Full-Reference Image Quality Assessment Based on Spatial-Domain Combined with Frequency-Domain Analysis
zhiming shi, Xinrong Wu · Research Square · 2023
Abstract This paper designs a full reference image quality assessment method. Firstly, the image features are extracted from the spatial domain of the image to represent the edge information of the image. Secondly, the spatial frequency and texture features of the image are extracted in the frequency domain. Thirdly, the acquired features are grouped and fused into new features. Finally, the random forest is used to model full reference method. Through many experiments, the effectiveness of the proposed methods is verified, and the accuracy of the assessment is improved better than classical image assessment method.