Face image quality assessment method, computer device and computer-readable storage medium

Abstract

The present invention provides a face image quality assessment method, a computer device, and a computer-readable storage medium. The method includes acquiring an initial image, dividing the initial image into multiple sub-images, and calculating the average gray value of each sub-image. and the variance, and perform normalized calculations on each sub-image; perform discrete cosine transform on each normalized sub-image and extract low-frequency coefficients, and use the gray value and low-frequency coefficients of each sub-image to calculate the sub-image Weighted features; calculate a similarity probability score for each sub-image with respect to a standard face sub-image, and calculate the final score for this initial image. The present invention also provides a computer device and a computer-readable storage medium for realizing the above method for evaluating the quality of a face image. The method of the present invention takes into account the influence of various conditions such as illumination, posture, and occlusion on the face image, and can more accurately evaluate the quality of each image.

Description

technical field [0001] The present invention relates to the technical field of image recognition, specifically, a face image quality assessment method, a computer device and a computer-readable storage medium for realizing the method. Background technique [0002] Since each person's face shape is discriminative, non-contact and easy to obtain, it is regarded as an ideal biometric feature for personal identity authentication. In the past two decades, face recognition has attracted widespread attention from academia and business circles, and many face recognition methods have emerged. With the development of intelligent technology, the identification of people by means of image recognition has been widely used in many fields such as security, among which, the recognition of people's faces is a key research field in the field of image recognition. [0003] In the process of face recognition by taking images, due to the existence of various factors such as illumination, postur...

AI Technical Summary

Problems solved by technology

Therefore, the existing technology proposes to use convolutional neural network for face quality assessment. However, the method based on deep learning requires a large amount of labeled data for training, and usually suffers from many constraints such as computing resources, power consumption, and model volume when deployed. factor constraints

[0007] It can be seen that most of the current face image quality assessment methods only consider some basic characteristics of face images, and do not consider the performance impact on subsequent face comparison and verification.

Images