An Analysis of RGB, Hue and Grayscale under Various Illuminations

Closed

Hurriyatul Fitriyah, Randy Cahya Wihandika

2018 3rd International Conference on Sustainable Information Engineering and Technology, SIET 2018 - Proceedings Conference paper Cited by 13 Quartile

Abstract

Hue is an appearance parameter model which theoretically shows only color information. In cases where illumination varies from locations and conditions, the hue in HSV/HSL/HSI color space is theoretically remains alike. This study gives an experimental analysis on whether hue value is similar in various illuminations. Analysis of RGB and grayscale under the same situation are also presented. The experiment used objects which have primary colors of red, green and blue in several variation such as dark vs. light color, indoor vs. outdoor location and dark vs. bright illumination. The result shows that hue has standard deviation of <16 (range 0°-360°) for all variations. It is very small compared to RGB which has standard deviation of > 52 (range 0-255) for all variations. Moreover, RGB values dropped drastically under a very low light condition. An analysis on grayscale was also performed. It results that objects' color cannot be differentiated in this color-space. Grayscale differentiates object well when based on brightness as its value on brighter objects or illuminations were larger than darker object or illumination. © 2018 IEEE.

Affiliations

Computer Engineering, Universitas Brawijaya, Malang, Indonesia; Informatics Engineering, Faculty of Computer Science, Universitas Brawijaya, Malang, Indonesia