USU Old Main mark

Yuxuan Wang

PH.D.
Department of Computer Science
Utah State University

Phone: (435)213-6781
Email: yuxuan.wang@aggiemail.usu.edu
Webpage: http://cvprip.cs.usu.edu/wang

Education

Ph.D. in Computer Science, Utah State University
05/2012
Dissertation:"A Novel and Effective Short Track Speed Skating Tracking System"

M.S. in Computer Science, Harbin Institute of Technology, China
07/2006
Thesis: "Discriminative Training of Hidden Markov Model in Chinese Sign Language Recognition"

B.S. in Computer Science, Harbin Institute of Technology, China
07/2004
Thesis: "Protein Motif Recognition and Generation"

Interests

  • Computer Vision, Pattern Recognition, Image Processing
  • Machine Learning
  • Computer/Mobile Games Developing
  • Computer-Aided Diagnosis (CAD)
  • Bioinformatics
  • Recommendation System

    • Projects and Experiences

    • A short track speed skating tracking system. Based on single panning camera, skaters in short track speed skating competitions are tracked automatically and lots of invaluable information can be generated such as trajectories, velocity analysis, and 2D reconstruction animations. The proposed novel algorithm solve bunch of challenging problems such as skating in groups, wearing the same uniform, difficulty for registration, etc.. Currently, Matlab is employed for research purpose. Latter, C++ will be employed to build the practical system. link

    • A pavement detection system: integrating line scan camera, GPS, encoder, splitter, and other equipments. Novel algorithms are designed to handle complex road situations. Cracks are detected and classified automatically. IDE is VS 2008; OpenCV and the SDK Sapera LT is employed for developing. link

    • An automatic vehicle detection and classification system. Cooperated with Utah Department of Transportation (UDOT), a video based vehicle classification system is developed, aiming at discovering large trucks. IDE is VS 2008 and OpenCV is employed.link

    • Participated in the development of building up a benchmark database for breast ultrasound images; and proposing bunch of novel algrithms to segment breast cancer ultrasound images.

    • A 3D game Engine "HEAT". Focus on developing VOIP system to allow communicating with each other through talkback system. The engine is based on C++ and many 3rd party open sources. link

    • A recommendation system for open education resources. Focus on new algorithm design to improve recommendation results. The web service is based on Ruby on Rails, and JAVA is employed for algorithm designs. link

    • A mobile app developing project called Geobob. GeoBob is a development platform that enables the design and development of location-based applications for a wide variety of uses on mobile devices like iPhone and Android phones. I was the major iPhone developper. As a by-product, the official USU app was released in early 2011 and got thousands of downloads. Another one GeePerS Math, which is a GPS based orientation game, will be released soon. link1 link2

    • A system to analyze the protein-DNA interaction using structure and physicochemical features of proteins, using support vector machine and decision tree, respectively. According to the study, four features were finally selected to be good features which can predict the specificity of amino acid-base contacts.

    Publication


    Journals
    1. Y. X. Wang, H. D. Cheng and J. Shan, "A Novel and Effective Short Track Speed Skating Tracking System ", Computer Vision and Image Understanding, in revision.

    2. J. Shan, H. D. Cheng and Y. X. Wang, "A Novel Segmentation Method for Breast Ultrasound Images Based on Neutrosophic L-Means Clustering", Med Phys. 2012 Sep;39(9):5669-82.

    3. J. Shan, H. D. Cheng and Y. X. Wang, "Completely Automated Segmentation Approach for Breast Ultrasound Images Using Multiple-Domain Features", Ultrasound in Medicine and Biology, Volume 38, Issue 2, Pages 262-275, February 2012.

    Conferences
    1. J. Shan, H. D. Cheng and Y. X. Wang, "A Novel Segmentation Method For Breast Ultrasound Images Based on Neutrosophic L-means Clustering", Int. Conf. on Image Processing (ICIP2012), Orlando, Florida, USA, 2012.

    2. J. Shan, Y. X. Wang and H. D. Cheng, "Completely Automatic Segmentation for Breast Ultrasound Using Multiple-Domain Features", Int. Conf. on Image Processing (ICIP2010), Hong Kong, China, September 26-29, 2010. (Oral presentation, acceptance rate 47%)

    3. Brett E. Shelton, Joel Duffin, Y. X. Wang, Justin Ball: Linking open course wares and open education resources: creating an effective search and recommendation system. Procedia CS 1(2): 2865-2870, 2010.

    4. Y.X. Wang, H. D. Cheng and J. Shan, "Detecting shadows of moving vehicles based on HMM", Proc. of the 19th International Conference on Pattern Recognition (ICPR 2008), Florida, USA, Dec 8-11, 2008. (Acceptance rate 38%)

    5. J. Shan, H. D. Cheng and Y.X. Wang, "A novel automatic seed point selection algorithm for breast ultrasound images", Proc. of the 19th International Conference on Pattern Recognition (ICPR 2008), Florida, USA, Dec 8-11, 2008. (Acceptance rate 38%)

    6. J. Shan, H. D. Cheng and Y.X. Wang, "A completely automatic segmentation method for breast ultrasound images using region growing", 11th Joint Conference on Information Science, Shenzhen, China, Dec 15-20, 2008.

    7. J. Shan, Y.X. Wang and C. Yan, "Toward the recognition code of protein-DNA recognition", Proc. of IEEE 7th International Symposium on BioInformatics and BioEngineering, pp. 1290 ¨C 1293, 2007. (Acceptance rate 13%)

    8. Y.X. Wang, H. D. Cheng and Chris Glazier, "An Improved Approach for Vehicle Detection and Classification", Joint Conference on Information Science, 2007.

    Programming Skills

    • Proficient: Matlab, C/C++, Objective C, OpenCV.
    • Familiar: Java, R, C#, SQL, Ruby on Rails
    • Platform: Windows, Linux, Mac/IOS
    • Special hardware: Line scan camera, Hasp digital rights management, rotary encoder.