【導(dǎo)讀】在當(dāng)今的視頻傳輸與廣播網(wǎng)絡(luò)中，版權(quán)的保護(hù)問題已經(jīng)變得越來越緊迫。因?yàn)橐曨l拷貝的出現(xiàn)并沒有降低原始視頻文件的品質(zhì)。一種保護(hù)版權(quán)的方法是在視。頻序列中嵌入一段數(shù)字密碼，這段數(shù)字密碼的學(xué)術(shù)名稱叫做水印。編碼標(biāo)準(zhǔn)決定了。這種算法充分使用了壓縮。列存在的條件下被檢測(cè)出來。從這個(gè)包含于一個(gè)宏塊上的16×16采樣區(qū)域中，我們?nèi)〕鲆粋€(gè)8×8的分塊。宏塊中的4種不同的4×4分塊。對(duì)于水印的嵌入過程，選擇I幀可以歸結(jié)于兩個(gè)主要的理由。為它們的存在對(duì)視頻序列來說是非?；镜?。其次，P幀和B幀被運(yùn)動(dòng)補(bǔ)償高度壓。8×8色度區(qū)域是可以被預(yù)測(cè)的。度預(yù)測(cè)模式適用于幀的平滑區(qū)域選擇的。整數(shù)形式，意味著所有工作可以用避免降低解碼精確度的。這意味著kP的提取不僅具有隨機(jī)性，而且具有魯棒性。行為的后果是可以預(yù)見的，盡管這種后果不會(huì)造成視頻幀品質(zhì)的改變。擊者完成這項(xiàng)操作，將會(huì)造成可見性被辨認(rèn)出的后果。另一個(gè)提取ikP的特征系數(shù)是垂直差分系數(shù)，(,)Y??

　　

【正文】 21 1 1 1 1 1( , ) ( ( ( , ) ( , ) ) , ( ( , ) ( , ) ) ,......., ( ( , ) ( , ) ) , ......, ( ( , ) ( , ) ) )M M M Mi i i i iM n M n M N M Ni i i in n N NY Y Y Y YY Y Y Y? ? ? ? ? ? ? ?? ? ? ? ? ? ? ?? ? ? ? ? ? ? ? ? ?? ? ? ? ? ? ? ?? ? ? ? ? ? ? ???? ? ? ? ? ? ? ? ? ?? ? ???? ? ? ? ? ? ? ?? ? ? ? ? ? ? ? Public Key, Pk Determination First of all we extract the qDC coefficients of 44B? blocks of the thi macroblock. Secondly, all those extracted qDC coefficients are supposed to be presented in a key scrambled style. To fulfill this, all the extracted qDC coefficients are presented in a vector form of the DC coefficients, vqiDC . The quantized DC coefficient of the thi MB in the vector form, vqiDC can be made up of twenty four elements. The luma ponent, Y contributes 16 elements to the vqiDC whereas the chroma ponent contributes 4 elements from each of its sub ponent, bC and rC as seen from the Figure below. Let thi represent the position of the bits that make up a public key in an thi macroblock, ikP . The bits of the public key in a macroblock, ()ikPj are obtained from the quantized DC coefficients that are presented in a vector style, vqiDC . Below is how the bits are obtained: 1 , ( ) ( 1 )()0 , ( ) ( 1 )vvq i q iik q i q iif D C j D C jPj if D C j D C j? ???? ? ???? VIDEO WATERMARK EMBEDDING PROCESS In a pressed domain before carrying out the watermarking process first decode the pressed video bitstream. This is to bring nearer the entropy coding level. Once the entropy coding level is brought nearer to the watermark embedding process, then the putation plex bees less and hence more suitable for real time applications. Furthermore, the nearer the watermark embedding process is to the DCT transform operations, the less induced degradation to the pressed video bitstream. It is highly desirable to carry out the watermark embedding process after quantization. This is simply to make sure that there is no erasure of the watermark and also the quantization being a lossy operation. In this thesis therefore we propose to embed the watermark in the reordered quantized AC coefficients. This is because apart from the quantization process being lossy, entropy coding and decoding is a fast procedure, and also watermark embedding and detection can be done in real time situations. We saw that the ikP is actually made up of several bits that total up to 24. All these bits are not utilized. It is only several of those bits that are utilized in the selection of the coefficient iA in the MB for watermark embedding. The selected coefficient iA from the thi MB is modified in order to embed the watermark iWM in the thi MB. The table shows how to modify the selected coefficient iA . If V is the original video sequence, *V is the watermarked video sequence, S is the scaling factor, iWM is the watermark to be embedded in the thi MB, and iA is the modified randomly selected coefficient from the thi MB for watermark embedding, then we obtain the watermarked video sequence from the expression below ? ?* iiV V S W M A? ? ????? （ 8） Let ? ?iiH WM A?? （ 9） and hence the expression (8) bees *V V SH?? It is just one level that is the maximum change made to the quantized coefficient selected for watermarking embedding. Therefore, the quantization step size modification of a block is as large as the unquantized DCT coefficient of that same block. As a result, the degradation induced by watermark embedding process is proportional to the quantization error caused.