Color Distribution Index (CDI) for Evaluating the Display Image Quality
نویسندگان
چکیده
In this research, we proposed a new image quality evaluation index Color Distribution Index (CDI). It considers color gamut, contrast ratio, maximum luminance, electro-optical characteristics such as gamma, bit depth and JND as human visual system to evaluate the ability for reproducing visually smooth gradation. When the CDI cell was defined, we verified JND from fundamental matter such as JND in luminance and chromaticity using the latest TV sets. According to our investigations, luminance JND was L/L=0.01, chromaticity JND was u’v’=0.001. We used these values for a unit CDI cell and evaluated bit depth needed for high image quality without any visible false contour. We clarified 12 bit is needed as necessary and sufficient condition to reproduce high image quality with latest LCD TV set designed based on BT.709. And we clarified that maximum luminance, contrast ratio, gamma characteristics and bit depth should be balanced to achieve TV sets of high image quality from the viewpoint of the total image quality improvement. Introduction Currently development of high image quality FPD is going on rapidly. Liquid Crystal Display (LCD) TV tends to have higher luminance, higher contrast ratio and wider color gamut. Now in latest LCD TV, the maximum luminance and contrast ratio are 450 cd/m and 3,000:1 respectively [1]. Furthermore LCD TV with contrast ratio over 100,000:1 is developed [2]. LCD TV whose color gamut is much wider than HDTV is recently commercialized [3]. As High Dynamic Range (HDR) Display, the maximum luminance and contrast ratio 3,000 cd/m and 50,000:1 respectively is being developed [4]. In the future, it is prospected that LCD TV will have higher contrast ratio and wider color gamut. To use high ability of LCDs effectively for reproducing image, higher bit depth will be needed for high image quality. Under these circumstances, considering luminance, contrast ratio and color gamut, this research aims to develop an evaluation method of display image quality and reveal the bit depth which is necessary to distribute reproduced colors perceptually uniformly in all over the color gamut of LCD TV. One purpose of high bit depth is to reproduce smooth picture without any false contour under any condition. For this purpose, to reveal the necessary bit depth, the investigations by Just Noticeable Difference (JND) which use uniform color space (L*a*b* color space or Munsell color space) and Eab* are performed [5], [6]. Yoshida [7] conducted the subjective evaluations under various conditions of adaptation luminance using LCD TV to reveal the bit depth needed for high image quality with smooth gradation. Then in the case of severe condition like observing low Average Picture Level (APL) images in dark room, some visible contour appeared even if Eab* was under 1. These results originated from the fact that L*a*b* color space assumed that observer evaluated under enough daylight and should set 100% white as reference. So Yoshida indicated that it was difficult to apply the result based on L*a*b* to the case of severe condition like observing low APL pictures in dark room and the evaluation of display should use the luminance JND based on absolute luminance. In this paper, we assume that colors are needed to be reproduced less than JND in all region of reproducible color gamut. For that purpose, we propose Color Distribution Index (CDI). For calculating CDI, JND is an important factor. There are many researches on JND [8]. But we confirm it again with the newest instruments and we inspect the suitability of luminance JND from the Weber fraction and the chromaticity JND from subjective evaluations. Then we estimate the bit depth needed for high image quality FPD TV using CDI considering the luminance and chromaticity JND. We discuss about the bit depth needed for high image quality TV considering physical characteristics of FPD TV such as the contrast ratio and maximum luminance from the viewpoint of JND. In this paper, we use CIE u’v’ which is known as uniform chromaticity scale and L as an absolute luminance L. Color Distribution Index; CDI Up to now, the evaluations of color reproduction system has been performed with color difference ( Eab*) in L*a*b* color space. In the evaluation, a 100% white was set as the reference and it was difficult to apply the result to the case of severe condition like observing low APL pictures in dark room [7]. And if the reproduced colors distribute in a narrow region, the value of color difference will be small, but large Eab* might appear in the other regions in color space which TV set can reproduce. Therefore there was a possibility that the ratio of large value of Eab* decreases even though the system have not enough color reproduction ability for reproducing visually smooth gradation. As a result, the actual image quality might not agree with the evaluation by color difference. Considering these problems, TV sets for reproducing visually smooth images need to have sufficiently small color difference 152 ©2008 Society for Imaging Science and Technology v ’ L-JND u’v’-JND L u’ between adjoining colors and reproduced colors distribute uniformly in reproducible color gamut. So in this research we propose Color Distribution Index (CDI) which does not set 100% reference white, instead it uses absolute luminance value. That is, the purpose of CDI is to evaluate the distribution of colors which TV set can reproduce in three dimensional color space. Fig. 1 shows a three dimensional feature of the CDI cell. First the reproducible color gamut is divided by the luminance JND and the chromaticity JND, and each cell is checked whether it includes one or more reproduced color or not. Each reproduced color is calculated by
منابع مشابه
Effect of Some Synthetic Parameters on Size and Polydispersity Index of Gelatin Nanoparticles Cross-Linked by CDI/NHS System
In our previous work, the effect of use of a water soluble CDI/NHS system as nontoxic cross-linking agent on fabrication of gelatin nanoparticles was investigated. In this research, the effect of variation in some synthetic parameters of gelatin nanoparticles cross-linked by CDI/NHS system such as type of gelatin and formulation of cross- linking agent on their size and distribution was examine...
متن کاملReduced-Reference Image Quality Assessment based on saliency region extraction
In this paper, a novel saliency theory based RR-IQA metric is introduced. As the human visual system is sensitive to the salient region, evaluating the image quality based on the salient region could increase the accuracy of the algorithm. In order to extract the salient regions, we use blob decomposition (BD) tool as a texture component descriptor. A new method for blob decomposition is propos...
متن کاملNo-Reference Image Quality Assessment Using Euclidean Distance Matrices
Image quality assessment (IQA) methods play important roles in many applications such as image communication, reception, compression, restoration, and display. No-reference IQA metrics are required to resolve an image when there is a lack of a reference image that is required for fullreference IQA metrics. We propose a no-reference IQA method to evaluate the image quality by using the differenc...
متن کاملModeling of Texture and Color Froth Characteristics for Evaluation of Flotation Performance in Sarcheshmeh Copper Pilot Plant, Using Image Analysis and Neural Networks
Texture and color appearance of froth is a discreet qualitative tool for evaluating the performance of flotation process. The structure of a froth developed on the flotation cell has a significant effect on the grade and recovery of copper concentrate. In this work, image analysis and neural networks have been implemented to model and control the performance of such a system. The result reveals...
متن کاملEvaluating colour image difference metrics for gamut-mapped images
The quality of several colour image difference metrics, pixelwise CIELAB DEab, S-CIELAB, iCAM, Structural Similarity Index, Universal Image Quality and the hue-angle algorithm, have been investigated. These results were compared with the results from a psychophysical experiment in which the perceptual image difference was evaluated. Six original images were reproduced using six different colour...
متن کامل