Uniform distorted scene reduction on distribution of colour cast correction

Muhamad Lazim Talib, Mohamad Faidzul Nasruddin, Siti Norul Huda Sheikh Abdullah

Research output: Contribution to journalArticle

Abstract

Scene in the photo occulated by uniform particles distribution can degrade the image quality accidently. State of the art pre-processing methods are able to enhance visibility by employing local and global filters on the image scene. Regardless of air light and transmission map right estimation, those methods unfortunately produce artifacts and halo effects because of uncorrelated problem between the global and local filter's windows. Besides, previous approaches might abruptly eliminate the primary scene structure of an image like texture and colour. Therefore, this study aims not solely to improve scene image quality via a recovery method but also to overcome image content issues such as the artefacts and halo effects, and finally to reduce the light disturbance in the scene image. We introduce our proposed visibility enhancement method by using joint ambience distribution that improves the colour cast in the image. Furthermore, the method is able to balance the atmospheric light in correspondence to the depth map accordingly. Consequently, our method maintains the image texture structural information by calculating the lighting estimation and maintaining a range of colours simultaneously. The method is tested on images from the Benchmarking Single Image Dehazing research by assessing their clear edge ratio, gradient, range of saturated pixels, and structural similarity metric index. The scene image restoration assessment results show that our proposed method had outperformed resuls from the Tan, Tarel and He methods by gaining the highest score in the structural similarity index and colourfulness measurement. Furthermore, our proposed method also had achieved acceptable gradient ratio and percentage of the number of saturated pixels. The proposed approach enhances the visibility in the images without affecting them structurally.

Original languageEnglish
Pages (from-to)24-31
Number of pages8
JournalInternational Journal on Advanced Science, Engineering and Information Technology
Volume9
Issue number1
DOIs
Publication statusPublished - 1 Jan 2019

Fingerprint

Visibility
Color
Image quality
color
Pixels
Image texture
Benchmarking
Image reconstruction
Epidemiologic Effect Modifiers
filters
Textures
Lighting
methodology
Recovery
Light
Artifacts
Processing
Air
texture
processing technology

Keywords

  • Colour cast
  • Colour correction
  • Uniform distorted scene

ASJC Scopus subject areas

  • Computer Science(all)
  • Agricultural and Biological Sciences(all)
  • Engineering(all)

Cite this

Uniform distorted scene reduction on distribution of colour cast correction. / Talib, Muhamad Lazim; Nasruddin, Mohamad Faidzul; Sheikh Abdullah, Siti Norul Huda.

In: International Journal on Advanced Science, Engineering and Information Technology, Vol. 9, No. 1, 01.01.2019, p. 24-31.

Research output: Contribution to journalArticle

@article{17519ed69e7646af9df5d2f757eba2e5,
title = "Uniform distorted scene reduction on distribution of colour cast correction",
abstract = "Scene in the photo occulated by uniform particles distribution can degrade the image quality accidently. State of the art pre-processing methods are able to enhance visibility by employing local and global filters on the image scene. Regardless of air light and transmission map right estimation, those methods unfortunately produce artifacts and halo effects because of uncorrelated problem between the global and local filter's windows. Besides, previous approaches might abruptly eliminate the primary scene structure of an image like texture and colour. Therefore, this study aims not solely to improve scene image quality via a recovery method but also to overcome image content issues such as the artefacts and halo effects, and finally to reduce the light disturbance in the scene image. We introduce our proposed visibility enhancement method by using joint ambience distribution that improves the colour cast in the image. Furthermore, the method is able to balance the atmospheric light in correspondence to the depth map accordingly. Consequently, our method maintains the image texture structural information by calculating the lighting estimation and maintaining a range of colours simultaneously. The method is tested on images from the Benchmarking Single Image Dehazing research by assessing their clear edge ratio, gradient, range of saturated pixels, and structural similarity metric index. The scene image restoration assessment results show that our proposed method had outperformed resuls from the Tan, Tarel and He methods by gaining the highest score in the structural similarity index and colourfulness measurement. Furthermore, our proposed method also had achieved acceptable gradient ratio and percentage of the number of saturated pixels. The proposed approach enhances the visibility in the images without affecting them structurally.",
keywords = "Colour cast, Colour correction, Uniform distorted scene",
author = "Talib, {Muhamad Lazim} and Nasruddin, {Mohamad Faidzul} and {Sheikh Abdullah}, {Siti Norul Huda}",
year = "2019",
month = "1",
day = "1",
doi = "10.18517/ijaseit.9.1.6437",
language = "English",
volume = "9",
pages = "24--31",
journal = "International Journal on Advanced Science, Engineering and Information Technology",
issn = "2088-5334",
publisher = "INSIGHT - Indonesian Society for Knowledge and Human Development",
number = "1",

}

TY - JOUR

T1 - Uniform distorted scene reduction on distribution of colour cast correction

AU - Talib, Muhamad Lazim

AU - Nasruddin, Mohamad Faidzul

AU - Sheikh Abdullah, Siti Norul Huda

PY - 2019/1/1

Y1 - 2019/1/1

N2 - Scene in the photo occulated by uniform particles distribution can degrade the image quality accidently. State of the art pre-processing methods are able to enhance visibility by employing local and global filters on the image scene. Regardless of air light and transmission map right estimation, those methods unfortunately produce artifacts and halo effects because of uncorrelated problem between the global and local filter's windows. Besides, previous approaches might abruptly eliminate the primary scene structure of an image like texture and colour. Therefore, this study aims not solely to improve scene image quality via a recovery method but also to overcome image content issues such as the artefacts and halo effects, and finally to reduce the light disturbance in the scene image. We introduce our proposed visibility enhancement method by using joint ambience distribution that improves the colour cast in the image. Furthermore, the method is able to balance the atmospheric light in correspondence to the depth map accordingly. Consequently, our method maintains the image texture structural information by calculating the lighting estimation and maintaining a range of colours simultaneously. The method is tested on images from the Benchmarking Single Image Dehazing research by assessing their clear edge ratio, gradient, range of saturated pixels, and structural similarity metric index. The scene image restoration assessment results show that our proposed method had outperformed resuls from the Tan, Tarel and He methods by gaining the highest score in the structural similarity index and colourfulness measurement. Furthermore, our proposed method also had achieved acceptable gradient ratio and percentage of the number of saturated pixels. The proposed approach enhances the visibility in the images without affecting them structurally.

AB - Scene in the photo occulated by uniform particles distribution can degrade the image quality accidently. State of the art pre-processing methods are able to enhance visibility by employing local and global filters on the image scene. Regardless of air light and transmission map right estimation, those methods unfortunately produce artifacts and halo effects because of uncorrelated problem between the global and local filter's windows. Besides, previous approaches might abruptly eliminate the primary scene structure of an image like texture and colour. Therefore, this study aims not solely to improve scene image quality via a recovery method but also to overcome image content issues such as the artefacts and halo effects, and finally to reduce the light disturbance in the scene image. We introduce our proposed visibility enhancement method by using joint ambience distribution that improves the colour cast in the image. Furthermore, the method is able to balance the atmospheric light in correspondence to the depth map accordingly. Consequently, our method maintains the image texture structural information by calculating the lighting estimation and maintaining a range of colours simultaneously. The method is tested on images from the Benchmarking Single Image Dehazing research by assessing their clear edge ratio, gradient, range of saturated pixels, and structural similarity metric index. The scene image restoration assessment results show that our proposed method had outperformed resuls from the Tan, Tarel and He methods by gaining the highest score in the structural similarity index and colourfulness measurement. Furthermore, our proposed method also had achieved acceptable gradient ratio and percentage of the number of saturated pixels. The proposed approach enhances the visibility in the images without affecting them structurally.

KW - Colour cast

KW - Colour correction

KW - Uniform distorted scene

UR - http://www.scopus.com/inward/record.url?scp=85062400298&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85062400298&partnerID=8YFLogxK

U2 - 10.18517/ijaseit.9.1.6437

DO - 10.18517/ijaseit.9.1.6437

M3 - Article

AN - SCOPUS:85062400298

VL - 9

SP - 24

EP - 31

JO - International Journal on Advanced Science, Engineering and Information Technology

JF - International Journal on Advanced Science, Engineering and Information Technology

SN - 2088-5334

IS - 1

ER -