Method for evaluating quality-lose referrence image quality base on Contourlet transformation

Abstract

The invention discloses a degraded reference image quality evaluation method based on Contourlet transformation. The method utilizes Contourlet transformation to catch the direction information of an image, effectively extracts the average value and the standard deviation of a transform coefficient matrix of an image Contourlet transformation domain sub-band reflecting the texture structure information of a reference image and a tested image as statistical attribute values of the image; by comparing the similarity of the texture structure information between the reference image and the tested image, the quality value of the tested image is finally obtained; image quality evaluation carried out by utilizing the similarity of the texture structure information between images does not require the reference image of very good visual quality, and the result of image quality evaluation purely reflects the similarity between the reference image and the tested image, namely, the result of evaluation can objectively reflects the influences of image processing or compression algorithm on the change of image quality.

Description

technical field [0001] The invention relates to an image quality evaluation technology, in particular to a quality-degraded reference image quality evaluation method based on Contourlet transformation. Background technique [0002] With the rapid development of image processing and compression technology, how to more effectively evaluate the visual quality of an image, and then compare the pros and cons of various processing or compression algorithms has become a research hotspot in the field of image processing and compression. Image quality evaluation algorithms mainly include subjective and objective quality evaluation algorithms. Subjective quality evaluation algorithms require the participation of many evaluators and follow complex procedures and steps, which are time-consuming and expensive, and the evaluation results are easily affected by evaluators and test conditions. And the impact of the test environment, poor stability, poor portability and other shortcomings, s...

AI Technical Summary

Problems solved by technology

Since the no-reference model does not need reference images, theoretically, the no-reference model effectively solves the practical problems of the full-reference model. However, the design difficulty of the no-reference model is that it distinguishes distortions based on the content of the test image to be evaluated. For example, most of the existing no-reference models can only be evaluated for specific content of interest or specific distortion types, and it is easy to confuse various distortion types

Images