skip to: page content | links on this page | site navigation | footer (site information)

Junqing Chen's Homepage

Resume

Junqing Chen

Contact

Unilever Research
45 River Road
Edgewater, NJ 07020
Home: (973 )482-7275
Office: (201) 840-2324
Email: junqing.chen@unilever.com
Webpage: http://commnet.ece.northwestern.edu/~jqchen

Education

1999-2003
Ph.D Electrical Engineering, Northwestern University
Thesis: "Perceptually-based Color and Texture Features for Image Segmentation and Retrieval"
Advisor: Thrasyvoulos N. Pappas

1996-1999
M.S. Electrical Engineering, Zhejiang University, China
Thesis: "Research on Genetic Algorithm Embedded Neural Network and Its Application in Control"
Advisor: Jingping Jiang

1992-1996
B.S. Electrical Engineering, Zhejiang University, China
Project: "Identification-free PSD Control System"
Supervisor: Peiti Feng

Research Experience

Aug. 2004-present
Imaging Scientist, Unilever Research at Edgewater, NJ
Facial photography standard. Skin image analysis. Comsumer perception.

Dec. 2003-July, 2004
Postdoctoral Research Fellow, Northwestern University
Continuing research on perceptual models for image segmentation and retrieval. Also, working on perceptual models for image and video quality evaluation.

1999-2003
Research Assistant, Northwestern University
Doctoral Research: "Perceptually-based Color and Texture Features for Image Segmentation and Retrieval"
 Developed spatially adaptive low-level features for image segmentation and retrieval that are based on perceptual principles about the processing of texture and color information. Proposed an algorithm that combines these features to obtain image segmentations that convey semantic information that can be used for content-based retrieval. The emphasis was on images of natural scenes. Designed and implemented subjective experiments for perceptual tuning of the algorithm. The novelty of the work is in the incorporation of knowledge of human perception and signal characteristics into feature extraction and algorithm design. Two types of features were considered. Spatially adaptive dominant colors were used as color composition features. Such features reflect the fact that the human visual system cannot simultaneously perceive a large number of colors, and also, the fact that image colors are spatially varying. Spatial texture features were based on a multiscale frequency decomposition, which approximates early processing in the human visual system. Local median energy of the subband coefficients was used to boost texture responses within regions while eliminating edge effects. The subjective experiments (available at http://peacock.ece.utk.edu/FeatureTest/) were used to obtain key segmentation parameters. Algorithm performance was demonstrated on large number of images, using objective and subjective performance evaluations.

Project: "Perceptual Metrics and Perceptual Coders"
 Examined perceptual metrics and used them to evaluate the quality of still image coders. Showed that mean-squared-error based metrics (such as PSNR) fail to predict image quality when comparing artifacts generated by different types of image coders. Considered three different types of coders: JPEG, the Safranek-Johnston perceptual subband coder (PIC), and the Said-Pearlman SPIHT algorithm with perceptually weighted subband quantization, based on the Watson et al. visual thresholds. Showed that incorporating perceptual weighting in the SPIHT algorithm results in significant improvement in visual quality. The metrics were used to evaluate the performance of the compression techniques for a wide range of bit rates. The experiments indicated that the PIC metric provides the best correlation with subjective evaluations. Also showed that the relative algorithm performance depends on image content, with the subband and DCT coders performing best for images with a lot of high frequency content, and the wavelet coders performing best for smoother images.

06-08, 2001
Summer Intern, IBM TJ Watson Research Center, Hawthorne, NY
Worked in the Visual Analysis Group on texture analysis of natural images for Content Based Image Retrieval. My mentors were Aleksendra Mojsilovic and Bernice Rogowitz (group manager). The goal was to develop algorithms that provide semantically meaningful image segmentations that can be used for semantic image browsing. Used Wavelet based texture classes and incorporated color to obtain overall image segmentation. This work laid the foundation for my Ph.D. thesis.

1996-1999
Research Assistant, Zhejiang University
 Project: "Genetic Algorithm Embedded Neural Network"
 Worked with the Real-time Control lab to design an optimal real-time controller for a chemical process reactor. The controller was in the structure of a neural network whose weights were first trained off-line with Genetic Algorithm and later adjusted based on real-time system feedback. The design was applied to the pro-type reactor and experiment results indicated that our GA trained neural network controller excelled in terms of performance, robustness and reliability.

Teaching Experience

2000-2002
Teaching Assistant, Northwestern University
For the following courses: Solid State Engineering, Fundamental of Circuits, Digital Communications, Engineering Analysis 3: Dynamic Systems, Probabilistic Systems and Random signals, Engineering Analysis 1: Linear Algebra, Numerical Methods and Signals and Systems. My duties included leading discussion sections, preparing assignments and solutions, assisting in exam preparation, as well as design and supervision of lab experiments.

1999-2000
International Scholars Program Northwestern University
I was paired with a trained undergraduate student (Teaching Partner), Jennifer Chuy, who helped me adjust to teaching in the US. Jennifer and I met twice a week during the whole academic year. Jennifer trained me in effective communication with students, classroom lecturing, most commonly used electrical engineering terms etc. At the end of each quarter, I gave a twenty-minute videotaped lecture session. Jennifer and I then studied my video together. Tremendous has been learned through such processes.

Academic Experience

2001-2002
Prepared NSF Proposal
 Assisted in preparing a grant proposal for research on "Deriving Perceptually-Based Texture and Color Features for Image Segmentation, Categorization, and Retrieval." The project was funded by NSF.

Reviewed articles for

  • IEEE Transaction on Image Processing
  • IEEE Transaction Circuits and Systems for Video Technology
  • Journal of Electronic Imaging
  • IEEE Signal Processing Magazine
  • Image and Vison Computing
  • International Journal of Remote Sensing
  • IEEE International Conference on Image Processing (ICIP)
  • IEEE International Conference on Multimedia & Expo (ICME)
  • IEEE International Conference on Acoustics, Speech, and Signal Processing (ICASSP)
  • International Conference on Pattern Recognition (ICPR)

Computer Skills

  • Programming languages: C & C++, Matlab & Simulink, HTML, JAVA, SHELL Programming
  • Operating Systems: UNIX, Windows98/2000/XP, LINUX, Mac OS X
  • Software: Adobe Photoshop, MySQL, Microsoft Office, Latex

Honors

  • Honorable Mention of Best Ph.D thesis, Northwestern University, 2004
  • P & G Scholarship and Excellent Graduate student, Zhejiang University, 1998.
  • Wang Guosong Scholarship, Zhejiang University, 1997.
  • Scholarships from Zhejiang University for undergraduate student, 1992-1996
  • Admission into Zhejiang Univ. exempted from Entrance Exams, 1992

Professional Memberships

  • IEEE Signal Processing Society
  • SPIE The International Society for Optical Engineering

Publications

Journal Articles
  • J. Chen, T. N. Pappas, A. Mojsilovic, and B. Rogowitz, "Adaptive Perceptual Color-Texture Image Segmentation", to appear, IEEE Trans. on Image Processing
  • T. N. Pappas, J. Chen and D. Depalov, Perceptually Based Color and Texture Features for Semantic Image Segmentation, Classification and Retrieval, accepteded for publication, IEEE Signal Processing Magazine, Special Issue on Semantic Retrieval of Multimedia.
  • J. Chen, T. N. Pappas, A. Mojsilovic, and B. Rogowitz, "Perceptually-tuned Color-Texture Image Segmentation", in preparation
Book Chapter
  • R. J. Safranek, T. N. Pappas and J. Chen, "Perceptual criteria for image quality evaluation", Handbook of Image and Video Processing, Al Bovik, Editor, Academic Press, forthcoming 2004
Conference Proceedings
  • J. Chen, T. N. Pappas, A. Mojsilovic, and B. E. Rogowitz, ``The Experimental determinatino of visual color and texture statistics for image segmentation'', in Human Vision and Electronic Imaging, SPIE conference, San Jose, CA, Jan. 2005 .
  • J. Chen, T. N. Pappas, A. Mojsilovic, and B. Rogowitz, "Perceptual Multiscale Image Segmentation based on Color and Texture features", IEEE International Conference on Image Processing (ICIP'04), Singapore, Oct. 2004
  • (Invited) J. Chen, T. N. Pappas, A. Mojsilovic, and B. Rogowitz, "Perceptual Tuning of Low-level Color and Texture Features for Image Segmentation", Proceedings of the 37th Asilomar Conference on Signals, Systems and Computers, Pacific Grove, CA, Nov. 2003
  • J. Chen, T. N. Pappas, A. Mojsilovic, and B. Rogowitz, "Image Segmentation by Spatially adaptive color and texture features", Proceedings of IEEE International Conference on Image Processing (ICIP'03), Barcelona, Spain, Sept. 2003
  • J. Chen, T. N. Pappas, A. Mojsilovic, and B. Rogowitz, "Perceptual Color and Texture Features for Segmentation", Proceedings of SPIE, in Human Vision and Electronic Imaging, Santa Clara, CA, Jan. 2003
  • J. Chen, T. N. Pappas, A. Mojsilovic, and B. Rogowitz, "Adaptive image segmentation based on color and texture", Proceedings of IEEE International Conference on Image Processing (ICIP'02), Rochester, New York, Sept. 2002
  • J. Chen and T. N. Pappas, "Perceptual Metrics and Perceptual Coders", Proceedings of SPIE, in Human Vision and Electronic Imaging, San Jose, CA, Jan. 2001
  • J. Chen and J.P. Jiang, "The State Estimation of the CSTR System based on a BP Neural Network trained by HGAs and ES", IASTED-SIP'98, Oct 28-31, 1998, #281-124
Conference Presentations
  • "The Experimental determination of visual color and texture statistics for image segmentation", Human Vision and Electronic Imaging session, SPIE, San Jose, CA, Jan. 2005
  • (invited) "Perceptual Tuning of Low-level Color and Texture Feature for Image Segmentation", the 37th Asilomar Conference on Signals, Systems and Computers, Pacific Grove, CA, Nov. 2003
  • "Perceptual Color and Texture Features for Segmentation", Human Vision and Electronic Imaging session, SPIE, Santa Clara, CA, Jan. 2003
  • "Perceptual Metrics and Perceptual Coders", Human Vision and Electronic Imaging session, SPIE, San Jose, CA, Jan. 2001

 

 

 

Contact | ©1999 Junqing Chen | Last updated