H.264/AVC (advanced video coding) compressed domain robust video watermark embedding and extracting methods free from intra-frame error drift

Abstract

The invention discloses H.264 / AVC (advanced video coding) compressed domain robust video watermark embedding and extracting methods free from intra-frame error drift. According to the video watermark embedding method, a residual coefficient matrix, whose number of non-zero residual coefficients is larger than the threshold, in a macro block with the predictive coding mode of Intra_4*4 in a frame I is analyzed, and then a watermark is embedded in a 4*4 sub-block of the macro block with the predictive coding mode of Intra_4*4 in the frame I by modulating the positive and negative correlation between the residual coefficient matrix and corresponding watermark template. The video watermark embedding method has the advantages as follows: the computation complexity is reduced, the video streaming code rate change after watermark embedding keeps stable, the error drift problem is solved, and the robustness of the compressed domain video watermark is further improved.

Description

technical field [0001] The invention relates to a video watermark technology, in particular to a method for embedding and extracting a robust video watermark in H.264 / AVC compression domain without intra-frame error drift. Background technique [0002] With the rapid development of network and multimedia technologies, operations such as video transmission and processing have become easier and easier. How to protect the legitimate rights and interests of video owners has become an urgent problem to be solved. Video watermarking technology is an effective means to solve this problem. The amount of video data is huge, and it is usually stored and transmitted in a compressed form. Traditional video watermarking algorithms need to completely decode the original video to embed and detect watermarks, and are not suitable for compressed videos. There are mainly two types of methods for embedding watermarks in the compressed domain. The first type is the compressed domain watermark...

AI Technical Summary

Problems solved by technology

There are mainly two types of methods for embedding watermarks in the compressed domain. The first type is the compressed domain watermarking method of joint encoding. When compressing a video stream, it is necessary to completely decode the original video and then jointly encode and embed the watermark, which is extremely complex; the second type is a compressed domain watermark method based on the video stream, which avoids the complicated encoding process and has low computational complexity. , but the watermark can only be embedded in the open loop, and the reconstruction value containing the watermark will cause prediction errors in the reconstruction of subsequent blocks, resulting in error drift

According to the processing method of error drift, this kind of method is divided into three types. The first type is to directly embed watermark without considering the error drift problem, which has good real-time performance, but because the error drift problem is not considered, the video quality is difficult to be guaranteed; Type 2 is to compensate for error drift, which increases the computational complexity. The existing methods of this type are poor in robustness and most of them are detectable but unreadable solutions, which are not suitable for some applications that need to extract watermarks; Type 3 It is to cleverly limit or avoid error drift, strong real-time performance and good video quality, but the existing watermarking methods are not robust enough

Images