Algorithm of extracting moving object silhouette based on frame fusion under complex background
复杂背景下基于图像融合的运动目标轮廓提取算法

Effective extraction of the moving object silhouette is essential for the subsequent processing, including target identification, tracking, behavior comprehension, and so on, In some cases, especially when there is low contrast in gray between the background and moving object, the recovered-object is often made up of holes and distorted parts. To improve the performance of extraction in such situation, a modified method based on the principle of frames subtraction was presented. Firstly pre-filtering was needed to alleviate the Gauss noise. Secondly, a maximum variance ratio threshold value was used to remove the remaining noise and background. Then some frames were fused to obtain more information about the moving object, and an area for reference was defined at the same time. After scanning the original image, the moving object silhouette was recovered. Experiment results prove that the modified method is more robust, and superior to other traditional ones.