Loading...
Please wait, while we are loading the content...
Similar Documents
Text-Line and Character Segmentation for Off-line Recognition of Handwritten Japanese Text
| Content Provider | Semantic Scholar |
|---|---|
| Author | Nguyen, Kha Cong Nakagawa, Masaki |
| Copyright Year | 2016 |
| Abstract | Text-line and character segmentation is one of the most important steps to develop an Optical Character Recognition (OCR) system for handwritten text. Due to distortions of handwriting, i.e., variations of character size, irregular gaps between characters and touching of characters, etc., however, the step is still a difficult task. This paper proposes a method for segmenting Japanese text-lines and characters from off-line handwritten text pages. The proposed method includes: text-line segmentation using a morphological method, zone projection and character separation for each segmented text-line by vertical projection, Stroke Width Transform (SWT), bridge finding and Voronoi diagrams. The method segments curved, touching and skew text-lines and complicated touching characters. |
| Starting Page | 53 |
| Ending Page | 58 |
| Page Count | 6 |
| File Format | PDF HTM / HTML |
| Volume Number | 115 |
| Alternate Webpage(s) | https://web.tuat.ac.jp/~nakagawa/pub/2015/pdf/PRMU2015-173(Text-Line_and_Character_Segmentation_for_Off-line_Recognition_of_Handw).pdf |
| Language | English |
| Access Restriction | Open |
| Content Type | Text |
| Resource Type | Article |
