A Novel Approach for Gujarati Handwritten Text Lines and Words Segmentation

摘要

Text line segmentation is an important step towards any automatic recognition of offline text document. The variation in inter-line gaps and skewed or curled text lines are some of the challenging issues in segmentation of handwritten text document. Moreover, Indian languages like Gujarati having modifier characters like matras and diacritical marks, which makes segmentation process more difficult. The selection of appropriate segmentation strategy is very important step in construction of character recognition system. This paper describes the process of line segmentation and word segmentation process in Gujarati text document. We propose technique to segment Gujarati handwritten text document image into text lines and words based on smearing technique and connected component method. This entire connected component contains extracted Gujarati words.