Improving ocraccuracy for scanned historical newspapers
OCR is part of the computer vision field and the use of it has grown rapidly in recent decade due to the incessant demand in document digitization. Different techniques are used by OCR tools to process varieties of input formats(.pdf, doc, .jpeg, etc.). However, from our point of view, no research...
Saved in:
| Main Author: | |
|---|---|
| Format: | Final Year Project Report |
| Language: | English |
| Published: |
Universiti Malaysia Sarawak, (UNIMAS)
2013
|
| Subjects: | |
| Online Access: | http://ir.unimas.my/id/eprint/39020/3/Naiker%20Nithyananthan%20ft.pdf http://ir.unimas.my/id/eprint/39020/ |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Summary: | OCR is part of the computer vision field and the use of it has grown rapidly in recent decade due to the incessant demand in document digitization. Different techniques are used by OCR tools to process varieties of input formats(.pdf, doc, .jpeg, etc.). However, from our point of view, no research has been done in applying the chain code technique on historical documents stored in image format. In this project, one variant of the chain code algorithm known as Compare images algorithm is presented when it has been tuned to process some samples of Sarawak Gazette. Experimental results show relatively high accuracy improvement (approximately 6.90%). Future works will focus on testing the algorithm to other historical documents. |
|---|