Zhang Xiaohui

Publications (3)0 Total impact

• Conference Proceeding: Detection and classification of bright lesions in color fundus images

  Zhang Xiaohui, A. Chutatape
  [show abstract] [hide abstract]
  ABSTRACT: Bright lesions, including exudates and cotton wool spots, are the main symptoms in diabetic retinopathy. Early detection and classification of such evidence is essential for an effective treatment. A three-stage approach is applied to detect and classify bright lesions. After a local contrast enhancement preprocessing stage, two-step improved fuzzy C-means is applied in Luv color space to segment candidate bright-lesion areas. The results are shown to be effective in dealing with the inhomogeneous illumination of the fundus images while reducing the influence of noise. Finally, a hierarchical support vector machine (SVM) classification structure is successfully applied to classify bright non-lesion areas, exudates and cotton wool spots.
  Image Processing, 2004. ICIP '04. 2004 International Conference on; 11/2004
• Conference Proceeding: Dynamic vector-space model for internet textual information categorization

  Zhang Xiaohui, Wang Huayong, Li Ying, Chang Guiran, Zhao Hong
  [show abstract] [hide abstract]
  ABSTRACT: First Page of the Article
  Systems, Man and Cybernetics, 2002 IEEE International Conference on; 02/2002
• Conference Proceeding: An autonomous system-based distribution system for web search

  Zhang Xiaohui, Wang Huayong, Chang Guiran, Zhao Hong
  [show abstract] [hide abstract]
  ABSTRACT: The tremendous growth in computer networks and storage has fueled the explosive growth of the web. The amount of information accessible from the web has dramatically increased by several orders of magnitude in the last few years, and shows no signs of abating. Developers of web search engines are confronted with the consequent information overload problem. Although many algorithms are widely used to optimize the performance of web search engines, it is still difficult to decrease the updating cycles of the web data for a traditional general-purpose search and indexing system. In this paper, an autonomous system (AS) based solution is presented. One of its main merits is to limit the network load inside an AS. Since AS is the element of the Internet interconnection and a uniformed route strategy is adopted inside an AS, it will be helpful to solve global network overload problem. At the other side, updating cycles are greatly decreased since the distributed system is working in a parallel way at different ASes. Some key algorithms and techniques are put forward and tested for management and synchronization of the whole system. The effectiveness of the solution is verified in laboratory simulation
  Systems, Man, and Cybernetics, 2001 IEEE International Conference on; 02/2001