Abstract
Today, we often use the video data in user personal equipment in multimedia communication services. Therefore, we consider the matching method of using the temporal characteristics in video searching system. In this paper, we pay attention to the luminance variances and the correlation value of the luminance variances between current frame and next frame. In addition, we propose matching algorithm the grouping correlation value by threshold and matching time can be short. Finally, we consider the structure of group and matching results in query video which have scene changes and not have ones.
