End-to-End Text Recognition Using Local Ternary Patterns, MSER and Deep Convolutional Nets

Author

Opitz, Michael ; Diem, Markus ; Fiel, Stefan ; Kleber, Florian ; Sablatnig, Robert

Author_Institution

Comput. Vision Lab., Vienna Univ. of Technol., Vienna, Austria

fYear

2014

fDate

7-10 April 2014

Firstpage

186

Lastpage

190

Abstract

Text recognition in natural scene images is an application for several computer vision applications like licence plate recognition, automated translation of street signs, help for visually impaired people or image retrieval. In this work an end-to-end text recognition system is presented. For detection an AdaBoost ensemble with a modified Local Ternary Pattern (LTP) feature-set with a post-processing stage build upon Maximally Stable Extremely Region (MSER) is used. The text recognition is done using a deep Convolution Neural Network (CNN) trained with backpropagation. The system presented outperforms state of the art methods on the ICDAR 2003 dataset in the text-detection (F-Score: 74.2%), dictionary-driven cropped-word recognition (F-Score: 87.1%) and dictionary-driven end-to-end recognition (F-Score: 72.6%) tasks.

Keywords

backpropagation; computer vision; convolution; image recognition; image retrieval; natural scenes; neural nets; text detection; AdaBoost ensemble; CNN; ICDAR 2003 dataset; LTP feature-set; MSER; automated translation; backpropagation; computer vision application; deep convolution neural network; deep convolutional nets; dictionary-driven cropped-word recognition; dictionary-driven end-to-end recognition; end-to-end text recognition system; image retrieval; licence plate recognition; local ternary pattern feature-set; maximally stable extremely region; natural scene images; post-processing stage; text-detection; visually impaired people; Character recognition; Convolution; Dictionaries; Feature extraction; Support vector machines; Text recognition; Training; adaboost; convolutional neural networks; dropout; end-to-end scene text recognition; lbp; ltp; neural networks; scene text detection; scene text recognition; text detection; text recognition;

fLanguage

English

Publisher

ieee

Conference_Titel

Document Analysis Systems (DAS), 2014 11th IAPR International Workshop on

Conference_Location

Tours

Print_ISBN

978-1-4799-3243-6

Type

conf

DOI

10.1109/DAS.2014.29

Filename

6830995