Downloads: 125 | Views: 134

M.Tech / M.E / PhD Thesis | Electronics & Communication Engineering | India | Volume 6 Issue 9, September 2017

Face Recognition across Age Using Auto Encoder Neural Network

Neha Rahman | Nitin Naiyar

Abstract: In this modern era of digitization and advanced technology, human face has become a demanding icon to authenticate ones identity. One peculiar feature which distinguishes it from other biometrics techniques is that it does not need the test subject to perform its working. Other conditions where face recognition does not work well include poor lighting, sunglasses, hats, scarves, beards, long hair, makeup or other objects partially covering the subjects face, and low resolution images. Facial Aging is such a process that affects both the shape as well as wrinkles on the face. These shapes and wrinkles changes degrade the performance of the automatic face recognition. In this paper, neural network is used for the performance evaluation of research work. Stack Auto-encoder is used for training purpose and serves as one of the input of deep network and confusion matrix is used for calculating the accuracy of the model. The evaluation is performed in the MATLAB environment.

Keywords: Face aging, Stacked Auto-encoder, Deep Networks, Age Invariants, softmax layer, confusion matrix

Edition: Volume 6 Issue 9, September 2017,

Pages: 1450 - 1454

Similar Articles with Keyword 'confusion matrix'