A multi-label age estimation method based on convolutional neural network

Abstract

The invention discloses a multi-label age estimation method based on a convolutional neural network, which includes: obtaining an input sample data set; performing face detection and alignment on each input sample, and performing normalization according to the position of the face; The age labels of the samples are processed with multiple labels, so that each sample is mapped to the same number of labels; all normalized images are used as the input of the convolutional neural network, and the multi-label set is used as the output, and the convolutional neural network is trained to obtain the age Estimation model; according to the principle of binary classification output and multi-label processing, combined with the ordered characteristics of age, the calculation of the estimated age of the face is completed. The present invention utilizes the micro-variability and orderliness among age features, uses the idea of ​​multi-label learning to construct a convolutional neural network model, and solves the problems of low age estimation accuracy and poor robustness in existing age estimation methods. question.

Description

technical field [0001] The invention belongs to the technical field of image processing and deep learning, and more specifically relates to a multi-label age estimation method based on a convolutional neural network. Background technique [0002] At present, face age estimation has been widely used in the fields of investigation and monitoring, information management, intelligent human-computer interaction, and social entertainment. However, face age estimation technology is not accurate enough in real application scenarios and is easily affected by expressions, poses, and lighting conditions. [0003] In face age estimation methods, most of them use traditional age estimation algorithms. Traditional age estimation methods are mainly divided into two stages: feature extraction and age estimation. In the feature extraction stage, most of them are explicit feature extraction, and the age features based on manual design are obtained. However, due to the limitations of manual ...

AI Technical Summary

Problems solved by technology

[0004] Aiming at the above deficiencies or improvement needs of the prior art, the present invention provides a multi-label age estimation method based on convolutional neural network. The output convolutional neural network model solves the problems of inaccurate age estimation and poor robustness of existing face age estimation methods

Images