A Moving Objects Detection Algorithm in Video Sequence

Background subtraction and frame differencing algorithm are widely used approaches for detecting moving objects from video sequences. Many different methods have been proposed over the recent years and many related researchers can be confused about their benefits and limitations. In order to overcome this problem, this paper provides an improved algorithm based on these two techniques. It uses surendra background algorithm to get background model and uses frame differencing algorithm to update it, then extracts the moving objects from the fusion of images extracted from these two methods. Because the moving objects we used in the video are bees which always fly in high speed and they are very small, the moving target becomes difficult to extract. In order to solve this problem and get more moving information, the circle segmentation dynamic threshold is used in this paper. It is shown through the result that the method can satisfy the precision and accuracy in the video sequences without affecting the requirement of real-time.