A two level algorithm for text detection in natural scene images

被引:12
|
作者
Rong, Li [1 ]
Wang Suyu [1 ]
Shi, ZhiXin [2 ]
机构
[1] Beijing Univ Technol, Sch Software Engn, Beijing, Peoples R China
[2] SUNY Buffalo, Ctr Document Anal & Recognit, Buffalo, NY 14260 USA
关键词
text detection; connected component group; conditional random field; support vector machine; graph cut; natural scene images;
D O I
10.1109/DAS.2014.41
中图分类号
TP18 [人工智能理论];
学科分类号
081104 ; 0812 ; 0835 ; 1405 ;
摘要
In this paper we present a two-level method to detect text in natural scene images. In the first level, connected components (referred as CCs) are got from the images. Then candidate text lines are extracted and groups of connected components that align in horizontal or vertical direction are got. We think CCs in these groups have high probability are texts. To validate which CC is text, a SVM is trained to make an initial decision. The output of SVM is calibrated to posterior probability. Then we use the information of posterior probability of SVM and information of whether the connected component is in a group to divide the connected components into four classes: texts, non-texts, probable texts and undetermined CCs. In the second level, a conditional random field model is used to make final decision. Relationship between CCs is modeled by a network G(V, E), Vertices of the graph correspond to CCs. The determination in the first level will influence the second level's determination by giving different parameters of data term for the four classes of CCs. By this way, we not only use information of a single CC's feature, but also use the information of whether a CC is in a group to make final decision of whether the CC is text or non-text. Experiments show that the method is effective.
引用
收藏
页码:329 / 333
页数:5
相关论文
共 50 条
  • [1] A robust algorithm for text region detection in natural scene images
    Park, Jonghyun
    Lee, Gueesang
    [J]. CANADIAN JOURNAL OF ELECTRICAL AND COMPUTER ENGINEERING-REVUE CANADIENNE DE GENIE ELECTRIQUE ET INFORMATIQUE, 2008, 33 (3-4): : 215 - 222
  • [2] Text detection of two major Indian scripts in natural scene images
    Department of Information Technology, Heritage Institute of Technology, Kolkata, India
    不详
    [J]. Lect. Notes Comput. Sci., (42-57):
  • [3] Text Detection in Natural Scene Images Using Two Masks Filtering
    Turki, Houssem
    Ben Halima, Mohamed
    Alimi, Adel M.
    [J]. 2016 IEEE/ACS 13TH INTERNATIONAL CONFERENCE OF COMPUTER SYSTEMS AND APPLICATIONS (AICCSA), 2016,
  • [4] Robust Text Detection in Natural Scene Images
    Yin, Xu-Cheng
    Yin, Xuwang
    Huang, Kaizhu
    Hao, Hong-Wei
    [J]. IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE, 2014, 36 (05) : 970 - 983
  • [5] Text Detection and Recognition in Natural Scene Images
    Pise, Amruta
    Ruikar, S. D.
    [J]. 2014 INTERNATIONAL CONFERENCE ON COMMUNICATIONS AND SIGNAL PROCESSING (ICCSP), 2014,
  • [6] Scene Text Detection in Natural Images: A Review
    Cao, Dongping
    Zhong, Yong
    Wang, Lishun
    He, Yilong
    Dang, Jiachen
    [J]. SYMMETRY-BASEL, 2020, 12 (12): : 1 - 26
  • [7] Text detection and restoration in natural scene images
    Ye, Qixiang
    Hao, Jianbin
    Huang, Jun
    Yu, Hua
    [J]. JOURNAL OF VISUAL COMMUNICATION AND IMAGE REPRESENTATION, 2007, 18 (06) : 504 - 513
  • [8] Text Detection and Recognition in Natural Scene Images
    Huang, Xiaoming
    Shen, Tao
    Wang, Run
    Gao, Chenqiang
    [J]. PROCEEDINGS OF 2015 INTERNATIONAL CONFERENCE ON ESTIMATION, DETECTION AND INFORMATION FUSION ICEDIF 2015, 2015, : 44 - 49
  • [9] Uyghur Text Detection in Natural Scene Images
    Li, Xinming
    Li, Junfang
    Gao, Qiag
    Yu, Xiao
    [J]. 2019 IEEE INTERNATIONAL CONFERENCE ON MECHATRONICS AND AUTOMATION (ICMA), 2019, : 1542 - 1547
  • [10] Cascade Detector for Text Detection in Natural Scene Images
    Hanif, Shehzad Muhammad
    Prevost, Lionel
    Negri, Pablo Augusto
    [J]. 19TH INTERNATIONAL CONFERENCE ON PATTERN RECOGNITION, VOLS 1-6, 2008, : 1917 - +