论文部分内容阅读
对低信噪比运动背景下序列图像中弱小目标的提取算法进行了研究,提出一种新的目标提取算法。根据运动背景的相关性和目标运动的连续性,采用背景模板匹配技术,计算各帧背景运动量,去除复杂背景,用自适应预估和阈值分割技术捕获跟踪目标。实验中,结合所研究的算法,对100帧包含低对比度弱小目标的运动图像序列进行了处理。实验结果表明,此方法能有效地实现目标的自动捕获与跟踪,跟踪效率高,跟踪稳定。
The algorithm of weak and small target extraction in sequence images with low SNR is put forward and a new target extraction algorithm is proposed. According to the correlation of motion background and the continuity of target motion, the background template matching technique is used to calculate the motion background of each frame, the complex background is removed, and the tracking target is captured by adaptive prediction and threshold segmentation. Experiments, combined with the study of the algorithm, on 100 frames containing low-contrast weak small moving target sequence was processed. Experimental results show that this method can effectively achieve the goal of automatic capture and tracking, tracking, high efficiency, stable tracking.