Affiliation:
1. Government First Grade College, Madhugiri, India & Tumkur University, India
2. BMS Institute of Technology, India
Abstract
Edge detection from handwritten text documents, particularly of Kannada language, is a challenging task. Kannada has a huge character set, amounting to 17,340 character combinations. Moreover, in handwritten Kannada, the character strokes are highly variable in size and shape due to varying handwriting styles. This chapter presents a solution for edge detection of Kannada handwritten documents. Sobel edge detection method, which efficiently enhances the image contrast and detects the character edges, is proposed. Experimentation of this edge detection approach yielded high F-measure and global contrast factor values.
Reference20 articles.
1. Alaei, Nagabhushan, & Pal. (2011). A benchmark Kannada Handwritten Document Dataset and its Segmentation. Proceedings of International Conference on Document Analysis and Recognition, 1520-5363.
2. Chandrakala, Thippeswamy, & Sahana. (2018). Enhancement of degraded document images using Retinex and Morphological operations. International Journal of Computer Science and Information Security, 16(4).
3. Chandrakala, H. T., & Thippeswamy, G. (2016). A Comprehensive Survey on OCR techniques for Kannada Script. International Journal of Science and Research, 5(4).
4. Epigraphic Document image enhancement using Retinex method;H. T.Chandrakala;Proceedings of 3rd international symposium of signal processing and intelligent recognition systems, Book chapter in Advanced in signal processing and intelligent recognition systems,2017
5. Robust text detection in natural images with edge-enhanced Maximally Stable Extremal Regions