As its constant false alarm probability and adaptive threshold, Constant False Alarm Rate(CFAR) detection is one of the most widely used algorithms for ship detection in Synthetic Aperture Radar(SAR) images. Though its performance is better in low-resolution SAR images, with the improvement of swath and resolution of SAR, CFAR cannot meet real-time requirements of ship detection. Aiming at ship detection in high resolution wide swath SAR images, a method is proposed for SAR ship detection based on block prescreening. Firstly, the SAR image is divided into blocks followed by a pre-trained SVM (Support Vector Machines) classification. Then blocks containing targets are detected by Signal to Clutter Ratio(SCR) method. Compared to traditional CFAR detection method, experiments based on real data show that the proposed method can achieve better performance with less computation.