A face image age recognition method based on transfer learning

Abstract

The invention discloses a face image age recognition method based on transfer learning. The method comprises the following steps: 1) adopting a preprocessing technology for improving the brightness balance of a picture; 2) using a deep convolutional neural network DCNN to realize image feature extraction, and training the DCNN by adopting a transfer learning method; 3) using a softmax classifier;using the softmax classifier for mapping a plurality of scalar parameter values output by the DCNN into a probability distribution array; wherein each probability is the possibility of corresponding classification tags. In the deep convolutional neural network model,an Adam optimizer is selected and used to solve parameters; a DCNN-based classifier is established by obtaining parameters through face image data set training and an Adam optimization objective function, and a classification label corresponding to a maximum component in a probability distribution array output by the softmax classifier is taken as a prediction result of the classifier in the prediction stage. According to the invention, the face image age identification accuracy is obviously improved.

Description

technical field [0001] The invention relates to a face image age recognition method, in particular to a face image age recognition method based on transfer learning. Background technique [0002] With the rapid development of computer vision, pattern recognition and biometric technology, computer-based face age estimation has attracted more and more attention in recent years. It has a wide range of computer vision application prospects, including security detection, forensics, human-computer interaction (HCI), electronic customer information management (ECRM), etc. In real life, the use of surveillance cameras and age recognition systems can effectively prevent vending machines from selling cigarettes and illegal drugs to minors. In social security, fraud and illegal activities at ATMs usually occur in a specific age group, so it can be confirmed and prevented in advance by introducing age information. In the field of biometrics, facial age estimation, as an important supp...

AI Technical Summary

Problems solved by technology

[0003] Although there are relevant studies on face age estimation at home and abroad, the estimation accuracy is not high due to the differences in individual age generation, the complexity of texture information, lack of data, and interference factors.

Images