Quality classification method for portrait photography image

Abstract

The invention discloses a quality classification method for portrait photography images. The method comprises: firstly, randomly selecting a plurality of portrait images used as sample images of the quality grade for each quality grade from a sample image library in which images are classified into different quality grades according to quality classification of portrait photography images; using a factial feature point detection algorithm to acquire feature points of a face, and then extracting the features of the face, including basic features in the face, position relation features, face shadow features, face proportion features, face saliency and expression features; through and combining with the global feature and distinguishing features of the sample images, performing learning training based on SVM on the samples, obtaining a classifier of quality classification; and calling the classifier obtained in the step (6) on the target portrait photography images, to perform quality classification. The quality classification method for portrait photography images deeply digs into face related features, and accuracy of the classification is high.

Description

technical field [0001] The invention relates to the field of image intelligent processing, in particular to a method for classifying the quality of photographic images of people. Background technique [0002] With the development of social network, more and more people conduct social communication through the network. Today, with the diversification of communication methods, images have gradually become one of the main social elements on the Internet. Images are more vivid than text, and can convey the user's current situation and emotions more quickly and accurately. Because online social networking is based on people, and according to the results of online surveys, nearly half of the social images are people-themed, and because of people's online viewing and reading habits, they tend to spend more time on social images with people. Lots of reading time. Therefore, images containing people play a very high role in social network elements. [0003] In terms of image qual...

AI Technical Summary

Problems solved by technology

[0005] Due to the particularity of character images, it is difficult to achieve a high classification accuracy in the quality classification of general image aesthetics

At present, some scholars have carried out relevant research on the aesthetic algorithm of human images, but the mining of relevant features of human faces is still not deep enough to fully reflect the quality of human photographic images.

Images