Novel approaches towards slope and slant correction for tri-script handwritten word images

ABSTRACT

Slope and slant correction of offline handwritten word images are two of the major pre-processing steps in document image processing, because these reduce the variations in writing, thereby make further processing of the same much easier. This paper presents novel slope and slant correction methods that are applied in three different script handwritten words namely Devanagari, Bangla and Roman. The language dependency and the computational complexity of state-of-the-art approaches towards the word level slope and slant correction are addressed here. A new technique for approximate core region detection is introduced here for skew detection and then linear regression is recursively applied to de-skew the word image. Whereas, in case of slant correction, a novel cost function over the vertical projection of de-skewed image is designed and optimized to fix the uniform slant angle of text words. A new benchmarked database is developed herein to evaluate the proposed methods both quantitatively and qualitatively. Comparison of the performances by our methods with some existing slope and slant correction methods reveals that our methods are more accurate and faster.

KEYWORDS:

• Slope
• slant
• core-region
• projection profile
• frequency curve
• handwritten word

Disclosure statement

No potential conflict of interest was reported by the authors.

Notes on contributors

Radib Kar is presently a third year student of Bachelor of Computer Science & Engineering at Jadavpur University, Kolkata, India. His research interests comprise Document Image Binarization, Image Processing, Pattern Recognition, Machine Learning and Analytics. He can be reached at [email protected].

Souvik Saha is presently a third year student of Bachelor of Computer Science & Engineering at Jadavpur University, Kolkata, India. His research interests comprise Document Image Binarization, Image Processing, Pattern Recognition, Machine Learning and Analytics. He can be reached at [email protected].

Suman Kumar Bera received his B.Sc. and M.Sc. degrees in Computer Science from Vidyasagar University in 2012 and 2014 respectively. He achieved gold-medal for securing the first position in M.Sc. and got selected as an INSPIRE Fellow (DST) in 2016. He is currently working as PhD Research Scholar in Department of Computer Science and Engineering, Jadavpur University. His areas of research interests are Image Processing, Pattern Recognition and Machine Learning. He can be reached at [email protected].

Ergina Kavallieratou was born in Kefalonia, Greece, in 1973. She received her Diploma in Electrical and Computer Engineering in 1996 from the Polytechnic School of the University of Patras and her PhD in Handwritten Optical Character Recognition and Document Image processing from the same department in 2000. She has worked as guest researcher in the Signals, Systems and Radiocomunications Laboratory of the Department of Telecommunications Engineering of the Polytechnic School of Madrid (1997–1998, 2018), in the Institute of Communication Acoustics of Ruhr-Universitaet Bochum, Germany (2000, 2001), Computer Science & Engineering, Lehigh University - USA (December 2009), CVC, Universidad de Barcelona (June 2011), Telecom ParisTech (2015). During the years 2002-2004, she was an Assistant Professor of Audio Processing in Department of Audio and Musical Instruments Technology in the Technological Educational Institute of Ionian Islands, Greece. She teaches in Greek Open University, 2001-2013. Since September 2004, she is a member of the teaching staff of the department of Information and Communication System Engineering, University of the Aegean, as Associate Professor since 2017. Her research interests include Optical Character Recognition, Document Image Analysis, Computer Vision and Robotics. She can be reached at [email protected].

Vikrant Bhateja is Associate Professor, Department of Electronics & Communication Engineering, Shri Ramswaroop Memorial Group of Professional Colleges (SRMGPC), Lucknow and also the Head (Academics & Quality Control) in the same college. He is doctorate in Bio-Medical Imaging & Signal Processing and has a total academic teaching experience of 16 years with around 125 publications in reputed international conferences, journals and online book chapter contributions. His areas of research include digital image and video processing, computer vision, medical imaging, machine learning, pattern analysis and recognition. Dr. Vikrant has edited 15 proceeding books/editorial volumes with Springer Nature. He is Editor-in-Chief of IGI Global–International Journal of Natural Computing and Research (IJNCR). He is also associate editor in International Journal of Synthetic Emotions (IJSE) and International Journal of Ambient Computing and Intelligence (IJACI) under IGI Global press. He is guest editor Special Issues in reputed Scopus/SCIE indexed journals under Springer-Nature: “Evolutionary Intelligence” and “Arabian Journal of Science and Engineering”. He can be reached at [email protected].

Ram Sarkar received his Bachelor degree in Computer Science & Engineering from University of Calcutta in 2003. He received his Master degree in Computer Science & Engineering, and PhD degree in Engineering from Jadavpur University in 2005 and 2012 respectively. He joined Jadavpur University as an Assistant Professor in 2008. He received Fulbright-Nehru Fellowship (by USIEF) for post-doctoral research in University of Maryland, College Park, USA during 2014-15. His areas of current research interest are Document Image Processing, Machine Learning and Bioinformatics etc. He is a senior member of the IEEE, USA. He can be reached at [email protected].

Additional information

Funding

The Suman Kumar Bera of this paper has been funded by DST-INSPIRE fellowship under Govt. of India.