Venue: the Conference Room of Optical Image Analysis and Learning Center (OPTIMAL), 3rd Floor, Building 3
Abstract:
Recent advances in multimedia and communication technologies have paved ways for people around the world to acquire, utilize and share multimedia contents. As multimedia communications become ubiquitous in our life, information security and content right management become increasingly important. A traditional approach to protect the multimedia content is to encrypt the whole bit stream with a conventional cryptographic algorithm. However, due to large volume of the multimedia data, this approach requires a considerable amount of computational power, and is impractical in resources-deprived scenarios (e.g., hand-held mobile devices). In addition, multimedia files typically have well-defined hierarchical structure, which is not recognizable in the ciphertext, and hence, is wasted.
In this talk, we give a survey of the recent progress in multimedia encryption with increasing emphasis on integrating encryption and compression for improved system efficiency and operability. By performing compression and encryption jointly in one unified step, one can greatly simplify the system design, reduce energy consumption, and facilitate advanced multimedia manipulations. In order to assess the security level of this class of methods, we investigate and evaluate the joint compression-encryption methods using Huffman coding, Exp-Golomb coding, arithmetic coding, and Lempel-Ziv-Welch coding. We show that most of the existing techniques are vulnerable against various attacks. Therefore, solely relying on the randomness offered by the compressors cannot achieve high level of security. Nevertheless, joint compression-encryption is still attractive for applications that require middle-level security and relatively short-term copyright protection (e.g., the first few months of newly released movies).
Speaker Profile:
Xiaolin Wu (born in Chengdu, China) graduated from Wuhan University in 1982, majoring in computerscience. In 1988 he received his Ph.D degree in computer science from University of Calgary, Alberta,Canada. From 1988 to present, he has been a faculty member of University of Lethbridge, Canada (assistantprofessor, 1988-1989), University of Western Ontario, Canada (associate professor 1989-1995, tenuredprofessor in 1995-2000), New York Polytechnic University, USA (professor 2000-2002), McMaster University,Canada (professor, senior research chair, 2002-present). He was also a visiting professor at ChineseUniversity of Hong Kong (1995-1996), Technical University of Denmark (2000), Microsoft Research, USA(1999). He won many international awards, including 1998 UWO distinguished research chair, Canada, 2000Monsteds fellowship, Denmark, 2004 Nokia fellowship, Finland, 2008 VCIP best paper award, and he iscurrently NSERC senior industrial research chair, Canada. Xiaolin Wu is an associated editor of IEEE Transactionson Image Processing, and was an associated editor of IEEE Transactions on Multimedia. He served on the technicalcommittees and was session chairs of many IEEE international conferences, including Data Compression,Multimedia, Information Theory, and Image Processing
Xiaolin Wu’s research interests include data compression (image coding in particular), joint source-channelcoding, multimedia signal processing, joint compression-encryption, steganalysis, quantization, high-fidelityimaging and superresolution. He has published over 200 research papers in these research fields, including60 papers in IEEE Transactions on Image Processing, Multimedia, Information Theory, Signal Processing,Communications, Pattern Analysis and Machine Intelligence, Medical Imaging. Xiaolin Wu has directedmultiple government and industrial research projects of funding level over $3,500,000, in the areas of digitalcinema, telemedicine, signal steganalysis, joint source-channel coding, and remote sensing.
Optical Image Analysis and Learning Center
Tel: 029 - 8888 9302
E-mail: OPTIMAL@opt.ac.cn
