This article proposes a scheme for automatic recognition of Bangla text extracted from outdoor scene images. For extraction, first the headline is obtained, then certain conditions are applied to distinguish between text and non-text. By removing the headline, the Bangla text is partitioned into two zones. Further, an association among the text symbols in these two different zones is observed. For recognition purpose, a decision tree classifier is designed with Multilayer Perceptron (MLP) at leaf nodes. The root node takes into account all possible text symbols. Further nodes highlight distinguishable features and act as a two-class classifiers. Finally, at leaf nodes, a few text symbols remain, that are recognized using MLP classifiers. The association between the two zones makes recognition simpler and efficient. The classifiers are trained using about 7100 samples of 52 classes. Experiments are performed on 250 images (200 scene images and 50 scanned images).
JungK., KimI.K., KurataT., KourogiM. and HanH.J., Text scanner with text detection technology on image sequences, Proc. of Int. Conf. on Pattern Recognition3 (2002), 473-476.
2.
LiangJ., DoermannD. and LiH., Camera based analysis of text and documents: A survey. Int. Journ. on Doc. Anal. and Recog. (IJDAR)7 (2005), 84-104.
3.
BhattacharyaU., ParuiS.K. and MondalS., Devanagari and bangla text extraction from natural scene images, Proc. of the Int. Conf. on Document Analysis and Recognition (2009), 171-175.
4.
PalU. and ChaudhuriB.B., Indian script character recognition: A survey, Pattern Recognition37 (2004), 1887-1899.
5.
MandalA.K., PalS., DeA.K. and MitraS., Novel approach to identify good tracer clouds from a sequence of satellite images, IEEE T. Geoscience and Remote Sensing43(4) (2005), 813-818.
6.
FigueiredoM.A.T. and JainA.K., Unsupervised learning of finite mixture models. IEEE Trans. on PAMI24(3) (2002), 381-396.
7.
BiernackiC., CeleuxG. and GovaertG., Assessing a mixture model for clustering with the integrated completed likelihood, IEEE Trans. Pattern Anal. Mach. Intell.22(7) (2000), 719-725.
8.
RoyA., ParuiS.K., PaulA. and RoyU., A color based image segmentation and its application to text segmentation, Proc. of Ind. Conf. on Computer Vision, Graphics & Image Processing, (2008), 313-319.
9.
Di ZenzoS., CinqueL. and LevialdiS., Run-Based Algorithms for Binary Image Analysis and Processing, IEEE Trans. Pattern Anal. Mach. Intell.18(1) (1996), 83-89.
10.
ChaudhuriB.B. and PalU., A complete printed bangla ocr system, Pattern Recognition31 (1998), 531-549.
11.
ParuiS.K., BhattacharyaU., DattaA. and ShawB., A database of handwritten bangla vowel modifiers and a scheme for their detection and recognition, Proc. of Workshop on Computer Vision Graphics and Image Processing, (2006), 204-209.
12.
BhowmikT.K., GhantyP., RoyA. and ParuiS.K., Svm-based hierarchical architectures for handwritten bangla character recognition, Int. Journ. on Doc. Anal. and Recog. (IJDAR)12 (2009), 83-96.
13.
ChenX. and YuilleA., Detecting and reading text in natural scenes, Proceedings of IEEE Conference of Computer Vision and Pattern Recognition (CVPR)2 (2004), 366-373.
14.
LeeJ.-J., LeeP.-H., LeeS.-W., YuilleA. and KochC., Adaboost for text detection in natural scene, Proceedings of International Conference of Document Analysis and Recognition (ICDAR) (2011), 429-434.
15.
EpshteinB., OfekE. and WexlerY., Detecting text in natural scenes with stroke width transform. Proceedings of IEEE Conference of Computer Vision and Pattern Recognition (CVPR) (2010), 2963-2970.
16.
YiC. and TianY., Text string detection from natural scenes by structure-based partition and grouping, IEEE Trans. on Image Processing20(9) (2011), 2594-2605.
17.
YiC. and TianY., Localizing text in scene images by boundary clustering, stroke segmentation, and string fragment classification, IEEE Trans. on Image Processing21(9) (2012), 4256-4268.
18.
PanY.-F., HouX. and LiuC.-L., A hybrid approach to detect and localize texts in natural scene images, IEEE Trans. on Image Processing20(3) (2011), 800-813.
19.
ShahabA., ShafaitF. and DengelA., ICDAR robust reading competition challenge 2: Reading text in scene images, Proceedings of the International Conference of Document Analysis and Recognition (2011), 1491-1496.
20.
ShahabA., ShafaitF. and DengelA., A head-mounted device for recognizing text in natural scenes, Proceedings of the Fourth International Workshop CBDAR, Beijing, China 2011, pp. 29-41.
21.
NeumannL. and MatasJ., Real-time scene text localization and recognition, Proceedings of the IEEE Conference of Computer Vision and Pattern Recognition (CVPR) (2012), 3538-3545.
22.
JungK., KimK. and JainA., Text information extraction in images and video: A survey, Pattern Recognition37(5) (2004), 977-997.
23.
YinX.-C., YinS., HuangK. and HaoH.-W., Robust text detection in natural scene images, IEEE Trans. Pattern Anal. Mach. Intell36(5) (2014), 970-983.
24.
GhoshalR., RoyA. and ParuiS.K., Recognition of Bangla text from Scene Images through Perspective Correction. Proc. of International Conference on Image Information Processing (ICIIP). (2011).
