Super Resolution Image Visual Quality Assessment Based on Feature Optimization

Lei Shu

Abstract:

Most existing no-referenced image quality assessment (NR-IQA) algorithms need to extract features ‹rst and then predict image
quality. However, only a small number of features work in the model, and the rest will degrade the model performance.
Consequently, an NR-IQA framework based on feature optimization is proposed to solve this problem and apply to the SR-IQA
‹eld. In this study, we designed a feature engineering method to solve this problem. Speci‹cally, the features associate with the SR
images were ‹rst collected and aggregated. Furthermore, several advanced feature selection algorithms were used to sort the
feature sets according to their importance, and the importance matrix of features is obtained. •en, we examined the linear
relationship between the number of features and Pearson linear correlation coe’cient (PLCC) to determine the optimal number
of features and the optimal feature selection algorithm, so as to obtain the optimal model. •e results showed that the image
quality scores predicted by the optimal model are in good agreement with the human subjective scores. Adopting the proposed
feature optimization framework, we can e”ectively reduce the numberoffeatures inthe modelandobtainbetter performance.•e
experimental results indicated that SR image quality can be accurately predicted using only a small part of image features. In
summary,weproposedafeatureoptimizationframework tosolvethe currentproblem ofirrelevant features in SR-IQA, and anSR
image quality assessment model was proposed consequently.