运动图像识别技术在谷物害虫检测中的应用

Application of Moving Image Detection Technology in the Grain Pest Classification

描述了采用差分图像法实现对谷物害虫检测判断和三帧差分法实现谷物害虫图像恢复与提取的方法。该方法在连续3帧视频图像中,分别在1、2帧之间、2、3帧之间作差分运算,根据未变化区域与运动变化区域服从不同的统计规律,对差分图像作运动变化检测。该算法适于实现对仓储物害虫的快速鉴定和分类,实验结果令人满意。

A method of using difference images to detect grain pests was described and three-image difference was adopted to restore and extract the pest images. In this method, the difference operations among three sequential frames were conducted firstly between the former frames and the latter ones in turn. According to the different distribution between the moving image and the unmov- ing part, the difference images were used for moving detection. This method was adaptive to be used in the real-time video surveillance system. The validity of the method is confirmed by implementing the quick identification and classification of grain pests .