视频水印技术按照水印与视频的结合形式，分为压缩域水印和原始视频水印。压缩域水印可以很好地利用视频序列在时间上的冗余，但嵌入水印信息量较小；原始视频水印技术没有充分利用视频序列时间冗余度，但嵌入水印的信息较多，耗费时间较多。 H&G算法可以不解码视频而快速嵌入水印信息，但嵌入容量有限。本文改进H&G算法通过分析非关键帧运动矢量，选择关键帧嵌入位置，使得在压缩域嵌入水印时，水印嵌入速度和嵌入量有较大幅的改进。%Video watermarking technology combined form in accordance with the video watermark,the wa-termark into the compressed domain and original video watermark. Compressed domain watermarking a video sequence can be a good use of the redundancy in time,but little information is embedded water-mark;original video watermark underutilized temporal redundancy of video sequences on the basis of the watermark information can be embedded more,time consuming. H & G method can not decode the video quickly embed watermark information,the program in the H&G algorithm is based on the analysis of non-key frame motion vector selection key frame embedding locations,such that when the compressed do-main embedded watermark, the watermark embedding rate and the doping amount of a substantial im-provement.
展开▼
