A No-reference Color Image Quality Evaluation Method Based on Local Binary Model

Abstract

The invention discloses a non-reference color image quality evaluation method based on a local binary pattern. The non-reference color image quality evaluation method comprises the steps of: calculating mutual information, a mean value, variance, a contrast ratio and information entropy of a distorted image, regarding the mutual information, the mean value, the variance, the contrast ratio and the information entropy as feature values of the distorted image, and constructing multi-dimensional feature vectors of the distorted image; training the multi-dimensional feature vectors of the distorted image and corresponding human eye subjective scores by utilizing support vector regression analysis, so as to obtain a function relationship model between the multi-dimensional feature vectors of the distorted image and the human eye subjective scores; and regarding multi-dimensional feature vectors of a distorted image to be evaluated as input values of the function relationship model, and acquiring an output value of the function relationship model serving as a quality evaluation value of the distorted image to be evaluated. The non-reference color image quality evaluation method fully considers color change of the image, and can conduct quality evaluation on the color image more efficiently and precisely.

Description

technical field [0001] The invention belongs to the field of color image quality evaluation, and more particularly relates to a reference-free color image quality evaluation method based on a local binary mode. Background technique [0002] With the continuous advancement of social informatization, the relationship between digital images and human production and life is becoming increasingly close, and people's requirements for image quality are also getting higher and higher. A good image should have the characteristics of beauty, clarity, rich layers, and prominent goals, which can give people more information and make it easy for people to accept and understand. However, in all links involved in digital image processing, including image acquisition, storage, encoding and compression, transmission, reconstruction, etc., images are easily affected by uncontrollable factors such as transmission media, processing technology, imaging systems, and object motion. Therefore, ima...

AI Technical Summary

Problems solved by technology

[0009] (1) This patent document needs to convert the color image into a grayscale image. The grayscale processing removes the color information and only retains the brightness component for subsequent processing. The features used in this method are mainly composed of the brightness information of the image, and there is no Extract features from the color itself, but color information plays an important role in the process of human perception of images, so this method has the problem of inaccurate evaluation results, and the applicable distortion range of this method is limited, only suitable for evaluating noise and blur image

[0010] (2) When measuring the quality of the distorted image in this patent document, low-pass filtering and downsampling are required for the reference image and the distorted image. Although the computational complexity is reduced to a certain extent, the image inevitably introduces Secondary distortion, which reduces the effectiveness and representativeness of subsequent feature extraction

The image quality evaluation method still has a large distance from the actual use requirements in terms of performance indicators, and the consistency with human subjective perception is poor

[0011] (3) The patent document requires that the original reference image must be compared for evaluation, and the reference image must be clear and of good quality, but in many practical application scenarios, the reference image cannot be obtained, which restricts the use of this method

Images