A novel scene text detection algorithm based on convolutional neural network

Xiaohang Ren, Kai Chen, Xiaokang Yang, Yi Zhou, Jianhua He, Jun Sun

Research output: Chapter in Book/Published conference outputConference publication


Candidate text region extraction plays a critical role in convolutional neural network (CNN) based text detection from natural images. In this paper, we propose a CNN based scene text detection algorithm with a new text region extractor. The so called candidate text region extractor I-MSER is based on Maximally Stable Extremal Region (MSER), which can improve the independency and completeness of the extracted candidate text regions. Design of I-MSER is motivated by the observation that text MSERs have high similarity and are close to each other. The independency of candidate text regions obtained by I-MSER is guaranteed by selecting the most representative regions from a MSER tree which is generated according to the spatial overlapping relationship among the MSERs. A multi-layer CNN model is trained to score the confidence value of the extracted regions extracted by the I-MSER for text detection. The new text detection algorithm based on I-MSER is evaluated with wide-used ICDAR 2011 and 2013 datasets and shows improved detection performance compared to the existing algorithms.

Original languageEnglish
Title of host publicationVCIP 2016 - 30th Anniversary of Visual Communication and Image Processing
Number of pages4
ISBN (Electronic)978-1-5090-5316-2
Publication statusPublished - 4 Jan 2017
Event2016 IEEE Visual Communication and Image Processing - Chengdu, China
Duration: 27 Nov 201630 Nov 2016


Conference2016 IEEE Visual Communication and Image Processing
Abbreviated titleVCIP 2016

Bibliographical note



  • CNN
  • deep learning model
  • I-MSER
  • scene text detection
  • text region extraction


Dive into the research topics of 'A novel scene text detection algorithm based on convolutional neural network'. Together they form a unique fingerprint.

Cite this