改进高斯混合模型的运动目标检测算法

Moving target detection algorithm based on improved Gaussian mixture model

传统的高斯模型无法检测比较复杂的场景或速度较低的运动目标,因此提出基于改进高斯混合模型的运动目标检测算法。使用多个高斯模型表示运动目标图像内各像素点特征,并基于图像内各像素点与高斯混合模型相匹配则视其为背景点,反之为前景点原理,更新高斯混合模型。通过更新前景模型并计算短时稳定度指标,提高运动目标检测效果,通过确定高斯分布与像素关系,设定新的参数构造背景模型,消除光照突变造成的影响。实验分析结果表明,该方法能够很好地检测与跟踪运动目标,且抗噪性能好、清晰度高,准确率高达98%。

The traditional Gaussian model fails to detect the complex scene or the moving target with low speed,so a moving target detection method based on the improved Gaussian mixture model is proposed.Several Gaussian models are used to represent the features of each pixel in the moving object image.The Gaussian mixture model is updated on the basis of the principle that if each pixel in the image is matched with the Gaussian mixture model,it is considered to be the background point,otherwise,it is the foreground point.The effect of moving target detection is improved by updating the foreground model and calculating the short⁃term stability index.By determining the relationship between Gaussian distribution and pixels,a new parameter is set to construct the background model to eliminate the impact of light mutation.The experimental analysis results show that this method can detect and track the moving target soundly,and has good anti⁃noise performance,high definition and high accuracy of 98%.