An approach of cluster validity on Gabor wavelet based adaptive face recognition

  • Authors:
  • Rezaul Bashar;Pankaj Raj Dawadi;Eun Sung Jung;Phill Kyu Rhee

  • Affiliations:
  • Department of Computer Science & Engineering, Inha University, Yong-Hyun Dong, Incheon, South Korea. E-mail: {bashar,pankaj,eunsung}@im.inha.ac.kr, pkrhee@inha.ac.kr;Department of Computer Science & Engineering, Inha University, Yong-Hyun Dong, Incheon, South Korea. E-mail: {bashar,pankaj,eunsung}@im.inha.ac.kr, pkrhee@inha.ac.kr;Department of Computer Science & Engineering, Inha University, Yong-Hyun Dong, Incheon, South Korea. E-mail: {bashar,pankaj,eunsung}@im.inha.ac.kr, pkrhee@inha.ac.kr;Department of Computer Science & Engineering, Inha University, Yong-Hyun Dong, Incheon, South Korea. E-mail: {bashar,pankaj,eunsung}@im.inha.ac.kr, pkrhee@inha.ac.kr

  • Venue:
  • International Journal of Knowledge-based and Intelligent Engineering Systems - Extended papers selected from KES-2006
  • Year:
  • 2007

Quantified Score

Hi-index 0.00

Visualization

Abstract

Though research on face recognition has been carried out for a decade, it has in trouble with different situations i.e. facial expression, view point, illumination conditions, noise, etc. To solve this problem, we propose to define situation specific actions for face recognition in this paper. The proposed system partitions face images into several image contexts (groups) based on cluster validity, and takes adaptation to individual partitioned groups. As there is no formal way to select whether the clustering algorithm is suitable or not, we propose a new adaptive cluster validity approach in comparison with Dunn's cluster validity measurement. After selecting proper image context, we train using genetic algorithm to individual feature elements generated by Gabor wavelet of a face image to produce weights. In Gabor wavelet based face recognition, we apply weights to individual elements of facial feature, and those weights are trained by Genetic algorithm. These weights are applied for classifying face images during face recognition time. We applied this concept to face recognition field in different situations and we achieved encouraging results in comparison with Dunn's measuring.