Abstract:As an effective technique to extend the depth-of-field (DOF) of optical lenses, multi-focus image fusion has recently become an active topic in image processing community. However, a major problem remaining unsolved in this field is the lack of universal criteria in selecting objective evaluation metrics. Consequently, the metrics utilized in different studies often vary significantly, leading to high difficulties in achieving unbiased evaluation. To address this problem, this paper proposes a statistic-based approach for verifying the effectiveness of objective metrics in multi-focus image fusion. The core idea is to adopt statistical correlation measures to evaluate the performance consistency between a certain fusion metric and some popular full-reference image quality assessment models. In addition, a convolutional neural network (CNN)-based fusion metric is presented to measure the similarity between the source images and the fused image based on the semantic features at multiple abstraction levels. A comparative study is conducted to evaluate 20 existing fusion metrics using the proposed statistic-based approach on a large-scale, realistic and with-ground-truth multi-focus image fusion dataset recently released. Experimental results demonstrate the feasibility of the proposed approach in evaluating the effectiveness of objective metrics and the advantage of our CNN-based metric. Resources will be released at https://github.com/yuliu316316.

A Multi-Metric Fusion Approach To Visual Quality Assessment

Multi-Metric Fusion Network for Image Quality Assessment

A new image fusion performance metric based on visual information fidelity

Novel Full-Reference Image Quality Assessment Metric Based on Entropy Fusion

Rethinking the Effectiveness of Objective Evaluation Metrics in Multi-focus Image Fusion: A Statistic-based Approach

A New Image Quality Approach Based on Decision Fusion

Selection of image fusion quality measures: objective, subjective, and metric assessment

Hyperspectral Image Quality Evaluation Based On Multi-Model Fusion

Perceived Assessment Metrics for Visible and Infrared Color Fused Image Quality Without Reference Image

Image Quality Assessment Using Contourlet Transform

Image Quality Assessment by Decision Fusion

Quality Assessment of Multimodal Medical Image Fusion Images Using Chromaticity Similarity and Information Similarity

A New Approach to Video Quality Assessment Based on Human Perception and Fusion

Image Fusion Quality Metrics By Directional Projection

Image Quality Assessment Based on Multi-scale Geometric Analysis

A Novel Metric Based on Mca for Image Quality

Superpixel-Based Quality Assessment of Multi-Exposure Image Fusion for Both Static and Dynamic Scenes

Paired Image-to-Image Translation Quality Assessment Using Multi-Method Fusion

Video Multimethod Assessment Fusion Based Rate-Distortion Optimization for Versatile Video Coding

Perceptual Quality Assessment for Multi-Exposure Image Fusion