An Adaptive and Fast CFAR Algorithm Based on Multithreading for Ship Detection in SAR Image

Abstract

In order to improve the speed of the ship detection, an adaptive and fast constant false alarm rate (CFAR) algorithm based on multithreading is proposed in this paper for ship detection in synthetic aperture radar(SAR) images. At first, the global threshold is calculated to prescreen the target pixels out rapidly by histogram statistics and CFAR algorithm, and obtain the index matrix. Then, the possible target pixels are detected in the sliding window based on the K distribution by the index matrix, and the multithreading technology is adopted to improve the local detection speed. Finally, the possible ships are screened out by making the four-connected neighborhood area statistics for the target pixels obtained in local detection according to the pixel area of the smallest ship. It is concluded that the accuracy and speed of the algorithm is improved greatly by analyzing the SAR image detection results, and the algorithm is more suitable for application to the ship detection system.