Performance Comparison of Six Algorithms for Page Segmentation

Shafait, Faisal; Keysers, Daniel; Breuel, Thomas M.

doi:10.1007/11669487_33

Cited by 52 publications

(36 citation statements)

References 17 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…[10] and Gorman [15] are the most widely cited algorithms to perform geometric page segmentation on non-Manhattan layouts of non overlapping zones. Evaluation experiments [17] have shown that the Voronoi based approach excels on a mixed dataset of both handwritten and machine printed documents for diverse scripts such as English and Arabic. Nevertheless, there are still challenges which the technique faces with such datasets.…”

Section: Non-manhattan Layout Basedmentioning

confidence: 99%

“…Extra edges are added on the boundaries, as voronoi segmentation does not produce those. [17,12] measure overall error as the percentage of ground-truth text-lines correctly contained within zones. A ground-truth text-line is said to lie completely within one detected zone if the area overlap between the two is significant.…”

Section: Polygonal Zonesmentioning

confidence: 99%

“…Page segmentation algorithms fall into the category of physical layout analysis. They perform segmentation of a document page into homogeneous zones, each consisting of only one physical layout structure [17]. Layout analysis can be pixel based or texture based segmentation [8], but we assume the final result is a region segmentation.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Voronoi++: A Dynamic Page Segmentation Approach Based on Voronoi and Docstrum Features

Agrawal

Doermann

2009

2009 10th International Conference on Document Analysis and Recognition

View full text Add to dashboard Cite

This paper presents a dynamic approach to document page segmentation. Current page segmentation algorithms lack the ability to dynamically adapt local variations in the size, orientation and distance of components within a page. Our approach builds upon one of the best algorithms, Kise et. al. work based on Area Voronoi Diagrams [10], which adapts globally to page content to determine algorithm parameters. In our approach, local thresholds are determined dynamically based on parabolic relations between components, and Docstrum based angular and neighborhood features are integrated to improve accuracy. Zone-based evaluation was performed on four sets of printed and handwritten documents in English and Arabic scripts and an increase of 33% in accuracy is reported.

show abstract

Section: Non-manhattan Layout Basedmentioning

confidence: 99%

Section: Polygonal Zonesmentioning

confidence: 99%

See 1 more Smart Citation

Voronoi++: A Dynamic Page Segmentation Approach Based on Voronoi and Docstrum Features

Agrawal

Doermann

2009

2009 10th International Conference on Document Analysis and Recognition

View full text Add to dashboard Cite

show abstract

“…We use a fast labeling algorithm to extract connected components from the document image. In recent comparative studies for the performance evaluation of page segmentation algorithms [1,10], it is shown that the constrained textline finding algorithm [3] has the lowest error rates among the compared algorithms for textline extraction, and the Voronoi-diagram based algorithm [5] has the lowest error rates for extracting zone-level information. Therefore, we use these two algorithms for extracting textlines and zones from the document image, respectively.…”

Section: Geometric Matching For Page Frame Detectionmentioning

confidence: 99%

Page Frame Detection for Marginal Noise Removal from Scanned Documents

2007

Self Cite

View full text Add to dashboard Cite

Abstract.We describe and evaluate a method to robustly detect the page frame in document images, locating the actual page contents area and removing textual and non-textual noise along the page borders. We use a geometric matching algorithm to find the optimal page frame, which has the advantages of not assuming the existence of whitespace between noisy borders and actual page contents, and of giving a practical solution to the page frame detection problem without the need for parameter tuning. We define suitable performance measures and evaluate the algorithm on the UW-III database. The results show that the error rates are below 4% for each of the performance measures used. In addition, we demonstrate that the use of page frame detection reduces the optical character recognition (OCR) error rate by removing textual noise. Experiments using a commercial OCR system show that the error rate due to elements outside the page frame is reduced from 4.3% to 1.7% on the UW-III dataset.

show abstract

“…The whole image is binarized using a single threshold value in global techniques (Otsu, 1979;White and Rohrer, 1983;Chi et al, 1996;Cattoni et al, 1998;Viola and Jones, 2004;Shafait et al, 2006). Grey value of each pixel is compared with the single threshold value for binarization.…”

Section: Introductionmentioning

confidence: 99%

Image Binarization Using Block Boundary Pixels Mean

Singh¹,

Singh²

2017

Journal of Computer Science

View full text Add to dashboard Cite

Input image pixels classification into two intensities like black and white as foreground and background respectively refers to image binarization. Binarization with a local threshold value is termed as adaptive image binarization. Locally adaptive thresholding techniques are local pixels dependent process. Local environment dependent process is normally time consuming one and hence its computational time complexity is also local region dependent. Adaptation depends on the local region contrast condition. Low contrast region may not be adapted correctly. This article presents a new locally adaptive fast binarization technique applied on contrast stretched domain of an input image. The local adaptation is based on the mean of the 9 local block boundary pixels. As it associates only 9 pixels for mean calculation, its computational time complexity is free from local reason block size. As it applies on contrast stretched domain image, it can adapt low contrast region for binarization while other techniques fail. Not only its low contrast adapting capability, it has a local block size free computational time complexity. Hence it is more convenient to adapt the low contrast region very fast as compared with other techniques. From the experimental results, it is observed that it yields fast better result than other related local techniques.

show abstract

Performance Comparison of Six Algorithms for Page Segmentation

Cited by 52 publications

References 17 publications

Voronoi++: A Dynamic Page Segmentation Approach Based on Voronoi and Docstrum Features

Voronoi++: A Dynamic Page Segmentation Approach Based on Voronoi and Docstrum Features

Page Frame Detection for Marginal Noise Removal from Scanned Documents

Image Binarization Using Block Boundary Pixels Mean

Contact Info

Product

Resources

About