Perceptual Analysis of Tone Mapping Operators

Abstract

Tone mapping is a crucial step in rendering, as it maps the High Dynamic Range (HDR) images to a Lower Dynamic Range (LDR) for display on standard devices. In this research, we address the challenge of selecting the most appropriate Tone Mapping Operator (TMO) based on perceptual differences, which are essential for achieving visually aesthetic and physically plausible images. In this article, we evaluate several existing TMOs using two novel perceptual parameters: Structural Similarity Index (SSIM) and Color Difference Formula (CIEDE2000). By utilizing these novel parameters, we aim to provide a comprehensive analysis of the performance of various TMOs under different scenarios. Our experimental results reveal significant variations in the performance of the evaluated tone mapping operators. Some operators exhibit superior performance in preserving structural information and color accuracy, while others may excel in certain specific scenarios. By shedding light on the perceptual differences among the evaluated tone mapping operators, this article facilitates the effective utilization of tone mappers for achieving desired visual outcomes. Additionally, the proposed SSIM and CIEDE2000 metrics serve as valuable tools for future research and benchmarking in the field of image processing.

Keywords

• Tone Mapping Operators
• High Dynamic Range
• Image Quality
• Perceptual
• SSIM