Modeling the Quality of Videos Displayed With Local Dimming Backlight at Different Peak White and Ambient Light Levels

Abstract

This paper investigates the impact of ambient light and peak white (maximum brightness of a display) on the perceived quality of videos displayed using local backlight dimming. Two subjective tests providing quality evaluations are presented and analyzed. The analyses of variance show significant interactions of the factors peak white and ambient light with the perceived quality. Therefore, we proceed to predict the subjective quality grades with objective measures. The rendering of the frames on liquid crystal displays with light emitting diodes backlight at various ambient light and peak white levels is computed using a model of the display. Widely used objective quality metrics are applied based on the rendering models of the videos to predict the subjective evaluations. As these predictions are not satisfying, three machine learning methods are applied: partial least square regression, elastic net, and support vector regression. The elastic net method obtains the best prediction accuracy with a spearman rank order correlation coefficient of 0.71, and two features are identified as having a major influence on the visual quality.

Subjective and Objective Evaluation of Local Dimming Algorithms for HDR Images

“…Besides PSNR, Burini et al [26] also used Mean Square Error (MSE) and labPSNR which comes from PSNR to assess the impact of colour distortion of LCD images. To determine which characteristics of local backlight displays influence quality assessment, Mantel et al [27] conducted subjective and objective evaluations to investigate which aspects, such as clipping and leakage, are relevant for perceptual quality. Mantel et al [28] extended their work to investigate the impact of ambient light and peak white levels on the perceived quality of videos displayed using BLD methods.…”

Subjective and Objective Quality Assessment of Swimming Pool Images

“…The characteristic of vision is non-linear, it being too bright or too dark will cause varying degrees of damage to the quality of the image. As the bottom feature of image, brightness feature will directly affect the result of image quality evaluation (Mantel et al, 2016 ). The basic information of the image or pixel can be obtained from the brightness characteristics.…”

Machine learning for soft and liquid molecular materials

“…80 ML methods, primarily neural networks, have also been tested to control the local backlight dimming, which is important for less loss of image detail, higher contrast ratio, and low power consumption of LCDs. [81][82][83] A comparative study of local backlight dimming prediction accuracy applied to the subjective evaluation of video quality, impacted by ambient light exposure and peak white (maximum display brightness), revealed that Elastic Net algorithm performed best compared to partial least squares regression and support vector regression. 81 The CNN-based algorithm made it possible to control the backlight intensity along with reducing the loss of detail while achieving a high contrast ratio by taking into account the diffusion property of light and leakage property of liquid crystal.…”

“…[81][82][83] A comparative study of local backlight dimming prediction accuracy applied to the subjective evaluation of video quality, impacted by ambient light exposure and peak white (maximum display brightness), revealed that Elastic Net algorithm performed best compared to partial least squares regression and support vector regression. 81 The CNN-based algorithm made it possible to control the backlight intensity along with reducing the loss of detail while achieving a high contrast ratio by taking into account the diffusion property of light and leakage property of liquid crystal. 82 However, this ML method required statistical information of pixel values in each local block.…”