摘要
针对帧差分法易产生空洞以及背景减法不能检测出与背景灰度接近的目标的问题,提出了一种将背景减和帧差法相结合的运动目标检测算法。首先利用连续两帧图像进行背景减法得到两种差分图像,并用最大类间与类内方差比法得到合适的阈值将这两种差分图像二值化,然后将得到的两种二值化图像进行或运算,最后利用图像形态学滤波得到准确的运动目标。实验结果表明,该算法简单、易实现、实时性强。
For the frame difference method is easy to produce the hollow and background subtraction is not detected if background is close to the target problem, a method of detection combining background subtraction and frame difference is presented. The first use of two successive frames of image background subtraction from the two image difference, and the maximum between class and within class variance ratio method to obtain the proper threshold will be the two difference images of the two values, and then obtained the two values of the two image or operation, finally using image morphological filtering to obtain accurate moving target. The experimental results show that, the algorithm is simple, easy to realize and strong real-time performance.
出处
《电子设计工程》
2013年第3期24-26,共3页
Electronic Design Engineering
关键词
运动目标检测
背景差分
帧差法
最大类间与类内方差比法
形态学滤波
moving target detection
background subtraction
frame difference
maximum between class and within class variance ratio method
morphological filtering
