In order to remove the effect of sea surface in ship target detection, a new algorithm based on image block chaos feature for moving target detection on the sea is proposed. In the algorithm, the largest Lyapunov exponent of image block is calculated by the use of small date sets. The change of background's chaos feature due to the existence of moving target is analyzed. Then ship target, which submerged by background, can be detected. The image sequence of 400 frames is used to the detection in the experiment, detection rate is 100% and false alarm rate is 5%. Result shows that the proposed algorithm can detect moving target effectively.%为了消除舰船目标检测中海面背景的影响,提出了一种基于图像块混沌特征的目标检测算法.该算法利用小数据量法计算图像块的最大Lyapunov指数,分析运动目标存在时背景信号混沌特征的变化,检测淹没在混沌背景信号中的目标信号.实验采用400帧图像进行检测,检测率为100%,虚警率为5%.结果表明,算法能有效地检测出海面背景下的舰船目标.
展开▼