Subspace-Based Face Recognition in Analog VLSI
Gonzalo Carvajal, Waldo Valenzuela, Miguel E. Figueroa · 2007
We describe an analog-VLSI neural network for face recognition based on subspace methods. The system uses a dimensionality-reduction network whose coefficients can be either programmed or learned on-chip to perform PCA, or programmed to perform LDA. A second network with userprogrammed coefficients performs classification with Manhattan distances. The system uses on-chip compensation techniques to reduce the effects of device mismatch. Using the ORL database with 12x12-pixel images, our circuit achieves up to 85 % classification performance (98 % of an equivalent software implementation). 1