Face Recognition Application of Artificial Neural Network

Abstract

A neural net is a collection of nodes which collectively perform a particular kind of computation. Each node is a small computational module and the nodes can operate in parallel. The operation of the network depends on the interactions between the nodes and the adaptation of their connective .strengths Artificial Neural Network as an engineering tool can be used for face recognition by training the network several times on noisy images and other various input ideas. The images contain faces and it is a common problem in the real world to deal with noise. This network is trained to output 1 in the actual position of the output vector, with this, its expected to fill the rest of the output vector with 0’s. However, irrespective of the noise generated by the input images there is a tendency that may result in the network not creating perfect 1’s and 0’s. The output will be passed through the competitive transfer function after the network has been trained. This function makes sure that the output corresponding to the face most like the noisy input image takes on a value of 1 and all others have a value of 0.

Cite as

Imaekhai Lawrence. (2021). Face Recognition Application of Artificial Neural Network. Research and Reviews on Experimental and Applied Mechanics, 4(2), 1–7. https://doi.org/10.5281/zenodo.5155053