Recognition of Low Quality Facial Image

A Study of Face Recognition with the Low Quality Images

Ying Dai, Yasuaki Nakano and Hidetoshi Miyao
Proceedings of the Third International Conference on Automatics, Robotics and Computer Vision, pp.1442-1446 (1994)

Abstract

Face recognition with low quality images has many applications such as security systems, criminal identifications and teleconferences. In this article, we tried to have a test and make some analysis based on the human perception and recognition processing. The experiment procedure about how to obtain the bound of the facial images with low quality for recognition was explained. The experimental evidences for the determination of the resolution bound of the facial images which can be identified was given on the basis of subjective tests. Machine recognition based on the pattern matching approach was also executed. The experimental results suggest that the bound for the correct recognition recognition by pattern matching method based on the similarity computation coincides with that of perceptional tests, and the pattern matching is also effective to the face recognition with low quality images.

[Face Recognition] [Research Themes of Prof. Nakano.]

mail address: <- Please enter the string into the address field

First Written Before June 17, 1998
Transplanted to KSU Before June 19, 2003
Transplanted to So-net April 22, 2007
Last Update April 22, 2007