Detection And Tracing Of The Motive Target In The Video Sequence

The purpose of this paper is to research the algorithms to extract the moving objects from the video sequence, to detect the shade area and to trace the moving objects.As for the extraction of the moving objects from the static background, two algorithms are introduced in this paper, algorithm based on the difference between the consecutive frames and the algorithm based on the difference from the background. The algorithm to extract the background, based on gray-level statistic, is improved. By analyzing the algorithm based on the single gauss model to extract the background, one improved algorithm based on multi-gauss model in HSV color space is proposed. From the experiment result, the stable background can be extracted from the video sequence accurately. And these two algorithms are adaptable to the change of the scenes.As for the shade detection, the algorithms in RGB and HSV color space are introduced in this paper. By comparing the algorithms in the two color spaces, one improved algorithm to detect the shade in both two color spaces is proposed in this paper. By using this algorithm, the accuracy and the time performance are greatly improved.As for the light change cancellation, algorithm based on the gray-level change statistic is introduced in this paper. In indoor scene, the moving objects can be extracted accurately by using the grade difference image with the lighting model.As for the tracing of the moving objects, two algorithms are introduced in this paper, algorithm using template to match in the image and algorithm based on the gray histogram character match in the image. Based on the second algorithm, one method is proposed in the paper, using the matched matrix to log the states of the moving objects.