基于动态规划和置信度检验的小目标检测

在目标检测、识别与跟踪系统中,弱小目标的检测是需要解决的关键技术之一。提出了一种基于动态规划与轨迹置信度检验的弱小目标检测方法。该方法利用动态规划方法积累目标能量,然后进行门限判决和对候选目标轨迹进行置信度检验,从而剔除假目标,给出目标信息。该算法针对白噪声背景条件下的图像目标检测,解决了低信噪比条件下运动点目标的检测问题,算法结构简洁,易于进行软硬件分割。最后给出其实验结果。

A Detection Algorithm for Small Moving Targets Based on Dynamic Programming and Confidence Test

Low SNR small moving target detection is a key technique in a target detection, recognition and tracking system. In this paper, we put forward a new method of low SNR small moving target detection based on dynamic programming and sliding trajectory confidence test technology. With this method the detection of a dim target in low SNR is performed based on dynamic programming, threshold decision and gliding confidence test. The method features a high detection probability, low false alarm probability, easy implementation for software and hardware segmentation, and excellent real-time performance. The experimental result is given in the paper.