论文部分内容阅读
为在低对比度低信噪比的条件下精确快速搜索跟踪空域机动目标,提出一种智能电视搜索跟踪算法。采用金字塔波门方法进行搜索,搜索区域大小逐级缩小,采用自递归OTSU对图像进行分割,再用连通域分析和灰度级门限判定检测目标,统计目标出现的概率,确定最优目标;锁定最优目标进入跟踪状态,可根据目标尺寸自适应的调整波门大小,对目标瞬时遮挡的情况采用综合预测得到目标位置,若不满足目标灰度级门限判定或预测条件,则转为重新搜索,实现无人工干预的智能搜索跟踪功能。实验结果表明,在DM642平台上搜索状态下最大区域的搜索时间小于20ms,在跟踪状态下目标识别时间小于5ms。完全满足现代空域机动目标电视智能跟踪系统速度、精度和稳定性的要求。
In order to search and track maneuvering targets accurately and quickly under the conditions of low contrast and low signal-to-noise ratio, a smart TV search and tracking algorithm is proposed. Using pyramidal wave method to search, the size of the search region is reduced step by step, the image is segmented by recursive OTSU, and the detection target is determined by the connected domain analysis and the gray level threshold, the probability of the target appears, and the optimal target is determined; The optimal target enters the tracking state, the size of the gate can be adaptively adjusted according to the target size, and the target position can be obtained by using the integrated prediction for the instantaneous occlusion of the target. If the target gray level threshold is not satisfied or the prediction condition is satisfied, the search is resumed , To achieve intelligent search and tracking without human intervention. The experimental results show that the search time of the largest area in the search state on the DM642 platform is less than 20 ms and the target recognition time in the tracking state is less than 5 ms. Fully meet the requirements of modern airstrip maneuvering target television intelligent tracking system speed, accuracy and stability.