TEDLESS – Text detection using least-square SVM from natural scene

Text detection from the natural scene is considered to be a challenging problem due to the complex background, varied light intensity at different locations, a large variety of colors, diverse font style and size. This paper focusses on detecting candidate text objects from the scene. The image is i...

Full description

Bibliographic Details
Main Authors: Leena Mary Francis, N. Sreenath
Format: Article
Language:English
Published: Elsevier 2020-03-01
Series:Journal of King Saud University: Computer and Information Sciences
Online Access:http://www.sciencedirect.com/science/article/pii/S131915781730126X
id doaj-766959e0478a40b2aa16ebef49dd8a20
record_format Article
spelling doaj-766959e0478a40b2aa16ebef49dd8a202020-11-25T01:48:29ZengElsevierJournal of King Saud University: Computer and Information Sciences1319-15782020-03-01323287299TEDLESS – Text detection using least-square SVM from natural sceneLeena Mary Francis0N. Sreenath1Corresponding author.; Department of Computer Science and Engineering, Pondicherry Engineering College, Pondicherry 605014, IndiaDepartment of Computer Science and Engineering, Pondicherry Engineering College, Pondicherry 605014, IndiaText detection from the natural scene is considered to be a challenging problem due to the complex background, varied light intensity at different locations, a large variety of colors, diverse font style and size. This paper focusses on detecting candidate text objects from the scene. The image is initially preprocessed to remove the noise and enhance the contrast. Then the various objects of the scene are marked and extracted forming a pool of objects. A set of candidate text objects are extracted from this pool of objects and given as output. In order to locate text candidates among these objects, we use Least-Square Support Vector Machine Technique, which trains the model using Char 74K character dataset and CIFAR 10 non-text image dataset. Finally, the trained model was applied to perform a binary classification of text and non-text objects. The results were evaluated over ICDAR 2015 scene images, MSRA500 and SVT datasets and also have been compared to other approaches acquiring encouraging results. Keywords: Text detection, Support Vector Machine, Least Square Support Vector Machine, Machine Learning, Natural scene text extractionhttp://www.sciencedirect.com/science/article/pii/S131915781730126X
collection DOAJ
language English
format Article
sources DOAJ
author Leena Mary Francis
N. Sreenath
spellingShingle Leena Mary Francis
N. Sreenath
TEDLESS – Text detection using least-square SVM from natural scene
Journal of King Saud University: Computer and Information Sciences
author_facet Leena Mary Francis
N. Sreenath
author_sort Leena Mary Francis
title TEDLESS – Text detection using least-square SVM from natural scene
title_short TEDLESS – Text detection using least-square SVM from natural scene
title_full TEDLESS – Text detection using least-square SVM from natural scene
title_fullStr TEDLESS – Text detection using least-square SVM from natural scene
title_full_unstemmed TEDLESS – Text detection using least-square SVM from natural scene
title_sort tedless – text detection using least-square svm from natural scene
publisher Elsevier
series Journal of King Saud University: Computer and Information Sciences
issn 1319-1578
publishDate 2020-03-01
description Text detection from the natural scene is considered to be a challenging problem due to the complex background, varied light intensity at different locations, a large variety of colors, diverse font style and size. This paper focusses on detecting candidate text objects from the scene. The image is initially preprocessed to remove the noise and enhance the contrast. Then the various objects of the scene are marked and extracted forming a pool of objects. A set of candidate text objects are extracted from this pool of objects and given as output. In order to locate text candidates among these objects, we use Least-Square Support Vector Machine Technique, which trains the model using Char 74K character dataset and CIFAR 10 non-text image dataset. Finally, the trained model was applied to perform a binary classification of text and non-text objects. The results were evaluated over ICDAR 2015 scene images, MSRA500 and SVT datasets and also have been compared to other approaches acquiring encouraging results. Keywords: Text detection, Support Vector Machine, Least Square Support Vector Machine, Machine Learning, Natural scene text extraction
url http://www.sciencedirect.com/science/article/pii/S131915781730126X
work_keys_str_mv AT leenamaryfrancis tedlesstextdetectionusingleastsquaresvmfromnaturalscene
AT nsreenath tedlesstextdetectionusingleastsquaresvmfromnaturalscene
_version_ 1725011866237796352