Quaternion discrete cosine transform-based color image quality evaluation method

Abstract

The invention discloses a quaternion discrete cosine transform-based color image quality evaluation method, and mainly solves the problem of low consistency of objective evaluation and subjective evaluation of color image quality at present. According to the scheme, the method comprises the steps of 1) representing color three-channel information of a color image by utilizing a quaternion matrix;2) obtaining a coefficient of the image by utilizing quaternion discrete cosine transform; 3) according to a spatial frequency position, dividing the obtained quaternion color spectrum coefficient into different sub-bands; and 4) calculating the similarity between the sub-bands of an original image and a distorted image, and obtaining image quality through weighted fusion. The calculation is simple; an experimental result shows that three channels of the color image are subjected to overall processing by utilizing a quaternion; the color image can be subjected to quality evaluation more effectively and accurately; and the method can be used for processing the color image in compression, storage and transmission of the color image.

Description

technical field [0001] The invention belongs to the technical field of color image processing, in particular to a color image quality evaluation method, which can be used for color image processing in color image compression, storage, and transmission. Background technique [0002] With the advent of the era of big data, multimedia information is experiencing explosive growth at an unprecedented speed. As one of the most widely used media in multimedia information, color images can describe the objective world truly and vividly. Compared with grayscale images, color images carry more real and rich information. In order to make color images better applied in practice, it plays an important role in many fields closely related to our lives, such as biomedicine, video multimedia, security monitoring and other fields, and makes a great contribution to the quality of color images. Quantitative evaluation has very important application value. [0003] For color images, most of th...

AI Technical Summary

Problems solved by technology

[0005] However, the above methods do not take into account the influence of color distortion on the quaternion spectrum characteristics of color images, so they cannot effectively measure color distortion, and most of the existing color image quality evaluation methods are incompatible with actual use requirements in terms of performance indicators. There is still a large distance

Images