The linear characteristics of the ship wakes in SAR image is analyzed. Due to the fact that there are many factors which can generate the linear features, so aiming at the problem of the uncertainty in detecting the ship wake peaks in the Radon transform domain, an algorithm based on the peak morphological information to detect the ship wakes in SAR ocean imagery is proposed. The algorithm extracts all possible peaks by thresholding the Radon coefficients, and perform the continuous wavelet transform to the peak match for all these 1-dimensional local peak sections, and makes decision in the feature space finally using the decision vectors formed by the extracted parameters. The simulation and real data processing results show that the algorithm is reliable and can efficiently improve the accuracy of detection.