基于灰度直方图的背景差目标检测算法

提出了一种基于灰度直方图的目标检测方法。背景差法和帧间差法是两种比较常用的目标检测方法。利用图像灰度直方图进行背景建模,然后提取视频序列中的一帧图像与背景模型进行对比分析,设定一个阈值来判断当前帧与背景模型之间的差异性,以此可以快速判断是否有目标出现。实验结果表明,此方法简单有效、计算复杂度低,能够快速地检测出视频序列中的目标。

A Background Subtraction Algorithms of Object Detection Based on Hirstogram

In this paper, a histogram-based object detection method is proposed. The background subtraction d are used in object detection. It analyzed histogram by comparing the images and background model,then set a threshold to determine whether it is object. The experimental results show that this method is simple and practical.