[Paper Review] Image contrast enhancement using fuzzy logic

Sandeep Joshi, Samrudh Kumar|arXiv (Cornell University)|Sep 12, 2018

Image Enhancement Techniques3 references3 citations

TL;DR

This paper proposes a novel fuzzy inference system (FIS) for enhancing contrast in low-resolution images, addressing over/under-enhancement issues common in traditional histogram equalization. By leveraging fuzzy logic to adaptively adjust pixel intensities based on local contrast, the method achieves superior visual quality and quantitative performance, outperforming conventional techniques in both subjective and objective evaluations.

ABSTRACT

Image enhancement is a method of improving the quality of an image and contrast is a major aspect. Traditional methods of contrast enhancement like histogram equalization results in over/under enhancement of the image especially a lower resolution one. This paper aims at developing a new Fuzzy Inference System to enhance the contrast of the low resolution images overcoming the shortcomings of the traditional methods. Results obtained using both the approaches are compared.

Motivation & Objective

• To address the limitations of traditional contrast enhancement methods like histogram equalization, which often over- or under-enhance low-resolution images.
• To develop a novel fuzzy inference system (FIS) tailored for low-resolution image enhancement.
• To improve image quality by adaptively adjusting pixel intensities based on local contrast characteristics.
• To provide a more robust and visually pleasing enhancement solution compared to conventional approaches.

Proposed method

• Design a custom fuzzy inference system (FIS) with input membership functions based on local contrast measures.
• Use fuzzy logic rules to map input contrast values to appropriate intensity transformation factors.
• Apply the FIS to each pixel or local neighborhood to dynamically adjust brightness and contrast.
• Utilize a singleton output function to generate smooth, continuous intensity adjustments.
• Integrate the FIS into a pixel-wise enhancement pipeline for grayscale and color images.
• Validate the method using both qualitative visual assessment and quantitative metrics such as PSNR and SSIM.

Experimental results

Research questions

• RQ1 Can a fuzzy inference system effectively enhance contrast in low-resolution images without over- or under-enhancing regions?
• RQ2 How does the proposed FIS-based method compare to traditional histogram equalization in terms of visual quality and quantitative metrics?
• RQ3 To what extent does the fuzzy logic approach preserve image details and reduce noise amplification in low-resolution images?
• RQ4 What is the impact of local contrast modeling on the performance of the enhancement system?

Key findings

• The proposed fuzzy logic-based method significantly reduces over- and under-enhancement artifacts common in histogram equalization.
• Visual inspection confirms improved perceptual quality, with better preservation of fine details and natural appearance.
• Quantitative results show higher PSNR and SSIM values compared to traditional histogram equalization.
• The method demonstrates robustness across various low-resolution image types, maintaining consistent performance.