Adaptive binarization is an important first step in many document analysis and OCR processes. This paper
describes a fast adaptive binarization algorithm that yields the same quality of binarization as the Sauvola
method,1 but runs in time close to that of global thresholding methods (like Otsu's method2), independent of
the window size. The algorithm combines the statistical constraints of Sauvola's method with integral images.3
Testing on the UW-1 dataset demonstrates a 20-fold speedup compared to the original Sauvola algorithm.

https://www.dfki.de/web/forschung/publikationen/renameFileForDownload?filename=FsDkTmbEfficientImplSpie2008.pdf&file_id=uploads_596

Efficient implementation of local adaptive thresholding techniques using integral images

Document image binarization is of great importance in the document image analysis and recognition pipeline since it affects further stages of the recognition process. The evaluation of a binarization method aids in studying its algorithmic behavior, as well as verifying its effectiveness, by providing qualitative and quantitative indication of its performance. This paper addresses a pixel-based binarization evaluation methodology for historical handwritten/machine-printed document images. In the proposed evaluation scheme, the recall and precision evaluation measures are properly modified using a weighting scheme that diminishes any potential evaluation bias. Additional performance metrics of the proposed evaluation scheme consist of the percentage rates of broken and missed text, false alarms, background noise, character enlargement, and merging. Several experiments conducted in comparison with other pixel-based evaluation measures demonstrate the validity of the proposed evaluation scheme.

Performance Evaluation Methodology for Historical Document Image Binarization

While modern off-the-shelf OCR engines show particularly high accuracy on scanned text, text detection and recognition in natural images still remains a challenging problem. Here, we demonstrate that OCR engines can still perform well on this harder task as long as appropriate image binarization is applied to input photographs. For such binarization, we systematically evaluate the performance of 12 binarization methods as well as of a new binarization algorithm that we propose here. Our evaluation includes different metrics and uses established natural image text recognition benchmarks (ICDAR 2003 and ICDAR 2011). Our main finding is thus the fact that image binarization methods combined with additional filtering of generated connected components and off-the-shelf OCR engines can achieve state-of-the-art performance for end-to-end text understanding in natural images.

/pdf/image-binarization-for-end-to-end-text-understanding-in-3us1gl15g5.pdf

Image Binarization for End-to-End Text Understanding in Natural Images

This work focuses on the most commonly used binarization method: Sauvola's. It performs relatively well on classical documents, however, three main defects remain: the window parameter of Sauvola's formula does not fit automatically to the contents, it is not robust to low contrasts, and it is not invariant with respect to contrast inversion. Thus, on documents such as magazines, the contents may not be retrieved correctly, which is crucial for indexing purpose. In this paper, we describe how to implement an efficient multiscale implementation of Sauvola's algorithm in order to guarantee good binarization for both small and large objects inside a single document without adjusting manually the window size to the contents. We also describe how to implement it in an efficient way, step by step. This algorithm remains notably fast compared to the original one. For fixed parameters, text recognition rates and binarization quality are equal or better than other methods on text with low and medium x-height and are significantly improved on text with large x-height. Pixel-based accuracy and OCR evaluations are performed on more than 120 documents. Compared to awarded methods in the latest binarization contests, Sauvola's formula does not give the best results on historical documents. On the other hand, on clean magazines, it outperforms those methods. This implementation improves the robustness of Sauvola's algorithm by making the results almost insensible to the window size whatever the object sizes. Its properties make it usable in full document analysis toolchains.

/pdf/efficient-multiscale-sauvola-s-binarization-2hx7mss9kn.pdf

Efficient multiscale Sauvola's binarization

Optimal combination of document binarization techniques using a self-organizing map neural network

Most of the document binarization techniques have many parameters that can initially be specified. Usually, subjective document binarization evaluation, employs human observes for the estimation of the best parameter values of the techniques. Thus, the selection of the best values for these parameters is crucial for the final binarization result. However, there is not any set of parameters that guarantees the best binarization result for all document images. It is important, the estimation of the best values to be adaptive for each one of the processing images. This paper proposes a new method which permits the estimation of the best parameter values for each one of the document binarization techniques and also the estimation of the best document binarization result of all techniques. In this way, document binarization techniques can be compared and evaluated using, for each one of them, the best parameter values for every document image.

/pdf/automatic-evaluation-of-document-binarization-results-1ocoxqp3iz.pdf

Automatic evaluation of document binarization results

This paper presents a system for binarization of digital documents. The system comprises the benefits of a set of other binarization techniques by combining their results. This is necessary for bad illuminated and degraded document where there are many pixels that cannot be easily classified as foreground or background. For this reason, it is necessary to perform the final binarization by exploiting the results of a set of binarization algorithms, especially for the document pixels that have high vagueness. Also, in this paper significant improvements are proposed for two of the methods used, i.e. for the Adaptive Logical Level Technique (ALLT) and the Improvement of Integrated Function Algorithm (IIFA). The entire system is extensively tested with a variety of degraded and bad-illuminated documents.

http://ieeexplore.ieee.org/iel5/9084/28824/01296408.pdf

E. Badekas

Papers

Automatic evaluation of document binarization results

A system for document binarization