A DWT-based Digital Video Watermarking Scheme with Error Cor(7)

Abstract. In this paper, a digital video watermarking algorithm is proposed. We present a novel DWT-based blind digital video watermarking scheme with scrambled watermark and error correcting code. Our scheme embeds different parts of a single watermark in

A test video is split into video stream and audio stream and watermarks are extracted separately by audio watermark extraction and video watermark extraction. Then the extracted watermark undergoes refining process.

Video Watermark Detection. The video stream is processed to get the video watermark. In this step, scene changes are detected from the tested video. Also, each video frame is transformed to the wavelet domain with 4 levels. Then the watermark is extracted with the following condition:

if WC[i] > median(WC[i], WC[i+1], WC[i+2], WC[i+3], WC[i+4])

W[j] = 1

else

W[j] = 0 (5)

where WC[i] is the ith DWT coefficient of a watermarked video frame, and W[j] is the jth pixel of an extracted watermark [13].

As an identical watermark is used for all frames within a scene, multiple copies of each part of the watermark may be obtained. The watermark is recovered by averaging the watermarks extracted from different frames. This reduces the effect if the attack is carried out at some designated frames. Then we can combine the 8 bit-planes and recover the 64×64 size image, i.e., 1/2n part of the original watermark. If enough scenes are found and all parts of the watermark are collected, the original large watermark image can be reconstructed. This can be shown in Fig. 8, where the original frame, the watermarked frame, and the extracted watermark are depicted. Moreover, if some of the watermark part is lost, the final watermark can still

survive. We will show this later.

(a) (b) (c) (d)

Fig. 8. (a) Original frame (b) Watermarked frame (c) Extracted watermark corresponding to Fig. 3(g) (d) Recovered watermark

Audio Watermark Detection and Refining. At the same time, error correcting codes are extracted from the audio stream and the video watermark extracted is refined by this information with the following equation

(6)

where k is the kth block of the average image, (i, j) is coordinate of the video watermark, and P: Q is a ratio of importance of extracted video watermark to audio watermark.