Image Communication and Information Processing (ICIP)

1、        Introduction

The research direction of image communication and information processing is one of the main research directions of the School of Electronic, Information, and Electrical Engineering of Shanghai Jiao Tong University. The basic mission of this direction is to foster innovative sources of imaging for China and supervise graduate students in the domain of advanced image communication and information processing technology, covering HDTV wireless transmission, ubiquitous digital media processing, network streaming media, digital film production and content protection. In present, there are 12 full professors and 9 associate professors in the direction of image communication and image processing. The research group in this direction founded the Shanghai key lab of digital media processing and transmission in 2005. There are advanced instruments and update software integrated in a 1200m2 test platform for digital video acquisition and analysis, video coding and quality measurement, High-speed wireless transmission and evaluation. During the past few years, close collaboration with many reputable universities (Gatech, NTU, UIUC, McMaster, etc.) and transnational companies (MSRA, Micronas, TI, Samsung, etc.) has been carried out through contract strategy, strategic joint lab and manpower exchange. It has made key contributions to the development of digital media technology in China. The outstanding research achievement on China DTV standard, named Advanced Digital Television Broadcasting for Terrestrial (ADTB-T) became a major part of the national Digital Television Terrestrial Broadcasting standard.

The research work in image communication and image processing mainly includes the following 5 areas: (1) HDTV prototyping and advanced digital television broadcasting-terrestrial, (2) intelligent video cooperative processing and transmission technique, (3) broadband network DTV pushing and content integration platform for IPTV, (4) biometrics and Bio-informatics, and (5) computer vision and remote sensing image processing. Many high level projects have been obtained in this research direction, such as the college discipline innovation wisdom plan “Intelligent video science and technology”, key projects from National Natural Science Foundation of China, projects of National High Technology Research and Development Program of China, National Natural Science Funds for Distinguished Young Scholar, and Shanghai Oriental Scholar plan, etc. The research fund of this group is about 4.7 million USD in recent 3 years. In this direction, there are now more than 300 graduate students, among which 25% are Ph.D candidates. More than 300 journal papers and international conference papers have been published in recent 3 years, including 78 SCI indexed paper and 120 EI indexed paper. Research results in this direction earned the First Prize of Shanghai Science and Technology Development Award in 2002, the Second Prize of National Science and Technology Development Award in 2003 and 2008, respectively.

 

2. Faculty Members

      In this research direction, there are 12 full professors, 9 associate professors and 6 assistant professors. The resume of some professors are given below.

     Zhang Wenjun is currently a distinguished professor and the vice president of Shanghai Jiao Tong University. He received Ph.D. degree from the Department of Electronics and Information Technology in Shanghai Jiaotong University, China, in 1989. From 1990 to 1993, he worked as a postdoctoral researcher at Philips Communications Industry Co. in Nuremberg, Germany, where he has been actively involved in developing the HD-MAC (former European HDTV) system. In 1995, he was elected as the head of HDTV Technical Executive Experts Group (TEEG) of China. Since then, he has been acting as the Chief Technical Expert for the Chinese government in the field of digital television and multimedia communications. In the past five years, he led TEEG toward successful development of several generations HDTV prototype systems for terrestrial broadcasting. Two formal TEEG schemes have been presented to the government as the candidates of Chinese digital TV terrestrial broadcasting standard. He is a member of Digital Television Standardization Committee (DTVSC) of China. He has published over 50 technical papers in the area of digital television and multimedia communications. Dr. Zhang received 5 patents and has 7 patents pending. His research interests include digital television, image communication, and wireless transmission. He was the general co-chair of IEEE SiPS2007 and is the general chair of IEEE BMSB 2010.

 

       Yang Xiaokang is a professor and the deputy director of the Institute of Image Communication and Information Processing, Department of Electronic Engineering, Shanghai Jiao Tong University, Shanghai, China. From August 2007 to July 2008, he visited the Institute for Computer Science, University of Freiburg, Germany, as an Alexander von Humboldt Research Fellow. From September 2000 to March 2002, he worked as a Research Fellow in Centre for Signal Processing, Nanyang Technological University, Singapore. From April 2002 to October 2004, he was a Research Scientist in the Institute for Infocomm Research (I2R), Singapore. He has published over 130 refereed papers, and has filed 12 patents. His current research interests include visual processing and communication, media analysis and retrieval, and pattern recognition. He actively participates in the International Standards such as MPEG-4, JVT, and MPEG-21. He received the Microsoft Young Professorship Award 2006, the Best Young Investigator Paper Award at IS&T/SPIE International Conference on Video Communication and Image Processing (VCIP2003) and awards from A-STAR and Tan Kah Kee foundations. He is currently a senior member of IEEE, a member of Design and Implementation of Signal Processing Systems (DISPS) Technical Committee of the IEEE Signal Processing Society and a member of Visual Signal Processing and Communications (VSPC) Technical Committee of the IEEE Circuits and Systems Society. He was the special session chair of Perceptual Visual Processing of IEEE ICME2006. He was the local co-chair of ChinaCom2007 and the technical program co-chair of IEEE SiPS2007.

       Liu Yuncai is a chair professor in Shanghai Jiao Tong University, China. He received a Ph.D. degree from the University of Illinois at Urbana-Champaign, in the Department of Electrical and Computer Science Engineering in 1990, and following as an Associate Researcher in the Beckman Institute of Science and Technology from 1990 to 1991. Since 1992, he had been a System Consultant and a Chief Consultant of Research in Sumitomo Electric Industries, Ltd., Japan. In October 2000, he joined the Shanghai Jiao Tong University as a Chair Professor of Changjiang Scholarship of the Ministry of Education and Honor Professor of the Shanghai Jiao Tong University. He has been an IEEE member since 1992, an associate editor of Pattern Recognition, a council member of Chinese Transport Engineering Society and a council member of China Society of Image and Graphics. Professor Liu is engaged in the wide research fields of computer vision and broad areas in Intelligent Transportation Systems. He made original contributions in the researches of 3D motion estimation, 3D human motion analysis and camera calibration. In the area of Intelligent Transportation Systems, he conducted many advanced projects, such as automatic digital map generation, advanced traffic management systems, advanced traffic information system, vehicle positioning, vehicle navigation, and so on. Presently, his research performances focus on video detection and cognition, ITS information fusion and the applications of computer vision in medical surgeries. Professor Liu published 4 books and more than 200 papers.

 

       Yang Jie is now the professor and director of institute of Image processing & Pattern recognition. He received Ph.D. degree in computer science in University of Hamburg, Germany. Prof. Yang is an expert of information technology in Shanghai Science Committee, the editor in journals of infrared and laser engineering, computer simulation, light mechanical electronics. He is also the committeeman of the Image information processing and control emphases lab academic committee of Ministry of Education, Member of Board of Directors, Shanghai System Simulation Association, director of the Shanghai Artificial Information Processing professional committee, vice-director of the Pattern Recognition Committee, Shanghai Image and Graphics Academy. He is vice-secretary-general of simulation machinery sub-association, National Computer User and high-ranking member of IEEE and Institute of National Automation. Prof. Yang is charged with more than 20 nation and ministry scientific research projects in image processing, pattern recognition, data amalgamation, data mining, and artificial intelligence, which includes one national defense 973 project, two national 863 projects, two national nature fund projects, Sino-French and Sino-Korean cooperation scientific project and so on. He has published a monograph in Germany, and issued 137 papers in important learned periodicals and conferences. And he was awarded the Rockwell Automation Outstanding Teacher and more.

 

3. Representative Research Achievements

 

Achievement 1—HDTV Prototyping and Advanced Digital Television Broadcasting-Terrestrial

 The research team developed the first HDTV prototype in 1998 and conducted the first live broadcasting of HDTV in 1999 in China. The new Digital Terrestrial Television Broadcasting (DTTB) technology called Advanced Digital Television Broadcasting-Terrestrial (ADTB-T) has become the key building block of the Chinese DTTB standard (GB20600-2006) which was published in August 2006, which has been used in.HDTV program of Beijing Olympics games in 2008. A series of ADTB-T demodulation ICs called HD2812, HD2910 and HD2815 has been developed. It has been deploying in rural broadcasting, railway mobile TV systems. This contribution has been awarded as the National Science and Technology Award in China in 2003 and 2008.

 ADTB-T features the single’s carrier’s high-capacity, high-speed digital TV terrestrial transmission and single-frequency network, and over 30 patents has been granted. The main novelties include:

1)  Proposed a first-order cycle frame structure with low PAPR, which constitutes to the high-performance receiver algorithm with a high spectral efficiency and solve the problem of small-scale transmitter.

2)  Invented five technical improvements for the channel’s distortion, such as overlap equalization, virtual center equalizer, turbo equalization with block code, leakage tap control and error rotated equalization.

3)  Proposed a multi-state control of the carrier capture and tracking methods, which enhance the reliability of the carrier capture.

4)  Devised a reliable digital AGC adaptive power control technology, which compensate to the signal jitter in different flight posture of helicopter.

 

3 Representative papers published in recent 5 years supporting achievement 1:

[1]   Wenjun. Zhang, Yunfeng. Guan, Weiqiang. Liang, Dazhi. He, Feng. Ju, Jun. Sun, “An introduction of the Chinese DTTB standard and analysis of the PN595 working modes,” IEEE Trans. on Broadcasting, vol. 53, no. 1, pp.8-13, March 2007.

[2]   Wenjun Zhang, Lin Gui, Wenfeng Ma, Bo Liu, Jian Xiong,” The television broadcasting network of Chinese High Speed Railway,” IEEE International Symposium on Broadband Multimedia Systems and Broadcasting, Apr. 2008, pp. 1–4.

[3]   Weiqiang Liang, Jing Chai, Yunfeng Guan, Wenjun Zhang, and Dazhi He, “A robust and adaptive carrier recovery method for Chinese DTTB receiver,” IEEE Trans. on Broadcasting, vol. 54, no.1, pp146-151, March, 2008.

 

Achievement 2—Intelligent video cooperative processing and transmission technique 

 An active-vision based video cooperative processing system is built up to realize optimal allocation of limited computational resources, power supply and communication capability. The research contents involved in this system mainly include reliable extraction of ROI (Region of Interest), robust ROI tracking, event detection and recognition, specializing in incorporating knowledge-based model or high-level feature model into video processing. The major innovations of this system can be summarized as:

1)  Robust ROI scalable coding strategy with leaky prediction, which successfully finds the trade-off between coding gain and error resilience, enhancing the decoding quality of ROI by restraining the cross error propagation between ROI and background.

2)  Scalable 3D wavelet video coding scheme based on human vision model, which proposes a spatio-temporal visual concealment model to guide the motion compensation process, resulting in great improvement of the decoded video quality.

3)  ROI tracking method based on Bayesian sequential analysis, which combines the relative strengths of particle filtering and mean-shift in tracking robustness and efficiency, being capable of tracking objects reliably in complex situations regardless of the difficulties such as fast motion, partial occlusion, and multiple objects.

4)  Event detection and recognition module developed to analyze ROI’s behavior, which provides semantic access or knowledge-based retrieval to video repositories, greatly alleviating the storage and transmission problem of large scale video dataset.

5)  The team won 2 best runs of 11 event detection tasks in NIST TRECVID2008 competition.

 

3 Representative papers published in recent 5 years supporting achievement 2:

[1]   X. K. Yang, W. S.  Lin, Z. K. Lu, X. Lin, S. Rahardja, E. P. Ong and S. S. Yao, “Rate Control for videophone using perceptual sensitivity cues,”IEEE Transactions on Circuits and Systems for Video Technology, Vol.15, No.4, pp.496- 507, April 2005.

[2]   X. K. Yang, W. S. Lin, Z. K. Lu, E. P. Ong and S. S. Yao, “Motion-compensated Residue Pre-processing in Video Coding based on Just-noticeable-distortion Profile,” IEEE Transactions on Circuits and Systems for Video Technology, Vol.15, No.6, pp.742-752, June 2005.

[3]   X. K. Yang, C. Zhu, Z. G. Li, X. Lin and N. Ling, “An Unequal Packet Loss Resilience Scheme for Video over the Internet,” IEEE Transactions on Multimedia, Vol.7, No. 4,  pp. 753-765, August 2005.

 

 

Achievement 3—Broadband Network DTV Pushing and Content Integration Platform for IPTV

 

High-performance Broadband Network (3TNet) is the national major science and technology project lunched in 2001. The aim of 3TNet is to self-develop Tbps transport, Tbps route, Tbps switch. A 3TNet test-bed is built in Shanghai to test next generation multimedia services run on 3TNet. IPTV is one of the most attractive services for 3TNet. We undertook the work to develop a set of head-end devices for IPTV, including DVB-IP media gateway and content integration platform, to push and manage SDTV, HDTV, and satellite DTV programs over 3TNet network. The main contributes of SJTU are:

1)  The broadband network DTV pushing and content integration platform head-end, including 3 sets of DVB-IP media gateway, a set of SSP, and 2 sets of IP-CAS, are made for 3TNet IPTV. 116 DTV programs, including 1 HDTV programs, are pushed into 3TNet for IPTV services. 24 satellite DTV programs, that are encrypted and organized on an independent portal, are run for a restricted area in 3TNet. All head-ends are running on 3TNet test-bed since Oct. 2005.

2)  Based on high-speed parallel buffering packet auto-filling and packet auto-recognize technologies, an IP-TS stream output rate re-shaping is promoted. Auto-resume of the interrupted input TS stream, and auto recognize and transform of 188/204 TS format are implemented in DVB-IP media gateway.

3)  Based on distributed input stream service information filtering and character recognition technologies, multi-channel DVB-SI meta data is converted to XML in real-time.

4)  The head-end devices developed for 3TNet IPTV are running on 3TNet Shanghai test-bed since Oct. 2005, which covered Shanghai Gubei District, The 9th in Pudong, Shanghai Jiao Tong University, Zhejiang University and Nanjing University of Aeronautics & Astronautics. With the pioneering work developed and tested by 3TNet, the State Administration of Radio File & Television brought forward to build NGB (Next Generation Broadcast) network based on the core-technologies of 3TNet.Based on previous works and followed NGB, we will focus on the design and development of the core technologies on interactive TV services.

 

3 Representative papers published in recent 5 years supporting achievement 3:

[1]   Liu YQ, Yu SY, Zhou J, “Adaptive segment-based patching scheme for video streaming delivery system, ” COMPUTER COMMUNICATIONS, 29 (11): 1889-1895, JUL 26 2006.

[2]   Liu YQ, Yu SY, Adaptive unequal loss protection for scalable video streaming over IP networks,  IEEE TRANSACTIONS ON CONSUMER ELECTRONICS, 51 (4): 1277-1282 NOV 2005.

[3]   Qian T. J., Sun J., Li D., Yang X. K. and Wang J., Transform Domain Transcoding from MPEG-2 to H.264 with Interpolation Drift-Error Compensation, IEEE Transactions on Circuits and Systems for Video Technology, Vol.15, No.4, pp.523-534, April 2006.

 

Achievement 4—Biometrics and Bio-informatics

 

Developed a system for face recognition from video camera based on 3D model and manifold learning; a system for iris capture and recognition. Seven projects are supported by Nature Science Foundation and National 863 Plan, China. Ten patents have been issued. Received 2 major awards for research achievement from Ministry of Education, China and Shanghai Municipality.

Research of bioinformatics focus on Protein subcellular location and N-terminal signal peptide prediction, Protein function prediction, Protein 3-D structure and fold prediction, Membrane protein topology modeling, Protein-DNA, protein-protein interactions etc. Fourty papers have been published in international journals and cited over 300 times. The bioinformatics servers we constructed have been visited over 20000 times

 

3 Representative papers published in recent 5 years supporting achievement 4:

[1]   Guitao Cao, Pengfei Shi, Bing Hu, “Ultrasonic Liver Discrimination Using 2-D Phase Congruency,” IEEE Transactions on Biomedical Engineering, 53(10): 2116 – 2119, Oct. 2006.

[2]   Tianhao Zhang, Dacheng Tao, Xuelong Li and Jie Yang, “Patch Alignment for Dimension Reduction,” IEEE Transactions on Knowledge and Data Engineering (TKDE), in press. 2009.

[3]   T. Zhang, X. Li, D. Tao and J. Yang, “Multimodal biometrics using geometry preserving projections, ” Pattern Recognition, 41: 805-813, 2008.

 

 

Achievement 5— Computer Vision and Remote Sensing Image Processing

1)   3D Motion estimation of articulated object
The problem of 3D motion estimation is insightfully explored without using any constraint condition. The 3D motion model, necessary condition and algorithm are presented for the first time. All other researches before are based on specific motion constraint assumption and unable to deal with general cases. Our research proposed a general algorithm for 3D motion estimation.

2)   3D Image reconstruction and processing 
The research focused on the registration and processing of 3D triangular meshes. A novel neural network strategy was proposed for efficient surface registration with unstructured mass data, based on our modified mesh PCA method for aligning the principal axes directions of 3D surfaces.

3)   Kernel-based nonlinear pattern recognition and machine learning 
In the field of kernel-based pattern analysis, we proposed to optimize the quasi-conformal kernel model using the Fisher criterion in the high-dimensional feature space. An efficient algorithm for optimizing the quasi-conformal kernel has been developed, and the optimized kernel was shown to be able to substantially improve the class separability of the input data in the feature space, and hence, the performances of various pattern recognition schemes.

4)   Remote sensing image processing 
In the field of remote sensing image processing, including similarity measures and search for remotely sensed images, nonlinear discriminance for scatter, shadow detection for color remotely sensed images, and search for geospatial objects in remote sensing based on visual selective attention, we have achieved important breakthroughs, and the research results have been published in the international famous journals.

 

3 Representative papers published in recent 5 years supporting achievement 5:

[1]    X. Y. Zhang, Y. C. Liu and T. S. Huang, “Motion Analysis of Articulated Objects from Monocular Images,” IEEE trans. Pattern Anal. Mach. Intell., Vol. 28, No.4, pp. 625-636, 2006.

[2]    Z.J. Zhao, Y.C. Liu and Z.Y. Zhang, “Camera Calibration with 3 Non-collinear Points Under Special Motions,” IEEE Transactions on Image Processing, Vol. 17, No.12, pp. 2393-2402, 2008.

[3]    Q.J. Kong, Z.P. Li, Y.K. Chen and Y.C. Liu, “An Approach to Urban Traffic State Estimation by Fusing Multi-Source Information,” IEEE Trans. Intelligent Transportation Systems, in press, 2008. 

 

[ 2011-09-07 ]