ridge-wavelet neural network for training and classifying. At last, compare and analyze the classification features through test the experiment results on SAR image show that this method is rapid and accurate and outperforms the traditional methods.