Data Mining of Image Segments Data with Reduced Neurofuzzy System

  • Authors:
  • Deok Hee Nam;Edward Asikele

  • Affiliations:
  • Engineering and Computer Science, Wilberforce University, Wilberforce, OH 45384;Engineering and Computer Science, Wilberforce University, Wilberforce, OH 45384

  • Venue:
  • ICDHM '09 Proceedings of the 2nd International Conference on Digital Human Modeling: Held as Part of HCI International 2009
  • Year:
  • 2009

Quantified Score

Hi-index 0.00

Visualization

Abstract

The target detection from raw images is a primary task in the image processing. Simultaneously, in order to perform the target detection in the image processing, a large number of variables or factors including unnecessary factors may be involved. This paper presents the pattern recognition through the image scaling based upon the characteristics of various images using the reduced dimension from the original characteristic dimension of the images. Using the less number of dimensions comparing to the original characteristic dimensions, the processing procedures can be simplified and able to overcome the restrictions of the systematic problems. To estimate the performance of the system, neurofuzzy systems with multivariate analysis including factor analysis, principal component analysis, and Fuzzy C-means clustering analysis, are applied. Using the proposed algorithm, the analyses of various image data can be compared.