Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

No-reference Quality Assessment of Contrast-distorted Images using Contrast-enhanced Pseudo Reference

Mohammad-Ali Mahmoudpour, Saeed Mahmoudpour

TL;DR

The paper tackles the challenge of no-reference image quality assessment for contrast-distorted images by introducing PRICCE, which generates pseudo-reference images via a set of contrast enhancement algorithms and uses a classifier to select the most suitable one for a given distorted image. This converts NR-IQA into a full-reference-like evaluation against the pseudo-reference, enabling accurate quality estimation without a true reference. A large dataset of 49,500 images is created by applying 33 distortion levels to 1,500 Waterloo images to train the classifier, which uses a ResNet-18 architecture and achieves 85% test accuracy. When integrated with FR-IQA metrics (notably MS-SSIM), PRICCE achieves competitive or superior performance on CCID2014, TID2013, and CSIQ compared to general NR-IQA methods and approaches RR-IQA methods, demonstrating strong MOS correlations and practical utility for contrast distortion assessment.

Abstract

Contrast change is an important factor that affects the quality of images. During image capturing, unfavorable lighting conditions can cause contrast change and visual quality loss. While various methods have been proposed to assess the quality of images under different distortions such as blur and noise, contrast distortion has been largely overlooked as its visual impact and properties are different from other conventional types of distortions. In this paper, we propose a no-reference image quality assessment (NR-IQA) metric for contrast-distorted images. Using a set of contrast enhancement algorithms, we aim to generate pseudo-reference images that are visually close to the actual reference image, such that the NR problem is transformed to a Full-reference (FR) assessment with higher accuracy. To this end, a large dataset of contrast-enhanced images is produced to train a classification network that can select the most suitable contrast enhancement algorithm based on image content and distortion for pseudo-reference image generation. Finally, the evaluation is performed in the FR manner to assess the quality difference between the contrast-enhanced (pseudoreference) and degraded images. Performance evaluation of the proposed method on three databases containing contrast distortions (CCID2014, TID2013, and CSIQ), indicates the promising performance of the proposed method.

No-reference Quality Assessment of Contrast-distorted Images using Contrast-enhanced Pseudo Reference

TL;DR

The paper tackles the challenge of no-reference image quality assessment for contrast-distorted images by introducing PRICCE, which generates pseudo-reference images via a set of contrast enhancement algorithms and uses a classifier to select the most suitable one for a given distorted image. This converts NR-IQA into a full-reference-like evaluation against the pseudo-reference, enabling accurate quality estimation without a true reference. A large dataset of 49,500 images is created by applying 33 distortion levels to 1,500 Waterloo images to train the classifier, which uses a ResNet-18 architecture and achieves 85% test accuracy. When integrated with FR-IQA metrics (notably MS-SSIM), PRICCE achieves competitive or superior performance on CCID2014, TID2013, and CSIQ compared to general NR-IQA methods and approaches RR-IQA methods, demonstrating strong MOS correlations and practical utility for contrast distortion assessment.

Abstract

Contrast change is an important factor that affects the quality of images. During image capturing, unfavorable lighting conditions can cause contrast change and visual quality loss. While various methods have been proposed to assess the quality of images under different distortions such as blur and noise, contrast distortion has been largely overlooked as its visual impact and properties are different from other conventional types of distortions. In this paper, we propose a no-reference image quality assessment (NR-IQA) metric for contrast-distorted images. Using a set of contrast enhancement algorithms, we aim to generate pseudo-reference images that are visually close to the actual reference image, such that the NR problem is transformed to a Full-reference (FR) assessment with higher accuracy. To this end, a large dataset of contrast-enhanced images is produced to train a classification network that can select the most suitable contrast enhancement algorithm based on image content and distortion for pseudo-reference image generation. Finally, the evaluation is performed in the FR manner to assess the quality difference between the contrast-enhanced (pseudoreference) and degraded images. Performance evaluation of the proposed method on three databases containing contrast distortions (CCID2014, TID2013, and CSIQ), indicates the promising performance of the proposed method.

Paper Structure

This paper contains 10 sections, 6 equations, 5 figures, 3 tables.

Table of Contents

  1. Introduction
  2. Proposed Method
  3. Contrast Enhancement Algorithms
  4. Dataset Generation
  5. Classification Network
  6. Experimental Results
  7. Conclusion
  8. Declarations
  9. Funding
  10. Conflict of Interest

Figures (5)

  • Figure 1: Overview of proposed PRICCE
  • Figure 2: Samples of the generated dataset in different types and levels of distortions
  • Figure 3: Labelling process of the generated dataset
  • Figure 4: Confusion matrix of image enhancement algorithms
  • Figure 5: Scatter plots of MOS vs proposed PRICCE scores. Pseudo-references and distorted images are compared under different FR-IQA methods. (a-c) DIST (d-f) GMSD (g-i) SSIM (j-l) VIF (m-o) MS-SSIM