论文部分内容阅读
提出用帧间方块编码差值法和邻域综合法提取运动目标．第一种方法是把当前帧方块的二值图像像素编码值与前帧对应的二值图像像素编码值求差值，由差值检测运动目标；第二种方法是检测２×２方块内各像素的帧差值符号位和帧差绝对值，由二者判别是否运动像素．与普通帧差法相比，提高了抗噪度，改善了运动目标的边缘特性，减少了硬件量．检测算法可以以视频速率工作．实验结果表明，所提出的检测方法是有效的．
Proposed to use the inter-frame coding difference method and neighborhood synthesis method to extract the moving target. The first method is to compare the code value of the binary image pixel of the current frame square with the code value of the binary image pixel corresponding to the previous frame to detect the moving target by the difference. The second method is to detect the difference between Pixel frame difference between the symbol bit and the absolute value of the frame difference, from the two to determine whether the motion pixel. Compared with the normal frame difference method, the anti-noise is improved, the edge characteristics of moving objects are improved, and the amount of hardware is reduced. Detection algorithms work at video rates. The experimental results show that the proposed detection method is effective.