Neural Networks - DCT for Face Identification 1
Luigi Rosa
Neural Networks - DCT for Face Identification. Matlab source code. High information redundancy and correlation in face images result in inefficiencies when such images are used directly for recognition. Discrete cosine transforms are used to reduce image information redundancy because only a subset of the transform coefficients are necessary to preserve the most important facial features such as hair outline, eyes and mouth.
We demonstrate experimentally that when DCT coefficients are fed into a backpropagation neural network for classification, a high recognition rate can be achieved by using a very small proportion of transform coefficients. This makes DCT-based face recognition much faster than other approaches. Matlab Image Processing Toolbox and Matlab Neural Network Toolbox are required.
CNET Networks is not responsible for the content of this Publisher's Description.
8
6
Format wars: The tech that should have won
Did you love Laserdisc? Were you bonkers over Betamax? Travel back in time with Captain Tech and check out the format losers that should have triumphed
Test your broadband speed
Thinking about changing your broadband Internet provider? Run our quick and simple test to find out how your connection measures up
