Camera tamper detection using codebook model for video surveillance

The proposed model, Camera Tamper Detection Model (CTDM), is used for the camera tamper detection in a video surveillance system. In this paper, an adoptive background codebook model is trained for classifying foreground and background pixels from video image sequence. According to the color model, both chroma and brightness of pixels are adopted as feature values and they are able to tackle the problem of light and shadow movement. During a training process of background pixels, CDTM automatically creates clusters in a background codebook model to classify pixels. After training stage, a well adaptive background model is acquired which is a basis for camera tamper detection. A warning interface of camera tamper detection is provided in the surveillance system which can classify tamper type, e.g. displacement tamper or obstruction tamper, in order to ensure the security of video surveillance system. The experimental results show that the accuracy rate of real-time camera tamper detection in the research can reach 100%, either in the indoor or outdoor experimental scenes.