Image quality model generation method and system, image quality model evaluation method and system, equipment and medium

Abstract

The invention provides an image quality model generation method and system, an image quality model evaluation method and system, equipment and a medium, and the image quality model generation method comprises: constructing a training image sample set with N types of image features, wherein N is a positive integer; inputting training image samples in the training image sample set into a to-be-trained convolutional neural network model, wherein the to-be-trained convolutional neural network model comprises at least two deep neural network basic models; performing feature extraction on the training image sample by using the to-be-trained convolutional neural network model to obtain training image feature information, and calculating a loss value between the training image feature information and the standard image feature information by using a loss function; and adjusting parameters in a to-be-trained convolutional neural network model according to the loss value, and generating an image quality model comprising the parameters. The image quality model provided by the invention is used for assisting a test engineer in evaluating the imaging quality of the terminal camera, and the test efficiency and the evaluation accuracy are improved.

Description

technical field [0001] The invention belongs to the technical field of image testing, and in particular relates to a method and system for generating and evaluating an image quality model and its equipment and media. Background technique [0002] With the fierce competition in the smart phone market, the camera mode of the smart phone largely determines the key to the success of the competition among manufacturers. [0003] Now people take pictures in daily life and travel, and the number of photos stored in personal albums is increasing explosively, which determines that image test engineers must measure visual perception when adjusting the camera mode of smartphones, but even It is also difficult for professional photographers to explain which features have a greater impact on image aesthetics. Although aesthetics is difficult to describe, image test engineers still evaluate images from basic factors such as auto exposure, white balance, and auto focus. [0004] The curre...

AI Technical Summary

Problems solved by technology

[0003] Now people take pictures in daily life and travel, and the number of photos stored in personal albums is increasing explosively, which determines that image test engineers must measure visual perception when adjusting the camera mode of smartphones, but even It is also difficult for professional photographers to explain which features have a greater impact on image aesthetics. Although aesthetics is difficult to describe, image test engineers still evaluate images from basic factors such as automatic exposure, white balance, and automatic focus.

[0004] The current image quality evaluation method mainly relies on image test engineers to evaluate the quality of camera image quality based on personal subjective preferences, which is inefficient, highly subjective, and insufficient in reliability of evaluation results.

Images