Author: Wang, Hongbing
Title: Reproducing colour appearance of stimuli under high dynamic range (HDR) scenes
Advisors: Wei, Minchen (BEEE)
Degree: Ph.D.
Year: 2024
Department: Department of Building Environment and Energy Engineering
Pages: xxiv, 182 pages : color illustrations
Language: English
Abstract: The fast development of multi-media devices and systems provides great improvements and opportunities for reproducing colour appearance of real scenes. For example, the high dynamic range (HDR) displays in recent years allowed to produce highlights with luminance values beyond the diffuse white luminance, which commonly appear in real scenes. The dynamic ranges and colour ranges of these displays, however, are always much smaller than those in real scenes, requiring the adjustments of luminance (i.e., tone mapping) to reproduce the colour appearance. This thesis aimed to investigate how colour appearance should be characterised and processed for stimuli in HDR scenes.
Firstly, comprehensive reviews, analyses, and comparisons were made on the nonlinearity introduced by the tone-curve-based and histogram-based tone mapping methods proposed and implemented in the image processing community. The results showed that these two categories of methods generally introduced similar tone (i.e., luminance) nonlinearity, though they are commonly applied in different ways. In addition, the review clearly showed that few studies specifically investigated the change of colour appearance for tone mapping in the colour science community.
Then, two psychophysical experiments were carefully designed and carried out to investigate how lightness and chroma should be simultaneously adjusted for performing tone mapping on a uniform colour stimulus. The human observers were presented with two reference stimuli (i.e., one black and one highlight), and were asked to adjust the other stimulus until it appeared in the middle of the two reference stimuli, in terms of the colour appearance, under a D65 adapting condition. The first experiment focused on neutral stimuli, while the second experiment focused on chromatic stimuli. In particular, the two experiments covered a very wide range of adapting luminance (up to 11,000 cd/m2), luminance contrast of stimuli (up to 72,045), and the colour gamut (close to the Rec. 2020 colour gamut), all of which were never achieved in the past. The results clearly suggested that all the existing colour spaces, including those that were proposed for HDR conditions, failed to accurately characterise the perceived lightness of highlight stimuli under HDR conditions, and the relationship can be better characterised using a power function (with the power value to be varied with the adapting luminance). Moreover, none of the existing colour spaces was able to characterise the relationship between lightness and chroma for performing tone mapping. Furthermore, it was found that a greater degree of chroma compression, together with hue correction, was always needed.
The last part of the thesis focused on the hue linearity of various colour spaces. The human observers viewed two reference stimuli (i.e., one black and one stimulus with a high chroma), and adjusted the other one at a fixed chroma level until it had the same hue appearance as the stimulus with the high chroma. In particular, the stimuli almost covered the Rec. 2020 colour gamut and a total of 21 hues were investigated. The results showed that none of the nine existing colour spaces had a perfect performance in hue linearity, with the CIELAB colour space having very poor performance in the blue and red regions and the ICTCP colour space having a relatively better performance. More importantly, the experiment results derived 21 constant hue loci for the nine existing colour spaces, which can be used for hue correction when performing tone mapping or gamut mapping.
In summary, this thesis comprehensively reviewed the tone mapping methods and investigated how to accurately adjust lightness, chroma, and hue when performing tone mapping on colour stimuli under HDR conditions. The findings not only help to further understand the response of the human visual system to colour stimuli under HDR conditions but also provide very useful insights and tools for achieving accurate colour reproduction in practical applications.
Rights: All rights reserved
Access: open access

Files in This Item:
File Description SizeFormat 
7534.pdfFor All Users7.33 MBAdobe PDFView/Open


Copyright Undertaking

As a bona fide Library user, I declare that:

  1. I will abide by the rules and legal ordinances governing copyright regarding the use of the Database.
  2. I will use the Database for the purpose of my research or private study only and not for circulation or further reproduction or any other purpose.
  3. I agree to indemnify and hold the University harmless from and against any loss, damage, cost, liability or expenses arising from copyright infringement or unauthorized usage.

By downloading any item(s) listed above, you acknowledge that you have read and understood the copyright undertaking as stated above, and agree to be bound by all of its terms.

Show full item record

Please use this identifier to cite or link to this item: https://theses.lib.polyu.edu.hk/handle/200/13085