摘要
针对传统运动目标检测方法存在的缺点和不足,在对现有目标检测算法进行分析对比的基础上,设计并实现了一种简单有效的目标检测方案。首先提出了一种基于像素灰度归类和单高斯模型的背景重构算法,进而以此为基础采用背景差分法进行目标的检测,同时采用分层背景更新算法较好地解决了"拖影"和光照大面积变化的情况,最后给出了一种解决阴影的简单算法。实验结果表明,该算法高效、快速,可以满足实时检测的需要。
In the view of the weakness and shortage of traditional moving targets detection method,This paper designed and realized a simple and effective object detection scheme based on review of the existed detection algorithms. At first,proposed a pixel intensity classification and the single Gaussian model based background reconstruction algorithm,Then used the background difference method for object detection.At the same time the background updating algorithm using hierarchical better solution the"smear"and the large-scale changes in illumination conditions,Finally,the paper proposed a simple algorithm to solve the problem of shadow.The experiment results show that the algorithm is fast and effective,which Can satisfy the requests of real-time detection of moving targets.
出处
《电脑开发与应用》
2010年第5期24-26,共3页
Computer Development & Applications
关键词
背景模型
背景差分
目标检测
背景重构
background modeling
background difference
object detection
background reconstruction