Gender Classification Using Artificial Neural Networks through Independent Components
In this paper, an efficient technique for gender classification is developed. It uses the information maximization approach of independent component analysis for extracting the features from the face images. Further, these features were tested using back propagation neural network (BPNN) and radial basis function neural network (RBFNN). The analysis were carried out on FERET database. The main objective of the paper is to build up an optimum classifier using neural networks. The performance of the classifier is estimated through confusion matrix and measured in terms of accuracy, sensitivity and specificity.
KeywordsICA infomax BPNN RBFNN Accuracy Sensitivity Specificity
Unable to display preview. Download preview PDF.
- 3.Haykin, S.: Neural Networks: A Comprehensive Foundation. IEEE (1999)Google Scholar
- 5.Jain, A., Huang, J.: Integrating independent components and support vector machines for gender classification, vol. 3, pp. 558–561 (2004)Google Scholar
- 7.Phillips, P.J., Moon, H., Rizvi, S.A., Rauss, P.J.: The feret evaluation methodology for face recognition algorithms. (a) Example 1 (b) Example 1 (c) Right ear (d) Mirroed Left earGoogle Scholar
- 9.Vipsita, S., Shee, B., Rath, S.: An efficient technique for protein classification using feature extraction by artificial neural networks. In: 2010 Annual IEEE India Conference (INDICON), pp. 1–5 (December 2010) Google Scholar