OCR binarization and image pre-processing for searching historical documents

by: Maya R Gupta, Nathaniel P Jacobson, Eric K Garcia

Pattern Recognition, Vol. 40, No. 2. (February 2007), pp. 389-397.

View FullText article

DOI, ScienceDirect

Reviews [Write a review of this article]

There are no reviews of this article

Find related articles from these CiteULike users

tgyork

Find related articles with these CiteULike tags

background, binarisation, historical, image-processing, ocr, preprocessing

Abstract

We consider the problem of document binarization as a pre-processing step for optical character recognition (OCR) for the purpose of keyword search of historical printed documents. A number of promising techniques from the literature for binarization, pre-filtering, and post-binarization denoising were implemented along with newly developed methods for binarization: an error diffusion binarization, a multiresolutional version of Otsu's binarization, and denoising by despeckling. The OCR in the ABBYY FineReader 7.1 SDK is used as a black box metric to compare methods. Results for 12 pages from six newspapers of differing quality show that performance varies widely by image, but that the classic Otsu method and Otsu-based methods perform best on average.

@article{ocrhistproc06, abstract = {We consider the problem of document binarization as a pre-processing step for optical character recognition (OCR) for the purpose of keyword search of historical printed documents. A number of promising techniques from the literature for binarization, pre-filtering, and post-binarization denoising were implemented along with newly developed methods for binarization: an error diffusion binarization, a multiresolutional version of Otsu's binarization, and denoising by despeckling. The OCR in the ABBYY FineReader 7.1 SDK is used as a black box metric to compare methods. Results for 12 pages from six newspapers of differing quality show that performance varies widely by image, but that the classic Otsu method and Otsu-based methods perform best on average.}, author = {Gupta, Maya R. and Jacobson, Nathaniel P. and Garcia, Eric K. }, citeulike-article-id = {2799352}, doi = {10.1016/j.patcog.2006.04.043}, journal = {Pattern Recognition}, keywords = {background, binarisation, historical, image-processing, ocr, preprocessing}, month = {February}, number = {2}, pages = {389--397}, posted-at = {2008-05-14 17:01:27}, priority = {0}, title = {OCR binarization and image pre-processing for searching historical documents}, url = {http://dx.doi.org/10.1016/j.patcog.2006.04.043}, volume = {40}, year = {2007} }

RIS record

TY - JOUR ID - ocrhistproc06 L3 - citeulike-article-id:2799352 TI - OCR binarization and image pre-processing for searching historical documents JF - Pattern Recognition VL - 40 IS - 2 SP - 389 EP - 397 N2 - We consider the problem of document binarization as a pre-processing step for optical character recognition (OCR) for the purpose of keyword search of historical printed documents. A number of promising techniques from the literature for binarization, pre-filtering, and post-binarization denoising were implemented along with newly developed methods for binarization: an error diffusion binarization, a multiresolutional version of Otsu's binarization, and denoising by despeckling. The OCR in the ABBYY FineReader 7.1 SDK is used as a black box metric to compare methods. Results for 12 pages from six newspapers of differing quality show that performance varies widely by image, but that the classic Otsu method and Otsu-based methods perform best on average. KW - background KW - binarisation KW - historical KW - image-processing KW - ocr KW - preprocessing AU - Gupta, Maya AU - Jacobson, Nathaniel AU - Garcia, Eric PY - 2007/02// UR - http://dx.doi.org/10.1016/j.patcog.2006.04.043 ER -

RIS BibTeX