Identifying Strikeouts and Predicting Synonym Words for Kannada Handwritten Documents

Computer Engineering

Handwritten character recognition (HCR) is a pivotal facet within the realm of machine learning and artificial intelligence. This research delves into the intricate challenges associated with detecting and recognizing Kannada handwritten document images. The paper introduces a novel methodology designed to identify strike-outs, addressing a significant aspect of document analysis. In this research, we employ dilation and contour analysis techniques to efficiently discriminate between sentences and individual strikeout words, enhancing the accuracy of our recognition system. Furthermore, we present an innovative synonym generation mechanism for the identified strikeout words, augmenting the utility of our approach in the context of document understanding and information retrieval. This work contributes to the advancement of HCR, especially in the context of Kannada script, and underscores the significance of tackling nuanced aspects such as strike-outs for comprehensive document analysis and interpretation.

This work is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License.