Image quality evaluation method and device

Abstract

This invention discloses an evaluating method and device of image quality. The method comprises: calculating the loss of the image to be evaluated at the space domain corresponding a reference image, obtaining the loss result of the image to be evaluated at the space domain; calculating the loss of the image to be evaluated at the frequency domain corresponding to the reference image, obtaining the frequency loss result of the image to be evaluated; obtaining the quality score of the image to be evaluated based on the loss result of the image to be evaluated at the space domain and the loss of the image to be evaluated at the frequency domain. The evaluating method and device of this invention synthetically evaluating the space domain loss and the frequency domain loss based on the characteristics of the human vision system improves the consistency of the object evaluation result and the subjective evaluation result and improves the accuracy of the evaluating result.

Description

technical field [0001] The present invention mainly relates to the technical field of image processing, in particular to an image quality evaluation method and an image quality evaluation device. Background technique [0002] Images are obtained by observing the objective world in different forms and means with various observation systems, and can directly or indirectly act on the human eye to produce visual entities. About 75% of the information that humans get from the outside world is obtained from images. With the development of signal processing and computer science and technology, image engineering has also become a subject with rich content and rapid development. An image system includes image acquisition, display, storage, communication, processing and analysis. It is widely used in various fields of the national economy, such as: scientific research, industrial production, medical and health care, education, entertainment, management and communication, and plays a...

AI Technical Summary

Problems solved by technology

[0006] In the prior art, the objective image quality evaluation method is mainly the peak signal-to-noise ratio (PSNR, PeakSignal-to-Noise-Ratio), which is mainly the statistics of the error between the pixels of the image from the perspective of pure mathematics, PSNR The evaluation is usually to evaluate the image quality on the RGB color domain or the brightness component. The evaluation results are not consistent with the subjective evaluation, the evaluation results are inaccurate, and the performance has great limitations.

Images