Research On Video Information Hiding Based On Quantization Modulation

Information hiding, also known as data hiding, is an important subdiscipline in the field of Information Security. With the development of network technology, multimedia industry and electronic commerce, information hiding technology finds its hotbed in modern digital world. Featured by hiding information without affecting its carrier, information hiding technology is warmly embraced by information security applications such as digital right management. Deemed as a paradigm of this technology, digital watermarking is drawing increasingly intense attention due to its contribution to the digital multimedia industry.Among diverse blind watermarking technologies, quantization modulation distinguishes itself by its desirable performance in watermark capacity, noise endurance, and fidelity maintenance. Accompanying these advantages is its apparent vulnerability to gain attack, with a traditional fixed quantization step. To solve this problem, some scholars study the relation between quantization step and signal gain, and break a new path for this technology’s development.Out of the current request of network video copyright protection and content management, this thesis investigates basic properties and extensive applications of information hiding technology, and presents two video watermarking schemes with quantization modulation, based on existing research results. Utilizing multiple signal processing technologies, these two schemes are typically aimed at two kinds of applications. One is the compression domain watermarking scheme for real-time video stream labeling such as user labeling, and the other deals with original data for robust applications like video classification. Both of them adopt adaptive quantization step to get rid of gain attack, improve the video’s visual quality, and hence better meet the performance requirements by different video service applications.