Key clip extraction from Surveillance Videos

The increasing demand of the cameras for surveillance systems not only requires the large storage devices but also requires the reduction in the time to browse the whole video. To meet such requirements video summarization is the only solution. A good visual summarization should collect as much as possible visual information from input video and also maintain the chronology between them. We propose a video summarization algorithm based on pyramidal reduction for surveillance videos. Surveillances video summarization is not only related to reduction of size of the video but also to track the important objects in the video by maintaining the chronology. We have implemented two methods for summarizing a given video sequence by pyramidal reduction using motion entropy (H) and ii) using mutual information (I). Here we take input from a fixed surveillance camera. The method uses the construction of a simple summary by discarding the unwanted segments or low level segments in the input video in a pyramidal way. The discarding criteria are applied based on the values of entropy (H) and the selection of key segments is done based on the mutual information (I).