近年來隨著網路技術的進步，各種多媒體資料能夠輕易地在網際網路上傳送，影像隱藏技術(Image Hiding Technique)在此亦有著相當的進展，在網路傳輸上JPEG壓縮影像廣泛的使用率使其成為一個很好的影像隱藏載體。由於一般影像資料量甚大，因此將欲隱藏之秘密影像利用高壓縮率的向量量化法壓縮後減少所需的容量，在藏入影像之後較能降低對掩護影像畫質之影響也較不亦被察覺。
　　先前將秘密影像經向量量化壓縮後藏入於JPEG掩護影像之研究，需修改量化表以達到可逆式地加以藏入，再利用掩護影像作為訓練編碼簿的影像，雙方不需事先建制編碼簿，增加使用上之機動性。量化表乃是放置於JPEG格式檔頭，因此可以輕易的取出觀察，竊聽者較容易判斷出是否為一張可疑之影像，對於安全性需求很高的秘密圖像通訊可能尚有不足。
　　本論文提出一個將向量量化影像藏入於JPEG影像之隱藏技術，利用JPEG掩護影像在壓縮後頻率域之特性將經向量量化壓縮之秘密影像藏入JPEG影像當中，而且掩護影像也可利用作為編碼簿訓練之用。此一不需要更改量化表之方式可使嵌入後所造成的掩護影像之失真降低於類似經過JPEG壓縮之失真，使得秘密圖像的傳遞更不易被察覺。
論文經由實驗顯示，所提出之方法，除了不需建制編碼簿，證明其利用JPEG影像可以作為秘密圖像傳遞的工具，較之於其他方法，能夠保有一定品質達到較不易被察覺之效果。

　　With the progress of network techniques in recent years, various multimedia data can be conveyed on internet easily, and image hiding technique also have great improvement in this respect. JPEG image is a suitable secret image carrier because it’s popular usage on internet. On the other hand, due to the capacity of secret image is high and it could degrade the cover image’s quality seriously after embedded. It’s suitable to use high compression ratio method such like vector quantization (VQ) to compress the secret image.
　　Pervious research about embed VQ secret image into JPEG cover image, it’s done by an invertible embed method with changing the coefficients of JPEG quantization table, so the JPEG cover image could directly use as the VQ codebook training image. Therefore, both side of secret channel no need to build up codebook in advance to increase the maneuverability of the system. But the JPEG quantization table is restored in it’s header and it could take out easily to observe if it’s a suspicious image, It is probably inadequate when high security demanded situation.
　　An image hiding method is proposed in this paper, VQ secret image is embedded into JPEG cover image utilizing the features in frequency domain of JPEG image and no need to modify the JPEG quantization table. The degradation due to embed is like the degradation of higher JPEG compression to reduce the chance to be suspicious. At the same time, cover image can also be use as the codebook training image to keep the maneuverability of the system. Experiment shows this method can maintain good imperceptibility than other methods to transmit secret image.

[1] CCITT Recommendation T.81, “Digital Compression and Coding of Continuous-tone Still Images”, 1992

[2] C.C. Chang, C.L. Chiang and J.Y. Hsiao, “A DCT-domain System for Hiding Fractal Compressed Images.”, Proceedings of the 19th International Conference on Advanced Information Networking and Applications, Vol. 2 Pages 83-86, 2005

[3] C.C. Chang, Y.C. Chou and J.J. Shen, “Improving Image Quality for JPEG Compression” Ninth International Conference on Knowledge-Based Intelligent Information & Engineering Systems, LNAI 3683, Pages 442-448, 2005

[4] C.K Chan, L.M. Cheng, K.C. Leung and S.L. Li, “Image Hiding Based on Block Difference”, 8th International Conference on Control , Automation, Robotics and Vision, Vol. 2, Pages 968-972, 2004

[5] D.C. Wu and W.H. Tsai, “Spatial-domain Image Hiding Using Image Differencing”, IEE Proc.-Vis. Image Signal Processing., Vol. 147, No.1, 2000

[6] G.C. Langelaar , R.L. Lagendijk and J. Biemond, “Watermarking by DCT Coefficient Removal: A Statistical Approach to Optimal Parameter Settings”, Proc. SPIE Vol. 3657 Security and Watermarking of Multimedia Contents, Pages 2-13, 1999

[7] G.K. Wallace, “The JPEG Still Image Compression Standard”, Consumer Electronics, IEEE Transactions, Vol. 38, Issue 1, Pages xviii-xxxiv, 1992

[8] http://www.ijg.org/

[9] H.W. Tseng and C.C. Chang, “Steganography Using JPEG-Compressed Images”, Proceedings of the Fourth International Conference on Computer and Information Technology, Pages 12-17, 2004

[10] H.W. Wong, C. Au and W.C. Wong, “A Data Hiding Technique in JPEG Compressed Domain”, Proc. SPIE Vol. 4314 Security and Watermarking of Multimedia Contents III, Pages 309-320, 2001

[11] H.W. Wong, C. Au and W.C. Wong, “Data Hiding and Watermarking in JPEG Compressed Domain by DC Coefficient Modification”, Proc. SPIE Vol. 3971 Security and Watermarking of Multimedia Contents II, Pages 237-244, 2000

[12] J. Fridrich, M. Goljan and R. Du, “Invertible Authentication Watermark for JPEG Images” IEEE International Conference on Information Technology: Coding and Computing, 2001. Proceedings, Pages 223-227, 2001

[13] M. Iwata, K. Miyake and A. Shiozaki, “Digital Steganography Utilizing Features of JPEG Images” IEICE Trans. Fundamentals, Vol. E87-A, No. 4, 2004

[14] R.M. Gary, “Vector Quantization”, IEEE ASSP Magazine, Apr. 1984, Vol.1, Issue 2, Pages 4-29, 1984

[15] S.F. Lin, S.C Shie and C.C. Chou, “A suitable Image Hiding Scheme for JPEG Images”, 2004 International Conference on Image Processing, Vol. 3, Pages 1565-1568, 2004

[16] T.S. Chen, C.C. Chang and M.S. Hwang, “A Virtual Image Cryptosystem Based upon Vector Quantization”, IEEE Transaction on Image Processing, Vol.7 No. 10 Page 1485-1488, 1998

[17] Y.C. Hu, “Grey-level Image Hiding Scheme based on Vector Quantization”, Electronics Letters, Vol. 39, No. 2, Pages 202-203, 2003

[18] Y. Linde, A. Buzo and R. Gray “An Algorithm for Vector Quantizer Design”, IEEE Transactions on Communications , Vol. COM-28, No. 1, Pages 84-95,1980

[19] Y. Noguchi, H. Kobayashi and H. Kiya, “A Method of Extracting Embedded Binary Data from JPEG Bitstreams using Standard JPEG Decoder”, IEEE International Conference on Image Processing, 2000 Proceedings, Vol. 1, Pages 577-580,2000

[20] 陳同孝,張真誠,黃國峰, ”數位影像處理技術”, 旗標出版社, 92年出版