TU Darmstadt / ULB / TUbiblio

Text Localization in Born-Digital Images of Advertisements

Siegmund, Dirk ; Wainakh, Aidmar ; Ebert, Tina ; Braun, Andreas ; Kuijper, Arjan (2018)
Text Localization in Born-Digital Images of Advertisements.
Iberoamerican Conference on Pattern Recognition (CIARP). Valparaíso, Chile (November 7–10, 2017)
doi: 10.1007/978-3-319-75193-1_75
Conference or Workshop Item, Bibliographie

Abstract

Localizing text in images is an important step in a number of applications and fundamental for optical character recognition. While born-digital text localization might look similar to other complex tasks in this field, it has certain distinct characteristics. Our novel approach combines individual strengths of the commonly used methods: stroke width transform and extremal regions and combines them with a method based on edge-based morphologically growing. We present a parameterfree method with high flexibility to varying text sizes and colorful image elements. We evaluate our method on a novel image database of different retail prospects, containing textual product information. Our results show a higher f-score than competitive methods on that particular task.

Item Type: Conference or Workshop Item
Erschienen: 2018
Creators: Siegmund, Dirk ; Wainakh, Aidmar ; Ebert, Tina ; Braun, Andreas ; Kuijper, Arjan
Type of entry: Bibliographie
Title: Text Localization in Born-Digital Images of Advertisements
Language: English
Date: 2018
Place of Publication: Cham
Publisher: Springer
Book Title: Progress in Pattern Recognition, Image Analysis, Computer Vision, and Applications
Series: Lecture Notes in Computer Science (LNCS)
Series Volume: 10657
Event Title: Iberoamerican Conference on Pattern Recognition (CIARP)
Event Location: Valparaíso, Chile
Event Dates: November 7–10, 2017
DOI: 10.1007/978-3-319-75193-1_75
URL / URN: https://doi.org/10.1007/978-3-319-75193-1_75
Abstract:

Localizing text in images is an important step in a number of applications and fundamental for optical character recognition. While born-digital text localization might look similar to other complex tasks in this field, it has certain distinct characteristics. Our novel approach combines individual strengths of the commonly used methods: stroke width transform and extremal regions and combines them with a method based on edge-based morphologically growing. We present a parameterfree method with high flexibility to varying text sizes and colorful image elements. We evaluate our method on a novel image database of different retail prospects, containing textual product information. Our results show a higher f-score than competitive methods on that particular task.

Uncontrolled Keywords: Patter recognition, Computer vision
Divisions: 20 Department of Computer Science
20 Department of Computer Science > Mathematical and Applied Visual Computing
Date Deposited: 10 Jul 2019 08:50
Last Modified: 05 Jul 2024 06:31
PPN:
Export:
Suche nach Titel in: TUfind oder in Google
Send an inquiry Send an inquiry

Options (only for editors)
Show editorial Details Show editorial Details