Wuhan University Journal of Natural Sciences

, Volume 11, Issue 5, pp 1152–1156 | Cite as

A new method of semantic feature extraction for medical images data

  • Xie Conghua
  • Song Yuqing
  • Chang Jinyi
Semantic Web and Intelligent Web


In order to overcome the disadvantages of color, shape and texture-based features definition for medical images, this paper defines a new kind of semantic feature and its extraction algorithm. We firstly use kernel density estimation statistical model to describe the complicated medical image data, secondly, define some typical representative pixels of images as feature and finally, take hill-climbing strategy of Artificial Intelligence to extract those semantic features. Results of a content-based medial image retrieve system show that our semantic features have better distinguishing ability than those color, shape and texture-based features and can improve the ratios of recall and precision of this syste smartly.

Key words

feature extraction kernel density estimation hill-climbing algorithm content-based image retrieve 

CLC number

TP 37 


Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.


  1. [1]
    Zhang Ji, Hsu Wynne, Lee Mongli. Image Mining: Issues, Frameworks and Techniques [C]//Proc MDM/KDD 2001, San Francisco, USA, Aug. 22–29, 2001: 13–20.Google Scholar
  2. [2]
    Peng Hanchuan, Long Fuhui. A Bayesian Learning Algorithm of Discrete Variables for Automatically Mining Irregular Features of Pattern Images [C]//Proc MDM//KDD 2001, San Francisco, USA, Aug. 22–29, 2001: 87–93.Google Scholar
  3. [3]
    Xie Conghua, Song Yuqing, Zhu Yuquan. A Grid-Based Approach to Extracting Irregular Features of Medical Images [J].Chinese Computer Engineering and Applications, 2005,28(41):52–54 (Ch).Google Scholar
  4. [4]
    Rosenblatt M. Remarks on Some Nonparametric Estimates of a Density Function [J].Annals of Mathematical Statistics, 1956,27:832–837.CrossRefMathSciNetGoogle Scholar
  5. [5]
    Parzen E. On Estimation of a Probability Density and Mode [J].Annals of Athematical Statistics, 1962,33(3):1065–1076.CrossRefMathSciNetGoogle Scholar
  6. [6]
    Hinneburg A, Keim D. An Efficient Approach to Clustering in Large Multimedia Databases with Noise [C].Proc of the 4th+Int'l Conf. New York: AAAI Press, 1998:58–65.Google Scholar
  7. [7]
    Li Cunhua, Sun Zhihui. A Mean Approximation Approach to a Class of Grid-Based Clustering Algorithms [J].Chinese Journal of Software, 2003,14(7): 1267–1274.MATHGoogle Scholar
  8. [8]
    Li Cunhua, Sun Zhihui, Song Yuqing. DENCLUE-M: Boosting DENCLUE algorithm by Mean Approximation on Grids [C].Proc 4th Int Conf on Advances in Web-Age Information Management. New York: Springer Verlag, 2003:202–213.Google Scholar
  9. [9]
    Hinneburg A, Keim D A General Approach to Clustering In Large Databases with Noise [J].Journal of Knowledge and Information Systems (KAIS), 2003,5(4):387–415.CrossRefGoogle Scholar
  10. [10]
    Jin Hua.Research and Implementation of CBIR Technology [D]. Zhenjiang: Jiangsu University, 2003 (Ch).Google Scholar
  11. [11]
    Wang Xiaoling, Xie Kanglin. Auto-Expanded Multi Query Examples Technology in Content-Based Image Retrieval [J].Journal of Southeast University, 2005,3(21):287–292.Google Scholar

Copyright information

© Springer 2006

Authors and Affiliations

  1. 1.Department of Computer Science and EngineeringChangshu Institute of TechnologyChangshu, JiangsuChina
  2. 2.School of Computer ScienceJiangsu UniversityZhenjiang, JiangsuChina

Personalised recommendations