Face image quality evaluation method based on lightweight regression network

Abstract

The invention relates to the technical field of computer vision, and discloses a face image quality evaluation method based on a lightweight regression network, and the method comprises the steps of collecting a face image data set; performing data preprocessing on the face image data set by using a face detection algorithm; utilizing a feature extraction algorithm to generate a quality score label, training, verifying and testing the deep learning regression network, and generating a face quality evaluation model; and performing quality evaluation on the face ID to be subjected to quality evaluation by utilizing the face quality evaluation model. According to the invention, the cosine similarity and the face confidence coefficient are used for marking the data, errors caused by manual marking are reduced, the marking speed is high, the lightweight deep learning network is used for regression of the quality score of the face image, the regression precision is guaranteed, the reasoningperformance of the face quality evaluation model is improved, the face image can be evaluated more comprehensively, and the recognition accuracy and the system operation efficiency of the face recognition system are improved.

Description

technical field [0001] The invention relates to the technical field of computer vision, in particular to a face image quality assessment method based on a lightweight regression network. Background technique [0002] The face recognition system is an important part of the intelligent video surveillance system. The face recognition system based on the surveillance video has a complex face image collection environment, which is affected by factors such as light, background, movement, and expression. There are many low-quality images in the face image, and the low-quality face image in the face recognition system will greatly reduce the recognition accuracy of the entire face recognition system, so the face quality estimation module will be added to the face recognition system , to estimate the quality of the face image, and select a good-quality face image for later feature comparison and other modules to improve the recognition accuracy of the entire face recognition system. ...

AI Technical Summary

Problems solved by technology

This invention is a global-based learning algorithm that requires a large amount of labeled data. Manual labeling only uses the prior knowledge of the human visual system, and does not consider the operating mechanism of the face recognition system itself, that is, the impact of face detection and feature extraction algorithms on quality evaluation. Influence, and the learning network uses a common convolutional network, the model accuracy is not high, and the running speed is slow

Images