Title :
A low cost compact optoelectronics neural networks for face recognition
Author_Institution :
Dept. of Electr. & Syst. Eng., Connecticut Univ., Storrs, CT, USA
fDate :
31 Oct-3 Nov 1994
Abstract :
Summary form only given. In this paper, we describe a nonlinear filter based optoelectronics neural networks associated with a supervised perceptron learning algorithm for real-time face recognition. The system is a two-layer neural networks. The first layer is implemented using a nonlinear joint transform correlator (JTC)34 and the second layer is implemented electronically because of the small number of the hidden layer neurons. The system is trained with a sequence of input facial images and is able to classify an input face in real-time
Keywords :
face recognition; face recognition; hidden layer neurons; input face; input facial images; low cost compact optoelectronics neural networks; nonlinear filter based optoelectronics neural networks; nonlinear joint transform correlator; real-time face recognition; supervised perceptron learning algorithm; trained; two-layer neural networks; Costs; Face recognition; Integrated optics; Lighting; Neural networks; Neurons; Nonlinear optics; Optical filters; Optical sensors; Robustness;
Conference_Titel :
Lasers and Electro-Optics Society Annual Meeting, 1994. LEOS '94 Conference Proceedings. IEEE
Conference_Location :
Boston, MA
Print_ISBN :
0-7803-1470-0
DOI :
10.1109/LEOS.1994.586864
