A Method for Measuring the Blurred Degree of Human Face Image

Abstract

The invention discloses a method for measuring the blurring degree of a human face image, which includes the following steps: 10. Obtaining a blurred human face image: performing a Gaussian blur operation on the original human face image P to obtain a blurred human face image P'; 20. Spectrum information Acquisition: Fourier transform the original face image P and the blurred face image P' respectively to obtain the spectral information F of P and the spectral information F' of P'; 30. Energy spectrum and calculation: according to the spectral information F and the spectral The information F' calculates the energy spectrum and H of P and the energy spectrum and H' of P' respectively; 40. Gaussian blur influence factor calculation: calculate the Gaussian blur influence factor X of the face image according to the energy spectrum and H and the energy spectrum and H' , the larger the Gaussian blur influence factor X is, the clearer the image will be, otherwise the blurrier the image will be. The present invention can accurately measure the fuzziness of human face images obtained in changing scenes, so as to select the clearest human face images from captured video streams and improve the accuracy of human face recognition.

Description

technical field [0001] The invention relates to the technical field of image communication, in particular to a method for measuring the blurring degree of a human face image. Background technique [0002] Face recognition technology has gradually penetrated into the public life. In the face recognition system, selection and recognition based on the blur degree of the face image is one of its important features. There are few methods for judging the blur of face images. The existing image blur judgment methods mainly include methods based on edge gradient information and methods based on frequency domain information. [0003] The degree of blurring of the image is related to its edge sharpness. The high edge sharpness of the image indicates that the image is clear, and the low edge sharpness of the image indicates that the image is blurred. Therefore, the blur detection of the image can be carried out by detecting the edge sharpness of the image. . The flow of the measureme...

AI Technical Summary

Problems solved by technology

However, this method mainly relies on the frequency domain information of the image, so the problem encountered is similar to the method based on edge gradient information. When the edge information contained in multiple face images changes greatly, the difference between images of the same definition The high-frequency information changes greatly, and at this time, the image blur detection method based on frequency domain information also loses its effect

Images