# Perception Evaluation -- A new solar image quality metric based on the multi-fractal property of texture features

## TL;DR

This paper introduces a novel solar image quality metric called perception evaluation, which leverages multi-fractal texture features extracted by deep neural networks and measures image quality based on the cosine distance of Gram matrices.

## Contribution

The paper proposes a new reduced-reference image quality metric for solar images using multi-fractal texture features and deep neural networks, which is robust across different scenes.

## Key findings

- Perception evaluation accurately estimates solar image quality.
- The metric performs well on both simulated and real images.
- It provides a robust assessment when using high-resolution reference images.

## Abstract

Next-generation ground-based solar observations require good image quality metrics for post-facto processing techniques. Based on the assumption that texture features in solar images are multi-fractal which can be extracted by a trained deep neural network as feature maps, a new reduced-reference objective image quality metric, the perception evaluation is proposed. The perception evaluation is defined as cosine distance of Gram matrix between feature maps extracted from high resolution reference image and that from blurred images. We evaluate performance of the perception evaluation with simulated and real observation images. The results show that with a high resolution image as reference, the perception evaluation can give robust estimate of image quality for solar images of different scenes.