Abstract Text detection and localization have great importance for content based image analysis and text based image indexing. The efficiency of text recognition depends on the efficiency of text localization. So, the main goal of the proposed method is to detect and localize text regions with high accuracy. To achieve this goal, a new and efficient method has been introduced for localization of Bangla text from scene images. In order to improve precision and recall as well as f-measure, Maximally Stable Extremal Region (MSER) based method along with double filtering techniques have been used. As MSER algorithm generates many false positives, we have introduced double filtering method for removing these false positives to increase the f-measure to a great extent. Our proposed method works at three basic levels. Firstly, MSER regions are generated from the input color image by converting it into gray scale image. Secondly, some heuristic features are used to filter out most of the false positives or non-text regions. Lastly, Stroke Width Transform (SWT) based filtering method is used to filter out remaining non-text regions. Remaining components are then grouped into candidate text regions marked by bounding box over each region. As there is no benchmark database for Bangla text, the proposed method is implemented on our own prepared database consisting of 200 scene images of Bangla texts and has got prominent performance. To evaluate the performance of our proposed approach, we have also tested the proposed method on International Conference on Document Analysis and Recognition( ICDAR) 2013 benchmark database and have got a better result than the related existing methods.

References

[1] Asaduzzaman A., Molla K., and Molla G., “Printed Bangla Text Recognition using Artificial Neural Network with Heuristic Method,” in Proceedings of International Conference on Computer and Information Technology, Dhaka, pp. 27-28, 2002.

[2] Banik P., Bhattacharya U., and Parul S., “Segmentation of Bangla Words in Scene Images,” in Proceedings of the 8th Indian Conference on Computer Vision, Graphics and Image Processing, Mumbai, pp. 1-7, 2012.

[3] Bhattacharya U., Parui S., and Mondal S., “Devanagari and Bangla Text Extraction from Natural Scene Images,” in Proceedings of the 10th international Conference on Document Analysis and Recognition, Barcelona, pp. 171- 175, 2009.

[4] Chen H., Tsai S., Schroth G., Chen D., Chandrasekhar V., Takacs G., Vedantham R., Grzeszczuk R., and Girod B., “Robust Text Detection In Natural Images With Edge- Enhanced Maximally Stable Extremal Regions,” in Proceedings of IEEE International Conference on Image Processing, Brussels, pp. 2609-2612, 2011.

[5] Chen X. and Yuille A., “Detecting and Reading the Text in Natural Scenes,” in Proceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition, Washington, pp. 366-373, 2004.

[6] Chowdhury A., Bhattacharya U., and Parui S., “Text Detection of Two Major Indian Scripts in Natural Scene Images,” in Proceedings of International Workshop on Camera-Based Document Analysis and Recognition, Beijing, pp. 42-57, 2012.

[7] Computer Vision Toolbox, available at: https://www.mathworks.com/help/vision/exampl e, Last Visited, 2016.

[8] Epshtein B., Ofek E., and Wexler Y., “Detecting Text in Natural Scenes with Stroke Width Transform,” in Proceedings of Computer Society Conference on Computer Vision and Pattern Proposed Samarbandu and Gllavata et al.

[9] Islam et al.

[14] Liu

[27] 384 The International Arab Journal of Information Technology, Vol. 17, No. 3, May 2020 Recognition, San Francisco, pp. 2963-2970, 2010.

[9] Gllavata J., Ewerth R., and Freisleben B., “A Robust Algorithm for Text Detection in Images,” in Proceedings of the 3rd International Symposium on Image and Signal Processing and Analysis, Rome, pp. 611-616, 2003.

[10] Ghoshal R., Roy A., Bhowmik T., and Parui S., “Headline based Text Extraction from Outdoor Images,” in Proceedings of the 4th International conference Pattern Recognition, and Machine Intelligence, Moscow, pp. 446-451, 2011.

[11] Ghoshal R., Roy A., and Parui S., “Recognition of Bangla Text from Scene Images through Perspective Correction,” in Proceedings of International Conference on Image Information Processing, Shimla, pp. 1-6, 03 2011.

[12] He T., Huang W., Qiao Y., and Yao J., “Text Attentional Convolutional Neural Network for Scene Text Detection,” IEEE Transaction on Image Processing, vol. 25, no. 6, pp. 2529-2541, 2016.

[13] Islam M. and Mondal A., “Towards a Standard Bangla PhotoOCR: Text Detection and Localization,” in Proceedings of 17th International Conference on Computer and Information Technology, Dhaka, pp. 198-203, 2014.

[14] Islam R., Islam M., and Talukder K., “An Approach to Extract Text Regions from Scene Image,” in Proceedings of International Conference on Computing, Analytics and Security Trends, Pune, pp. 1-6, 2016.

[15] Kartaz D., Shafait F., Uchida S., Iwamura M., BigordaL., Mestre S., Mas J., Mota D., Almazan J., and HerasL.,” ICDAR 2013 Robust Reading Competition,” in Proceedings of the 12th International Conference on Document Analysis and Recognition, Washington, pp. 1484-1493, 2013.

[16] Li Y. and Lu H., “Scene Text Detection via Stroke Width,” in Proceedings of 21st International Conference on Pattern Recognition, Tsukuba, pp. 681-684, 2012.

[17] Maximally Stable Extremal Regions, available at: https://en.wikipedia.org/wiki/Maximally_stable_e xtremal_regions, Last Visited, 2016.

[18] Measure properties of image regions, available at: http://www.mathworks.com/help/images/ref/regio nprops.html, Last Visited, 2016.

[19] Matas J., Chum O., Urban M., and Paul T., “Robust Wide Baseline Stereo From Maximally Stable Extremal Regions,” Image and Vision Computing, vol. 22, no. 10, pp. 761-767, 2004.

[20] Nahar K., “Off-line Arabic Hand-Writing Recognition Using Artificial Neural Network with Genetics Algorithm,” The International Arab Journal of Information Technology, vol. 15, no. 4, pp. 701-707, 2018.

[21] Neumann L. and Matas J., “On Combining Multiple Segmentations in Scene Text Recognition,” in Proceedings of the 12th International Conference on Document Analysis and Recognition, Washington, pp. 523-527, 2013.

[22] Neumann L. and Matas J., “Scenetextlocalizationand Recognition with Oriented Stroke Detection,” in Proceedings of IEEE International Conference on Computer Vision, Sydney, pp. 97-104, 2013.

[23] Neumann L. and Matas J., “Efficient Scene Text Localization and Recognition with Local Character Refinement,” in Proceedings of the 13th International Conference on Document Analysis and Recognition, Tunis, pp. 746-750, 2015.

[24] Otsu N., “A Threshold Selection Method From Gray-Level Histograms,” IEEE Transactions on Systems, Man, and Cybernetics, vol. 9, no. 1, pp. 62-66, 1979.

[25] Pan F., Hou X., and Liu L., “A Hybrid Approach to Detect and Localize Texts in Natural Scene Images,” IEEE Transaction on Image Processing, vol. 20, no. 3, pp. 800-813, 2011.

[26] Region Detectors, available at: http://micc.unifi.it/dElbimbo/wpcontent/uploads/ 2011/03/slide_coroso/A34%20MSER.pdf, Last Visited, 2017.

[27] Samarabandu J. and Liu X., “An Edge-based text Region Extraction Algorithm for Indoor Mobile Robot Navigation,” International Journal of Computer, Electrical, Automation, Control and Information Engineering, vol. 1, no. 7, pp. 2043- 2050, 2007.

[28] Shahab A., Shafait F., and Dengel A., “ICDAR 2011 Robust Reading Competition Challenge 2: Reading Scene Images,” in Proceedings of International Conference on Document Analysis and Recognition, Beijing, pp. 1491-1496, 2011.

[29] Shi C., Wang C., Xio B., Zhang Y., and Gao S., “Scene Text Detection Using Graph Model Built Upon Maximally Stable Extremal Regions,” Pattern Recognition Letters, vol. 34, no. 2, pp. 107-116, 2013.

[30] Shivakumara P., Phan T., and Tan C., “A Laplacian Approach to Multi-Oriented Text Detection in Video,” IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 33, no. 2, pp. 412-419, 2011.

[31] Sun L., Huo Q., Jia W., and Chen K., “A Robust Approach for Text Detection from Natural Scene Images,” Pattern Recognit, vol. 48, no. 9, pp. 2906-2920, 2015.

[32] Yin X., Yin X., Huang K., and Hao H., “Robust Text Detection in Natural Scene Images,” IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 36, no. 5, pp. 970-983, 2014. An Enhanced MSER Pruning Algorithm for Detection and Localization of Bangla ... 385

[33] Zarechensky M., “Text Detection in Natural Scenee with The Multilingual Text,” in the Proceedings of the 10th Spring Researcher’s Colloquium on Database and Information Systems, Veliky Novgorod, 2014.

[34] Zhang Z., Shen W., Yao C., and Bai X., “Symmetry-Based Text Line Detection In Natural Scenes,” in Proceedings of IEEE Conference on Computer Vision and Pattern Recognition, Boston, pp. 2558-2567, 2015. Rashedul Islam received B.Sc. degree in Computer Science and Engineering (CSE) from Khulna University, Khulna, Bangladesh in 2002 and received M.Sc. degree in Computer Science and Engineering (CSE) from Uttara University, Dhaka, Bangladesh in 2011. He is working as an Assistant Professor in the Department of Information and Communication Technology (ICT) of Rajuk Uttara Model College, Uttara, Dhaka, Bangladesh. Currently, he is a Ph.D. student in the discipline of Computer Science and Engineering, Khulna University, Khulna, Bangladesh. He is an author of the book Higher Secondary Information and Communication Technology-for class XI-XII (English version). He has published four papers, which have been published in journals as well as in refereed international conference proceedings published by IEEE and other. Rashedul Islam is a member of the Institution of Engineers, Bangladesh (IEB). His present research interest includes Im age processing, Artificial Neural Network, Bio-metrics etc. Rafiqul Islam obtained Ph.D. in Computer Science from Universiti Teknologi Malaysia (UTM) in 1999 and a combined Master (MS) and Bachelor Degree in Engineering (Computers) from Azerbaijan Polytechnic Institute (Azerbaijan Technical University at present) in 1987. He was a visiting fellow (a postdoctoral researcher) in Japan Advance Institute of Science and Technology (JAIST) in 2001. He worked as head of the Discipline of Computer Science and Engineering of Khulna University and as the Dean of the School of Science, Engineering and Technology of Khulna University. He worked as a Professor in the Department of Computer Science of American International University- Bangladesh (AIUB) from 2009 to 2015. Currently, he is a senior professor of Computer Science and Engineering Discipline of Khulna University, Khulna, Bangladesh. He has 27 years of teaching and research experiences. He has published more than 100 papers, which have been published in international and national journals as well as in refereed international conference proceedings published by IEEE, Springer and others. His research areas include design and analysis of algorithms in the area of image processing, secure cloud computing, external sorting, Information security, Network security data compression, bio- informatics, grid computing, cloud computing etc. Currently, he is doing researches on optimization process using metaheuristic algorithms. Recently his several papers have been published in the journals with high impact factors. Kamrul Talukder has been a Professor in Computer Science and Engineering (CSE) Discipline of Khulna University (KU) in Bangladesh since 2011. Prof. Talukder was the head of CSE Discipline, KU for three years. He completed his B.Sc. in CSE with distinction from Khulna University in 1999, M.Sc. in Computer Science (by research) from National University of Singapore (NUS) in 2004 and D.Eng. from Hiroshima University in 2008. He was a recipient of the prestigious Japan Society for the Promotion of Science (JSPS) Postdoctoral fellowship for the duration of two years. Dr. Talukder is a life fellow of the Institution of Engineers, Bangladesh (IEB). His research interest is mainly focused on Image Processing, Formal Verification and Software Engineering. He has published more than 50 research publications in the various proceedings of international conferences and in journals.