Abstract

Text line segmentation is one of the pre-stages of modern optical character recognition systems. The algorithmic approach proposed by this paper has been designed for this exact purpose. Its main characteristic is the combination of two different techniques, morphological image operations and horizontal histogram projections. The method was developed to be applied on a historic data collection that commonly features quality issues, such as degraded paper, blurred text, or presence of noise. For that reason, the segmenter in question could be of particular interest for cultural institutions, that want access to robust line bounding boxes for a given historic document. Because of the promising segmentation results that are joined by low computational cost, the algorithm was incorporated into the OCR pipeline of the National Library of Luxembourg, in the context of the initiative of reprocessing their historic newspaper collection. The general contribution of this paper is to outline the approach and to evaluate the gains in terms of accuracy and speed, comparing it to the segmentation algorithm bundled with the used open source OCR software.

Highlights

  • While adapting modern open source OCR software to reprocess a large collection of historic newspapers, ranging from years 1841 to 1954, the National Library of Luxembourg (BnL) explored ways to segment the newspaper scans into individual text lines

  • To running faster than BENCH, the aim was for the method to only take up a fraction of the time needed for the OCR pipeline’s character recognition functionality

  • The final but essential step of COMBISEG is the analysis of horizontal histogram projections, one created for every bounding box stored in boxes

Read more

Summary

CONTEXT

While adapting modern open source OCR software to reprocess a large collection of historic newspapers, ranging from years 1841 to 1954, the National Library of Luxembourg (BnL) explored ways to segment the newspaper scans into individual text lines Pursuing this goal, a method was developed that integrates into a larger OCR pipeline by sitting just in between the binarization and font recognition processes. That’s in the form of the kraken.pageseg.segment function that mostly relies on different filters from scipy.ndimage (Virtanen et al [2020]) This algorithm, as published with version 2.0.8 1 and in the following referred to as BENCH, essentially served as a benchmark in the context of the development of an own solution, designed for the precise needs of BnL. The subsequent section evaluates the method and draws the comparison to BENCH

ALGORITHM
Input Assumption
Segmentation
Morphology
Components
Histogram
Output
Parameters
Value Determination
EXPERIMENTAL RESULTS
Evaluation Method
Postprocessing
Results
CONCLUSION
Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call