【作者】 陈曦； 韩金芬； 刘绍从； 尚为； 张少茹；

【Author】 CHEN Xi, HAN Jin-fen, LIU Shao-cong, SHANG Wei, ZHANG Shao-ru(Institute of Automation, Hebei University of Technology, Tianjin 300130, China; Tianjin Entry-exit Insinspection and Quarantine Bureau, Tianjin 300042, China; College of Physics Science and Information Engineering, Hebei Normal University, Shijiazhuang 050016, China.

【机构】 河北工业大学自动化研究所； 天津出入境检验检疫局； 河北师范大学物理科学与信息工程学院；

【摘要】 运动目标的检测一直是应用视觉领域的重要课题．对于烟花爆竹炸点目标飞行检测,运用动态图像序列的运动目标跟踪技术,以及基于卡尔曼滤波理论的时域递归低通滤波算法,得到炸点目标的二值图像,并利用聚类的方法去掉噪声点．实验结果表明,该方法能在低信躁比下有效地提取出序列图像中的炸点飞行目标．更多还原

【Abstract】 Detection of moving objects is a problem in the research field of application vision. It deals with a detection of flying explosive target for cannon cracker. The binaryzation image of a explosive target can be obtained by a time domain recursive low-pass filtering algorithm based on Kalman filtering theory and the technique combination of moving target tracking based on dynamic image sequence. Furthermore,noise is removed by sorting method .The experiental results show that this method can effectively extract the flying explosive target from the sequence images with low SNR.更多还原