A Semi-reference Image Quality Evaluation Method Based on Weighted Dimension of Gabor Difference Box

Abstract

The invention discloses a semi-reference image quality evaluation method based on the Gabor difference box weighted dimension. The method first performs two-dimensional Gabor transformation on the input distorted image and the reference image after grayscale, and obtains the distorted grayscale image and the reference grayscale. The Gabor sine coefficient and cosine coefficient of the degree image, and then calculate the fractal dimension of the Gabor sine coefficient and cosine coefficient of the distorted gray image and the reference gray image, and use the difference box method to obtain the weighted dimension of the difference box, and then calculate the Gabor coefficient Gabor entropy and sine weighting coefficient and cosine weighting coefficient, and finally calculate the semi-reference image quality evaluation score according to the difference box weighting dimension and weighting coefficient. The image quality evaluation method of the present invention overcomes the shortcomings of excessively large extracted image feature dimensions and inconvenient transmission, and uses Gabor entropy to weight the absolute difference of the weighted dimension features of the difference box to obtain the final objective image quality evaluation score, which improves Semi-reference image quality assessment accuracy.

Description

technical field [0001] The invention belongs to the field of image processing, in particular to a semi-reference image quality evaluation method based on Gabor difference box weighted dimensions. Background technique [0002] Image quality evaluation is to judge the image quality and give a quantitative score to measure the degree of image distortion; image quality evaluation is a key issue in the field of image processing, and image quality evaluation methods can be divided into subjective image quality evaluation methods and objective image quality evaluation methods . The subjective image quality evaluation method is based on the subjective feelings of the observers. Although it is accurate, it has the disadvantages of high cost and time-consuming; the objective image quality evaluation method uses a calculation model to automatically predict the image quality, which is low in cost and short in time. application prospects. Objective image quality assessment methods can ...

AI Technical Summary

Problems solved by technology

For semi-reference image quality evaluation, scholars at home and abroad have carried out extensive research. Wang used the natural image statistical model in the wavelet domain to extract the features of the reference image and the distorted image, and used the KL (Kullback-Leibler) distance between the features for semi-reference. Gao uses multi-order curvelets, bandlets, wavelets and contourlets transformation, and extracts the normalized histogram of the transformation coefficients for image quality evaluation; Engelke calculates the measurement of blocking effect, blurring and ringing effect respectively for image quality evaluation ; but these methods have shortcomings such as the evaluation accuracy to be improved and the extraction of too many semi-reference features that are not conducive to transmission.

Images