In this paper, we propose a new scheme for data embedding in color video. Traditionally for data embedding in color images, the blue component is modified based on embedded data. This method fails when the picture has very low density of pixel with strong blue component. We devise an adaptive embedding scheme in which the data are embedded in blue or red component of some pixels in each video frames based on their density. This scheme requires no knowledge of the original video for the recovery of the embedded data. We experiment the proposed scheme for embedding an average of 30 bits per frame in the spatial domain of a color host video with QCIF format. Simulation results show that this data embedding scheme low visible distortions in the host video.

No abstract