下載電子全文宣告This thesis is authorized to indicate in-campus access at 3 years and off-campus access at 3 years
You can not download at the moment.
Your IP address is 18.104.22.168
The defense date of the thesis is 2016-08-29
The current date is 2019-02-24
This thesis will be accessible at 2019-08-29
URN etd-0828116-214630 Statistics This thesis had been viewed 420 times. Download 0 times. Author Chyong-Tai Chen Author's Email Address No Public. Department Computer Science and Enginerring Year 2015 Semester 2 Degree Master Type of Document Master's Thesis Language zh-TW.Big5 Chinese Page Count 58 Title Anti-tamper Digital Watermarking Mechanism Based on Genetic Algorithm Keyword Genetic Algorithm Image Attacks Digital Wat DCT DCT Digital Wat Image Attacks Genetic Algorithm Abstract With the fast development of the Internet, the transmission bandwidth has greatly increased and it has been more convenient for people to transfer digital images. Therefore, there is news that some information is altered or illegally stolen when they are passed around on the Internet. With the ever developing image processing technology, image tempering is not easy to detect from the appearance by the naked eye. Therefore, it is necessary to use computers to help analyze whether the image has been altered. Digital watermarking technology provides an effective solution.
Semi-fragile watermarking, as one digital watermarking technique, is robust to average image processing and is fragile to hostile attack in order to differentiate between ordinary compressing and deliberate hostile tempering. Because of the wide use of JPEG (Joint Photographic Expert Group) compression, how to ensure the image is not mistaken for a tempered one after multiple JPEG compression processes is one important subject in semi-fragile watermarking study.
This research proposes a DCT based semi-fragile watermarking method and uses genetic algorithm to find the best embedding location for watermark. JPEG is used as the original image, while binary image, used as watermark, is scrambled with Arnold transformation and embedded into the most suitable position according to genetic algorithm. Besides, JPEG compression is set as rational image processing. The result shows, when watermark-embedded image is JPEG compressed twice with QF both set as 50, the watermark can be completely taken out and the image would not be judged as a tempered one; when the image is deliberately tempered, the altered area can be correctly located.
Advisor Committee Shang-Lin Hsieh - advisor
none - co-chair
none - co-chair
Files Date of Defense 2016-07-28 Date of Submission 2016-08-29