Xiaobo Zhang's Homepage

  • 中文版
  • Home
  • Publications
  • Honors & Awards
  • Projects
  • Research Activities
  • Call For Papers
  • Picture Gallery
  • Contact Me
  • Links


Projects

  • 2012.7~Now


1. Study on Heterogeneous Feature Fusion based on Robust Pictorial Representation and Relationship for Multibiometrics (2014.1~2016.12)

Sources of Funding: National Natural Science Foundation of China (NSFC)


2. Intelligent Media Assets Retrieval System (2013.5~2014.12)

Sources of Funding: The 3rd Research Institute of China Electronics Technology Group Corporation (CETC3)


3. Intelligent Surveillance System via Multiple Sensors

Motivation:

            Construct a multiple sensor network for intelligent surveillance.

Content:

            1) Multiple sensor network construction;

            2) Multiple source data fusion;

            3) Video content analysis.

  •  Before 2012


1. Image Quality Evaluation  for Iris Recognition

Motivation:

            Select the best quality images in a video sequence for iris recognition

Content:

            1) Defocus blur and motion blur estimation;

            2) Signal to noise ratio (SNR) esimation;

            3) Saturation and exposure esimation.

2. Iris Segmentation via An Improved Level Set

Motivation:

           Segment iris regions in iris images using level set.

Content:

            Robust curve evoluation via an improved level set.

3. Long Distance Biometric System using Face, Iris, Palmprint

Motivation:

           Use visible light face images, near-infrared (NIR) iris images, and visible light palmprint images for personal identification at a distance.

Content:

           1) Face detection, eye detection and palmprint detection;

           2) Face recognition, iris recognition, and palmprint recognition.       

4. Bin-ocular Biometrics

Motivation:

          Use the double-eye regions in face images for personal identification, including iris biometrics and periocular biometrics.

Content:

          1) Haar and AdaBoost eye detection;

          2) Head tilt estimation based on eye position;

          3) Fusion of periocular biometrics and iris biometrics.

5. Gabor and AdaBoost based Near-Infrared (NIR) Face Recognition

Motivation:

           Use Gabor filter and AdaBoost learning for NIR face recognition

Content:

           1) Gabor filter and AdaBoost learning;

           2) Linear discirinant analysis (LDA) and nearest neighborhood (NN) based classification.

Make a Free Website with Yola.