基于改进帧差法的视觉监控实验系统

Visual Monitoring System Based on Improved Frame Difference Method

为了解决监控系统中对移动目标的漏检、误检,以及快速移动目标检测的问题,提出一种改进的帧差法对视频序列识别检测。该帧差法是对连续的3帧图像进行两两差分二值化,然后相邻二值化图像与运算,再把与运算的结果相或。把当前帧的前后两帧的差分二值化图像进行生态学膨胀、腐蚀等一系列处理,处理后的图像与前面或运算的结果作相或运算,这样基本解决误检的问题,再结合边缘提取的Robert算法就可以完整地检测出移动目标。经测试结果表明:在像素值相近时,该检测算法能够较好地解决检测时出现的误检和漏检,并且移动目标速度较快也能够检测出来,检出率高达90%以上,误检率降低到2%左右。

In order to solve the leak and mistake of moving targets in monitoring system,and the fast moving target detection problem,an improved frame differential method is presented for video sequence identification test.The improved frame difference method has three consecutive frames and does difference binarization for each two frames,then the adjacent binary images do“AND”operation,and the results do“AND-OR-NOT”operation.Two binarized frames which are before and after the current frame are expanded and corroded,the processed images do do“AND-OR-NOT”operation with the afore results.Thus the mistakenly identified problem can be solved.Coupled with Robert algorithm of edge detection,the moving targets can be fully detected.The test results show when the pixel values are similar the detection algorithm can better resolve the mistaken and leak detection problem,and can detect moving targets faster.Its detection rate is as high as 90%,error detection rate is down to 2%.