Big Data Identity Recognition Based on Improved Residual Neural Networks

Zi-ang LI

Abstract


Recently, facial recognition technology has recently been widely developed and applied everywhere. Deep learning is one of the most exciting machine learning method which has many definitions. In these learning methods, convolutional neural networks have fewer connections and parameters than standard feedforward neural networks. As the convolutional neural network deepens, there is a phenomenon that the accuracy of the training set decreases. The residual neural network is proposed, which allows the network to be deepened as much as possible. In this paper, we propose a face recognition method based on a novel residual neural network, which randomly discarding certain parameters of some layers in Residual neural networks and increasing the width of the residual block during training. The method was trained and tested on a standard face classification recognition database and had good results on the closed subset.

Keywords


Big data, Identity recognition, Residual neural networks, Deep learning, Machine learning


DOI
10.12783/dteees/icepe2019/28970

Full Text:

PDF

Refbacks

  • There are currently no refbacks.