An optical character recognition approach to qualifying thresholding algorithms

Authors:
Margaret Sturgill;Steven J. Simske
Affiliations:
Hewlett Packard Labs, Fort Collins, CO, USA;Hewlett Packard Labs, Fort Collins, CO, USA
Venue:
Proceedings of the eighth ACM symposium on Document engineering
Year:
2008

Citing 2
Cited 3

Minimum error thresholding

Pattern Recognition
Meta-algorithmic systems for document classification

Proceedings of the 2006 ACM symposium on Document engineering

Document engineering approaches toward scalable and structured multimedia, web and printable documents

Multimedia Tools and Applications
Layout-aware limiarization for readability enhancement of degraded historical documents

Proceedings of the 9th ACM symposium on Document engineering
Enhancement of historical printed document images by combining Total Variation regularization and Non-local Means filtering

Image and Vision Computing

Quantified Score

Hi-index	0.00

Visualization

Abstract

Pre-processing for raster image based document segmentation begins with image thresholding, which is a binarization process separating foreground from background. In this paper, we compare an existing (Otsu), modified existing (Kittler-Illingworth) and simple peak-based thresholding approach on a set of 982 documents for which existing ground truth (full text) is available. We use the output of an open source OCR engine which incorporates an adaptive/dynamic thresholder that can be bypassed by one of the three global thresholds we tested. This allowed comparison of these three approaches in the aggregate. We then used an independently-generated dictionary as a means of characterizing thresholder efficacy. Such an approach, if successful, will provide the means for selecting an optimal thresholder in the absence of a large set of ground truthed documents. Our preliminary findings here indicate that this approach may provide a reliable means for thresholder comparison and eventually preclude the need for time-intensive human ground truthing.