RBF networks

Partial Face Recognition using Radial Basis Function Networks

KIMINORI SATO My HomePage is here.

Abstract

This paper describes a face recognition system that uses partial face images (eye, nose, and ear images) for input data. The recognition technique is based on using Radial Basis Function (RBF) networks. As compared with using a standard back-propagation (BP) learning algorithm, the RBF networks prove to be superior for the face recognition task. From the experimental results of face recognition by partial face image data on a database of over 100 persons, we have achieved a recognition accuracy of 100% for the recognition of registered persons and a rejection rate of 100% for the rejection of unknown samples.

Keywords
 partial face images, Radial Basis Function (RBF) networks, recognition rate, rejection rate

Summary

  1. Face Recogniton Methodology using Radial Basis Function (RBF) networks
  2. Three layers: input layer, hidden layer, and output layer
  3. Use MATLAB
  4. Image acquisition system and how to capture face images
  5. Partial face images (eye and ear images) for input data
  6. Extraction Method using Sobel operater from this image and Extraction by human
  7. Mosaic preprocessing for input unit of a neural network
  8. We have a database of 130 person face image data.
  9. A recognition accuracy of 100% for the recognition of registered persons
  10. A rejection rate of 100% for the rejection of unknown samples
  11. Results are here.

My HomePage is here.