摘要： 针对树叶飘落、树枝摇动等自然背景的变化对目标检测带来的影响，提出一种结合分形维的高斯混合模型(GMM)目标检测方法。利用差分盒子维求取图像分形维数，通过设定分形维阈值去除自然背景，采用GMM方法进行目标检测。结果证明，该方法比传统的目标检测方法具有更好的检测效果。

关键词: 目标检测, 高斯混合模型, 差分盒子维

Abstract: In this paper, a Gaussians Mixtures Model(GMM) object detection method combining fractal dimension is put forward aiming at effects caused by changes of natural background. Differential Box Counting(DBC) is used to get the image fractal dimension. A fractal dimension threshold is set to eliminate the natural background. The GMM method is utilized to detect the object. Experimental results show that the performance of the proposed method is better than that of traditional methods.

Key words: object detection, Gaussians Mixture Model(GMM), Differential Box Counting(DBC)

中图分类号: 

• TP391.4