Robust Binaritation of Non-Uniform Illuminated Images
Image binarization is a basic image processing technique with many applications. Binarization methods have been incrementally improved to deal with real world conditions. A particularly important consideration is non-uniform illumination, present when using general purpose devices for image acquisit...
Main Authors: | , , , |
---|---|
Format: | Article |
Language: | Spanish |
Published: |
Universitat Politecnica de Valencia
2018-06-01
|
Series: | Revista Iberoamericana de Automática e Informática Industrial RIAI |
Subjects: | |
Online Access: | https://polipapers.upv.es/index.php/RIAI/article/view/8847 |
id |
doaj-2b04ab305afe4ff6b65d9fe427cf906b |
---|---|
record_format |
Article |
spelling |
doaj-2b04ab305afe4ff6b65d9fe427cf906b2021-02-02T09:02:57ZspaUniversitat Politecnica de ValenciaRevista Iberoamericana de Automática e Informática Industrial RIAI1697-79121697-79202018-06-0115325226110.4995/riai.2017.88476133Robust Binaritation of Non-Uniform Illuminated ImagesEdgar Molina0Julia Diaz1Hugo Hidalgo-Silva2Edgar Chávez3CICESE-Ciencias de la ComputaciónCICESE-Ciencias de la ComputaciónCICESE-Ciencias de la ComputaciónCICESE-Ciencias de la ComputaciónImage binarization is a basic image processing technique with many applications. Binarization methods have been incrementally improved to deal with real world conditions. A particularly important consideration is non-uniform illumination, present when using general purpose devices for image acquisition (e. g. mobile phones). In this paper we present a new binarization method for non uniformly illuminated text images. The main idea is to improve illumination using two complementary illuminated images. To achieve this we estimate the luminance of the original image and we create a new complementarily illuminated image.We then fuse the two images and for each pixel we estimate in which of the two images the pixel is correctly illuminated. The proposed method is evaluated with synthetic and real images, using several metrics, showing better performance than state-of-the-art techniques.https://polipapers.upv.es/index.php/RIAI/article/view/8847BinarizaciónIluminaciónFusión de Imágenes |
collection |
DOAJ |
language |
Spanish |
format |
Article |
sources |
DOAJ |
author |
Edgar Molina Julia Diaz Hugo Hidalgo-Silva Edgar Chávez |
spellingShingle |
Edgar Molina Julia Diaz Hugo Hidalgo-Silva Edgar Chávez Robust Binaritation of Non-Uniform Illuminated Images Revista Iberoamericana de Automática e Informática Industrial RIAI Binarización Iluminación Fusión de Imágenes |
author_facet |
Edgar Molina Julia Diaz Hugo Hidalgo-Silva Edgar Chávez |
author_sort |
Edgar Molina |
title |
Robust Binaritation of Non-Uniform Illuminated Images |
title_short |
Robust Binaritation of Non-Uniform Illuminated Images |
title_full |
Robust Binaritation of Non-Uniform Illuminated Images |
title_fullStr |
Robust Binaritation of Non-Uniform Illuminated Images |
title_full_unstemmed |
Robust Binaritation of Non-Uniform Illuminated Images |
title_sort |
robust binaritation of non-uniform illuminated images |
publisher |
Universitat Politecnica de Valencia |
series |
Revista Iberoamericana de Automática e Informática Industrial RIAI |
issn |
1697-7912 1697-7920 |
publishDate |
2018-06-01 |
description |
Image binarization is a basic image processing technique with many applications. Binarization methods have been incrementally improved to deal with real world conditions. A particularly important consideration is non-uniform illumination, present when using general purpose devices for image acquisition (e. g. mobile phones). In this paper we present a new binarization method for non uniformly illuminated text images. The main idea is to improve illumination using two complementary illuminated images. To achieve this we estimate the luminance of the original image and we create a new complementarily illuminated image.We then fuse the two images and for each pixel we estimate in which of the two images the pixel is correctly illuminated. The proposed method is evaluated with synthetic and real images, using several metrics, showing better performance than state-of-the-art techniques. |
topic |
Binarización Iluminación Fusión de Imágenes |
url |
https://polipapers.upv.es/index.php/RIAI/article/view/8847 |
work_keys_str_mv |
AT edgarmolina robustbinaritationofnonuniformilluminatedimages AT juliadiaz robustbinaritationofnonuniformilluminatedimages AT hugohidalgosilva robustbinaritationofnonuniformilluminatedimages AT edgarchavez robustbinaritationofnonuniformilluminatedimages |
_version_ |
1724295734671441920 |