Face Recognition/Detection by Clustering and Probabilistic Neural Networks
In this paper a face recognition/detection system is presented, composed out of three main blocks: a face extraction block, which identifies and extracts individual faces from the input image; a feature extraction block, which converts each face identified into a suitable set of features; and a PNN classifier, which recognizes the face presented to it as a feature set. Experiments carried out on a large set of images achieved excellent results in terms of speed and precision, compared with typical figures featured by most state-of-the-art systems.
KeywordsFace Recognition Input Image Probabilistic Neural Network Face Database Face Height
Unable to display preview. Download preview PDF.
- 2.Capizzi G., Coco S., and Pappalardo G. “Image Recognition By Wavelet Transform”, Proceedings of SOCO ‘89, pages 492–496. Genova, June 1999.Google Scholar
- 3.Capizzi G., Coco S., and Pappalardo G. “A New Clustering Based System For Automated Object Recognition”, Proceedings of SOCO ‘89, pages 503–507. Genova, June 1999.Google Scholar
- 4.Capizzi G., Coco S., and Pappalardo G. “Face Recognition by Pseudo-Spectral Transform and PNN Classification”, Proceedings of SCS 2001, pages 297–300. IEEE. Iasi, Romania. July 2001.Google Scholar
- 5.Kelly M. D. “Visual identification of people by computer”, Stanford AI Project, Stanford, CA, Tech. Rep. AI-130, 1970.Google Scholar
- 6.Kohonen T. “Correlation Matrix Memories”, IEEE Transactions on Computers, Vol. C21, no. 4, Apr 1972.Google Scholar
- 7.Lawrance S., Giles C.L., Tsoi A.C., and Back A.D. “Face Recognition: A Convolutional Approach”, NEC Res. Inst., Princeton, NJ Tech. Rep. 1995.Google Scholar