摘要
运动目标的检测是计算机视觉和图像编码研究的主要内容,有着广泛的应用领域。减背景法是常用的方法之一,但其受环境光照影响;而帧差法又不能完全提取出目标的所有相关点。针对传统运动目标检测算法的不足,提出基于动态阈值的减背景法与帧差法相结合的运动目标检测算法。实验证明,算法既能完全提取出目标的所有相关点,又适应复杂变化的环境,运动目标检测的精确度和速度都有了明显的提高。
The moving target detection is the main research content of the computer vision and image encoding which has an extensive application prospect. One of the common methods is based on the background subtraction. But it is influenced by the light and environmental change and the inter-frame difference cannot get all the related pixels of the object. In view of the faults of the traditional methods a moving object detecting method is proposed based on the combination of the background subtraction and the inter-frame difference using the dynamic threshold. The calcula- tion results show that the algorithm ha the advantages that it not only gets all the related pixels of the object, but also adapts to the complex environmental change so that the accurateness and rapidness are increased.
出处
《成都信息工程学院学报》
2008年第4期370-373,共4页
Journal of Chengdu University of Information Technology
关键词
减背景法
帧差法
动态阈值
运动目标检测
background subtraction algorithm
inter-frame difference algorithm
dynamic threshold
target detection