Source IEEE TRANSACTIONS ON CIRCUITS AND SYSTEMS FOR





![§ Upper bound of the minimum Hamming distance [27] J. E. Mac. Donald, “Design § Upper bound of the minimum Hamming distance [27] J. E. Mac. Donald, “Design](https://slidetodoc.com/presentation_image_h/191a8ff38c45636d3e31fa97d5480102/image-6.jpg)
![Two-class SVM classifer Schematic of the data extraction [29] C. -C. Chang and C. Two-class SVM classifer Schematic of the data extraction [29] C. -C. Chang and C.](https://slidetodoc.com/presentation_image_h/191a8ff38c45636d3e31fa97d5480102/image-7.jpg)












- Slides: 19
Source: IEEE TRANSACTIONS ON CIRCUITS AND SYSTEMS FOR VIDEO TECHNOLOGY, 2016, 26(3): 441 -452. Authors: Jiantao Yuan Yan Tang Zhou, Weiwei Sun, Li Dong, Xianming Liu, Oscar C. Au, Speaker :Xiaozhu Xie Date : 2018/04/26 1
§ Related work § Hamming distance § Entropy § Proposed scheme § Experimental results § Conclusions 2
Hamming distance between two strings of equal length is the number of positions at which the corresponding symbols are different. p "karolin" and "kerstin" is 3. p 1011101 and 1001001 is 2. p 2173896 and 2233796 is 3. 3
Shannon defined the entropy Η of a discrete random variable X with possible values {x 1, . . . , xn} and probability mass function P(X) as: 4
n bits Each block size of Image encryption: n bits Data embedding: Schematic of data hiding over encrypted domain [26] H. Lipmaa, P. Rogaway, and D. Wagner. CTR-Mode Encryption. [Online]. Available: http: //csrc. nist. gov/encryption/modes/workshop 1/papers/lipmaa-ctr. pdf, accessed Sep. 2000. 5
§ Upper bound of the minimum Hamming distance [27] J. E. Mac. Donald, “Design methods for maximum minimum-distance error-correcting codes, ” IBM J. Res. Develop. , vol. 4, no. 1, pp. 43– 57, Jan. 1960. 6
Two-class SVM classifer Schematic of the data extraction [29] C. -C. Chang and C. -J. Lin, “LIBSVM: A library for support vector machines, ” ACM Trans. Intell. Syst. Technol. , vol. 2, no. 3, pp. 27– 53, Apr. 2011. 7
Original image 125 76 155 62 90 76 50 62 57 24 96 118 57 24 96 138 78 76 125 62 50 76 60 62 82 97 211 114 125 216 49 62 148 66 89 62 96 24 96 118 57 24 96 160 125 144 56 41 125 44 56 41 33 211 114 82 97 21 14 97 Encrypted image divided into 2 x 2 blocks 97 Embed 010 (n=3) 243 175 155 62 90 76 50 62 250 84 96 118 57 24 96 138 78 76 125 62 50 76 60 62 82 97 211 114 125 216 49 62 148 66 89 62 96 24 96 118 57 24 96 160 125 144 56 41 125 44 56 41 33 211 114 82 97 21 14 97 8
243 175 80 208 103 232 … Marked encrypted block 70 … 250 84 43 50 103 165 164 11011111 … … Two-class SVM classifer 43 70 103 232 9
§ 10
11
100 images of size 512 × 512 with various characteristics, including natural images, synthetic images, and highly textured images. n=3 [18] X. Zhang, “Reversible data hiding in encrypted image, ” IEEE Signal Process. Lett. , vol. 18, no. 4, pp. 255– 258, Apr. 2011. [19] W. Hong, T. -S. Chen, and H. -Y. Wu, “An improved reversible data hiding in encrypted images using side match, ” IEEE Signal Process. Lett. , vol. 19, no. 4, pp. 199– 202, Apr. 2012. 12
Six test images for fine-grained comparison. (a) Lena. (b) Baboon. (c) Man. (d) Lake. (e) Cactus. (f) Texture mosaic 1. 13
Comparison of the extraction accuracy for six representative test images 14
15
Erroneous blocks in two problematic images. 16
17
§ Suggested a public key modulation mechanism. § No need of managing the data embedding key. § Performed extensive experiments. 18
Thank you! 19



















