Please use this identifier to cite or link to this item:
|Title:||Camera document restoration for ocr||Authors:||Lu, S.
|Issue Date:||2005||Citation:||Lu, S.,Tan, C.L. (2005). Camera document restoration for ocr. Proceedings of the 1st International Workshop on Camera-Based Document Analysis and Recognition, CBDAR 2005 : 17-24. ScholarBank@NUS Repository.||Abstract:||As camera resolution increases, high-speed noncontact text capture through a digital camera is opening up a new channel for text capture and understanding. Unfortunately, the skew, perspective, and geometric distortions coupled within the captured images make it hard to recognize the document text using the generic OCR systems. In this paper, we propose a document restoration technique, which is capable of removing the three types of distortions, and reconstructing the fronto-parallel view of the document text using a single document image captured through a digital camera. Different from the reported techniques, the proposed restoration technique is carried out based on the vertical stroke boundary and the top line and base line of text lines. Experimental results show the proposed technique is fast, accurate, and robust.||Source Title:||Proceedings of the 1st International Workshop on Camera-Based Document Analysis and Recognition, CBDAR 2005||URI:||http://scholarbank.nus.edu.sg/handle/10635/78052|
|Appears in Collections:||Staff Publications|
Show full item record
Files in This Item:
There are no files associated with this item.
checked on Apr 6, 2019
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.