基于改进ViBe算法的运动目标检测

针对ViBe算法中鬼影消除速度慢和固定阈值反映每个像素情况不详细的问题,分别提出了改进帧间差分与ViBe结合算法和自适应阈值的ViBe算法.改进帧间差分法,首先将连续六帧图像进行差分,并通过动态阈值进行校正,进而消除空洞现象,再与ViBe结合来判断鬼影像素.自适应阈值的ViBe算法,将像素值的变化作为像素值的设定依据,...

Moving Target Detection Based on Improved ViBe Algorithm

In the ViBe algorithm,the slow speed of ghosting elimination and the fixed threshold reflecting the incomplete situation of each pixel,respectively proposed the improved inter-frame difference and ViBe combined algorithm and the adaptive threshold ViBe algorithm.Improve the inter-frame difference method.First,the six consecutive frames of images are differentiated,and the dynamic threshold is used for correction to eliminate the hole phenomenon,and then combined with ViBe to determine the ghost image element.The ViBe algorithm with adaptive threshold takes the change of pixel value as the basis for setting the pixel value,so that the pixel situation is reflected in detail,thereby improving the accuracy of detection.The experimental results show that the former of the two improved algorithms nearly doubles the speed of removing ghosts,and the latter increases the accuracy of foreground detection by 1.6%on average.