Performance Evaluation of Document Structure Extraction Algorithms

Liang, Jisheng; Phillips, Ihsin T.; Haralick, Robert M.

doi:10.1006/cviu.2001.0933

Cited by 20 publications

(13 citation statements)

References 11 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Layout analysis attempts to extract this hierarchical structure from an image representation [7,8]. To segment a document at a specified level, previous work on machine printed document analysis relies on the homogeneity inside a region and a significant gap between neighboring regions [9]. Two methodologies, bottom-up or top-down, were used frequently in previous work for layout analysis [10].…”

Section: Layout Hierarchy In Document Imagementioning

confidence: 99%

Script-Independent Text Line Segmentation in Freestyle Handwritten Documents

Zheng

Doermann

et al. 2008

IEEE Trans. Pattern Anal. Mach. Intell.

172

View full text Add to dashboard Cite

Text line segmentation in freestyle handwritten documents remains an open document analysis problem. Curvilinear text lines and small gaps between neighboring text lines present a challenge to algorithms developed for machine printed or hand-printed documents. In this paper, we propose a novel approach based on density estimation and a state-of-the-art image segmentation technique, the level set method. From an input document image, we estimate a probability map, where each element represents the probability that the underlying pixel belongs to a text line. The level set method is then exploited to determine the boundary of neighboring text lines by evolving an initial estimate. Unlike most connected component based methods [1,2], the proposed algorithm does not use any script-specific knowledge. Extensive quantitative experiments on freestyle handwritten documents with diverse scripts, such as Arabic, Chinese, Korean, and Hindi, demonstrate that our algorithm consistently outperforms previous methods [3,1,2]. Further experiments show the proposed algorithm is robust to scale change, rotation, and noise. Keywords: Handwritten Text Line Segmentation, Document Image Analysis, Density Estimation, Level Set MethodsThe support of this research by the Department of Defense under contract MDA-9040-2C-0406 is gratefully acknowledged. 1 Report Documentation PageForm Public reporting burden for the collection of information is estimated to average 1 hour per response, including the time for reviewing instructions, searching existing data sources, gathering and maintaining the data needed, and completing and reviewing the collection of information. Send comments regarding this burden estimate or any other aspect of this collection of information, including suggestions for reducing this burden, to Washington Headquarters Services, Directorate for Information Operations and Reports, 1215 Jefferson Davis Highway, Suite 1204, Arlington VA 22202-4302. Respondents should be aware that notwithstanding any other provision of law, no person shall be subject to a penalty for failing to comply with a collection of information if it does not display a currently valid OMB control number.

show abstract

Section: Layout Hierarchy In Document Imagementioning

confidence: 99%

Script-Independent Text Line Segmentation in Freestyle Handwritten Documents

Zheng

Doermann

et al. 2008

IEEE Trans. Pattern Anal. Mach. Intell.

172

View full text Add to dashboard Cite

show abstract

“…Discussion over the need for and role of rigorous performance evaluation of vision algorithms was raised as a specific question within the academic community from 1986 [1,2]; again in the early 1990s [3]; continued by Haralick with support from DARPA and elsewhere [4][5][6][7] and by Forstner [8]; and was taken up within certain sub-communities, in particular OCR [9][10][11], document understanding [12,13], graphics recognition [14][15][16][17][18][19][20], and photogrammetry [21,22].…”

Section: Background and Motivationmentioning

confidence: 99%

“…For some tasks, such as face localization [94] and graphics recognition [13,19], researchers use existing data sets. For other tasks, data sets tend to be created specifically [79,85,86].…”

Section: Are There Data Sets In Common Use?mentioning

confidence: 99%

“…Similarly, the UCID subset has been proposed. 12 The COIL-100 database of labeled everyday objects from Columbia University is an alternative, as is the data set from Washington University, 13 and the SOIL-47 color image data set from Surrey. However, published works on these data sets by groups other than their originators could not be found.…”

Section: Is There a Data Set For Which The Correct Answers Are Known?mentioning

confidence: 99%

See 1 more Smart Citation

Performance characterization in computer vision: A guide to best practices

Thacker

Clark

Barron

et al. 2008

Computer Vision and Image Understanding

View full text Add to dashboard Cite

“…Yanikoglu et al [3] presented a region-based page segmentation benchmarking environment, named Pink Panther. Liang et al [4] proposed a performance metric for document structure extraction algorithms by finding the correspondences between detected entities and ground-truth. The quality of page segmentation algorithms was also evaluated by analyzing the errors in the recognized text [5].…”

Section: Introductionmentioning

confidence: 99%

Performance Comparison of Six Algorithms for Page Segmentation

Shafait

Keysers

Breuel

2006

Document Analysis Systems VII

View full text Add to dashboard Cite

Abstract. This paper presents a quantitative comparison of six algorithms for page segmentation: X-Y cut, smearing, whitespace analysis, constrained text-line finding, Docstrum, and Voronoi-diagram-based. The evaluation is performed using a subset of the UW-III collection commonly used for evaluation, with a separate training set for parameter optimization. We compare the results using both default parameters and optimized parameters. In the course of the evaluation, the strengths and weaknesses of each algorithm are analyzed, and it is shown that no single algorithm outperforms all other algorithms. However, we observe that the three best-performing algorithms are those based on constrained text-line finding, Docstrum, and the Voronoi-diagram.

show abstract

Performance Evaluation of Document Structure Extraction Algorithms

Cited by 20 publications

References 11 publications

Script-Independent Text Line Segmentation in Freestyle Handwritten Documents

Script-Independent Text Line Segmentation in Freestyle Handwritten Documents

Performance characterization in computer vision: A guide to best practices

Performance Comparison of Six Algorithms for Page Segmentation

Contact Info

Product

Resources

About