基于变参数混合高斯模型的动态目标检测算法

A Dynamic Target Detection Algorithm Based on Variable Parameter Gaussian Mixture Model

针对机器视觉系统中运动目标自动检测的实时性和准确性问题,提出了一种改进的目标检测算法。该算法在混合高斯模型适宜动态环境的基础上,提出了变参数模型,并对Otsu分割算法进行了改进,提高了智能系统检测目标计算速度,而且有效地消除了这类算法固有的目标虚警问题。试验结果表明:在复杂背景下,该方法有良好的适应能力和鲁棒性,适于变化和运动的实际场景。

An improved target detection algorithm for the real-time and accuracy of dynamic target automatic detection in machine vision systems is proposed in this paper.Based on Gaussian mixture model suitable for dynamic environments,variable parameter model is proposed and Otsu segmentation algorithm is improved,which not only improves the speed of detecting targets in the intelligent systems,but also solves inherent problems efficiently in false alarm of targets.The test results show:under complex background,this algorithm has an excellent adaptability and robustness to fit the complicated situation and changing environment.