Segmentation of Overlapping Handwritten Gurumukhi Characters
- Author Manas Kaur
- Co-Author Ashok Kumar Bathla
- DOI
- Country : India
- Subject : Engineering
OCR (Optical Character Recognition) is used to read the digitally scanned documents and convert these digitally scanned documents to computer readable format. The segmentation of offline handwritten Gurmukhi characters is quite a difficult task. This problem can be worse when there are overlapping Gurumukhi characters in same or different zones. The proposed research has adopted water reservoir algorithm. This algorithm is to detect the overlapping characters in different zones and segment them. Large amount of dataset has been tested in current technique. 94% positive results have been generated.
Comments
No have any comment !