Academia.eduAcademia.edu

Outline

Title
Abstract
FAQs
Figures
Introduction
Results
Conclusion
References
All Topics
Business and Management
Other Business

Newspaper Page Decomposition Using a Split and Merge Approach

Profile image of Karim HadjarKarim Hadjar

2001

https://doi.org/10.1109/ICDAR.2001.953972
visibility

…

description

4 pages

link

1 file

Abstract

Indexing large newspaper archives requires automatic page decomposition algorithms with high accuracy. In this paper, we present our approach to an automatic page decomposition algorithm developed for the First International Newspaper Segmentation Contest. Our approach decomposes the newspaper image into image regions, horizontal and vertical lines, text regions and title areas. Experimental results are obtained from the data set of the contest

FAQs

sparkles

AI

What novel approach does the paper suggest for newspaper page decomposition?add

The paper presents a split and merge technique that segments newspaper pages using extracted vertical and horizontal lines, departing from traditional top-down methods like Nagy's X-Y cut algorithm.

How effective is the proposed method on historical newspaper pages?add

Tested on 20 scanned front pages from the First International Newspaper Segmentation Contest, the method shows promising results despite limitations in handling complex layouts and image noise.

What preprocessing methods are utilized to enhance newspaper image segmentation?add

Only basic preprocessing occurs, primarily filtering small connected components; however, the algorithm does not leverage advanced techniques for image enhancement.

What limitations hinder the accuracy of the segmentation process?add

Key limitations include linear decision-making in the algorithms and inadequate consideration of background lines, leading to potential misclassification of components.

Which future enhancements are recommended for improving the system's performance?add

Future improvements could focus on refining the merge operation, enhancing title detection within text blocks, and integrating font recognition to optimize results further.

Figures (6)
During the split phase, the whole image newspaper is split according to the extracted vertical and horizontal lines as well as the boundaries of recognized images. This results in many small zones as illustrated in figure 3. Figure 3. Splitting the newspaper image into small zones.
During the split phase, the whole image newspaper is split according to the extracted vertical and horizontal lines as well as the boundaries of recognized images. This results in many small zones as illustrated in figure 3. Figure 3. Splitting the newspaper image into small zones.
Figure 1. In a text region, there are almost no overlapping bounding boxes of connected components. 2.4 Split
Figure 1. In a text region, there are almost no overlapping bounding boxes of connected components. 2.4 Split
Figure 5 shows a small part of a page that has been cor- rectly segmented. The results we have obtained show that our approach is feasible and give many hints for improve- ment. The most important limitation is that the whole ap- proach is a linear succession of algorithms. Once a decision is made by an algorithm, this decision cannot be rejected by a later algorithm if it is wrong. Figure 5. A part of a correctly segmented page.
Figure 5 shows a small part of a page that has been cor- rectly segmented. The results we have obtained show that our approach is feasible and give many hints for improve- ment. The most important limitation is that the whole ap- proach is a linear succession of algorithms. Once a decision is made by an algorithm, this decision cannot be rejected by a later algorithm if it is wrong. Figure 5. A part of a correctly segmented page.
Figure 4, The zone pairs (1,2) and (3,4) are mergeable, whereas the pairs (1,3), (2,4) and (5,6) are separated. All other pairs are unknown. merge operation will not try to merge them. If, on the other hand, there are black pixels between the two zones, they will be marked mergeable. All other pairs of zones are marked unknown. This procedure is illustrated in figure 4.
Figure 4, The zone pairs (1,2) and (3,4) are mergeable, whereas the pairs (1,3), (2,4) and (5,6) are separated. All other pairs are unknown. merge operation will not try to merge them. If, on the other hand, there are black pixels between the two zones, they will be marked mergeable. All other pairs of zones are marked unknown. This procedure is illustrated in figure 4.
Figure 7. Over-segmentation due to our sim- plistic merge operation. Figure 6. The merge operation does not rec- ognize that a bigger rectangle can be formed by not merging the bottom small rectangle of the middle column.
Figure 7. Over-segmentation due to our sim- plistic merge operation. Figure 6. The merge operation does not rec- ognize that a bigger rectangle can be formed by not merging the bottom small rectangle of the middle column.
The merge operation should be smarter. The choice of merging first vertically and then horizontally neighboring zones has considerable impact on the result. Consider fig- ure 6 where a smarter algorithm would recognize that split- ting off the bottom small rectangle in the middle column would allow to merge the three columns to form a bigger rectangle. In fact, the merge operation is an optimization problem with the goal of finding the optimal disposition of rectangular zones. Over-segmented results such as the one shown in figure 7 are due to this problem.
The merge operation should be smarter. The choice of merging first vertically and then horizontally neighboring zones has considerable impact on the result. Consider fig- ure 6 where a smarter algorithm would recognize that split- ting off the bottom small rectangle in the middle column would allow to merge the three columns to form a bigger rectangle. In fact, the merge operation is an optimization problem with the goal of finding the optimal disposition of rectangular zones. Over-segmented results such as the one shown in figure 7 are due to this problem.

Related papers

Automatic Article Extraction in Old Newspapers Digitized Collections
Pierrick Tranouez

We present a complete method for article segmentation in old newspapers, which deals with complex layouts analysis of degraded documents. The designed workflow can process large amounts of documents and generates digital objects in METS/ALTO format in order to facilitate the indexing and the browsing of information in digital libraries. The analysis of the document image is performed by a two stages scheme. Pixels are labelled in a first stage with a Conditional Random Field model in order to intent to label the areas of interest with a low logical level. Then this first logical representation of the document content is analysed in a second stage to get a higher logical representation including article segmentation and reading order.

View PDFchevron_right
Performance comparison of six algorithms for page segmentation
Thomas Breuel

Document Analysis Systems VII, 2006

View PDFchevron_right
Automated Processing of Digitized Historical Newspapers beyond the Article Level: Sections and Regular Features
Robert B Allen

Millions of pages of historical newspapers have been digitized but in most cases access to these are supported by only basic search services. We are exploring interactive services for these collections which would be useful for supporting access, including automatic categorization of articles. Such categori-zation is difficult because of the uneven quality of the OCR text, but there are many clues which can be useful for improving the accuracy of the categoriza-tion. Here, we describe observations of several historical newspapers to determine the characteristics of sections. We then explore how to automatically identify those sections and how to detect serialized feature articles which are repeated across days and weeks. The goal is not the introduction of new algorithms but the development of practical and robust techniques. For both analyses we find substantial success for some categories and articles, but others prove very difficult.

View PDFchevron_right
A multiresolution approach for page segmentation
luca lombardi, Gianmarco Manzini

Pattern recognition letters, 1998

In this work we propose a new page segmentation method for recognizing text and graphics based on a multiresolution representation of the page image. Our approach is based on the analysis of a set of feature maps available at different resolution levels. The final output is a description of the physical structure of a page. A page image is broken down into several blocks which represent components of a page, such as text, line-drawings, and pictures. The result, which uses only a small amount of memory in addition to that for the image, may be the first step for a more detailed analysis such as optical character recognition. q

View PDFchevron_right
Combining Visual and Textual Features for Semantic Segmentation of Historical Newspapers
Simon Clematide

Journal of Data Mining & Digital Humanities, 2021

The massive amounts of digitized historical documents acquired over the last decades naturally lend themselves to automatic processing and exploration. Research work seeking to automatically process facsimiles and extract information thereby are multiplying with, as a first essential step, document layout analysis. If the identification and categorization of segments of interest in document images have seen significant progress over the last years thanks to deep learning techniques, many challenges remain with, among others, the use of finer-grained segmentation typologies and the consideration of complex, heterogeneous documents such as historical newspapers. Besides, most approaches consider visual features only, ignoring textual signal. In this context, we introduce a multimodal approach for the semantic segmentation of historical newspapers that combines visual and textual features. Based on a series of experiments on diachronic Swiss and Luxembourgish newspapers, we investigate...

View PDFchevron_right
Performance evaluation and benchmarking of six-page segmentation algorithms
Thomas Breuel

IEEE Transactions on Pattern …, 2007

View PDFchevron_right
Evaluating SEE: a benchmarking system for document page segmentation
Stefan Agne

Seventh International Conference on Document Analysis and Recognition, 2003. Proceedings., 2003

The decomposition of a document into segments such as text regions and graphics is a significant part of the document analysis process. The basic requirement for rating and improvement of page segmentation algorithms is systematic evaluation. The approaches known from the literature have the disadvantage that manually generated reference data (zoning ground truth) are needed for the evaluation task. The effort and cost of the creation of these data are very high. This paper describes the evaluation system SEE and presents an assessment of its quality.. The system requires the OCR generated text and the original text of the document in correct reading order (text ground truth) as input. No manually generated zoning ground truth is needed. The implicit structure information that is contained in the text ground truth is used for the evaluation of the automatic zoning. Therefore, an assignment of the corresponding text regions in the text ground truth and those in the OCR generated text (matches) is sought. A fault tolerant string matching algorithm underlies a method, able to tolerate OCR errors in the text. The segmentation errors are determined as a result of the evaluation of the matching. Subsequently, the edit operations which are necessary for the correction of the recognized segmentation errors are computed to estimate the correction costs. Furthermore, SEE provides a version of the OCR generated text, that is corrected from the detected page segmentation errors.

View PDFchevron_right
Information Extraction and Search in Newspaper Front Pages
Tiago Varela

This study analyses existing newspaper archives and reviews the state of the art regarding page segmentation methods, text recognition in documents and web search interfaces. We also detail a new keyword extraction method that specifically targets current newspaper front pages. This method splits the newspaper front page image into rectangular regions and applies a background/foreground segmentation method that targets the contained text. Using a test bed designed specifically for this study, this method obtained an 8.62 percentage point increase in accuracy over the default keyword extraction method. We also developed a prototype content-aware newspaper archive, aptly named Gazeta, which allows the users to search for keywords and to view the front pages containing those particular keywords. It also graphically displays the number occurrences for the specified keywords, over a particular timespan.

View PDFchevron_right
Pixel-accurate representation and evaluation of page segmentation in document images
Thomas Breuel

Pattern Recognition, 2006. …, 2006

View PDFchevron_right
Automated Processing of Digitized Historical Newspapers: Identification of Segments and Genres
Robert B Allen

Lecture Notes in Computer Science, 2008

Many historical newspapers are being digitized. We aim to support access to them via text analysis of the OCRd content. However, the OCR includes many errors; so extracting meaningful content from it is difficult. A pipeline of processing steps is proposed. Here, we describe the first two steps: segmentation and genre identification. The segmentation procedure based on headings was quite successful. Genre identification worked well for easily defined genre categories such as weather reports. We also propose additional techniques which may improve the accuracy still farther.

View PDFchevron_right

References (10)

  1. F. Bapst, R. Brugger, and R. Ingold. Towards an Interactive Document Recognition System. Internal working paper 95- 09, IIUF-Université de Fribourg, March 1995.
  2. B. Gatos and A. Antonacopoulos. First International News- paper Segmentation Contest. http://www.lpa.gr/ contest/, 2001.
  3. B. Gatos, S. L. Mantzaris, K. V. Chandrios, A. Tsigris, and S. J. Perantonis. Integrated algorithms for newspaper page decomposition and article tracking. In ICDAR'99: Fifth Inter- national Conference on Document Analysis and Recogntion, pages 559-562, Bangalore, India, Sept. 1999.
  4. V. Govindaraju, S. W. Lam, D. Niyogi, D. B. Sher, R. Srihari, S. N. Srihari, and D. Wang. Newspaper image understand- ing.
  5. In S. Ramani, R. Chandrasekar, and K. S. R. Anjaneyulu, editors, Knowledge Based Computer Systems, pages 375-84. Narosa Publishing House New Delhi, India, 1990.
  6. O. Hitz, L. Robadey, and R. Ingold. Analysis of Synthetic Document Images. In ICDAR'99: Fifth International Confer- ence on Document Analysis and Recogntion, pages 555-558, Bangalore, India, Sept. 1999.
  7. G. Nagy, S. Seth, and M. Viswanathan. A prototype docu- ment image analysis system for technical journals. Computer, 25(7):10-22, July 1992.
  8. L. Robadey, O. Hitz, and R. Ingold. Segmentation de docu- ments ideaux structure complexe. In CIFED'2000: Colloque International Francophone sur l'Ecrit et le Document, pages 383-392, Lyon, France, jul 2000.
  9. D. Wang and S. N. Srihari. Classification of newspaper image blocks using texture analysis. Computer Vision, Graphics, and Image Processing, 47(3):327-352, Sept. 1989.
  10. A. Zramdini. Study of Optical Font Recognition Based on Global Typographical Features. PhD thesis, University of Fribourg, 1995.

Related papers

Integrated Algorithms for Newspaper Page Decomposition and Atricle Tracking
Ανδρέας Τσίγκρης

The conversion of newspaper pages into digital resources is an important task that greatly contributes to the preservation and access to newspaper archives.In this paper, an integrated methodology is presented for segmenting newspaper page and identifying newspaper articles. In a first stage, a succession of image processing and document analysis algorithms is employd for segmenting newspaper page images into various objects (text, images and drawings, titles). A rule based approach is subsequently applied to the objects identified during the page segmentation phase for reconstructing individual articles. Experimental results, obtained from a large testbed of old newspaper issues, are presented which clearly demonstrate the applicability of our integrated approach to successful newspaper page segmentation an identification of newspaper articles

View PDFchevron_right
Segmentation of Unstructured Newspaper Documents
IJAERS Journal

— Document layout analysis is one of the important steps in automated document recognition systems. In Document layout analysis, meaningful information is retrieved from document images by identifying, categorizing and labeling the semantics of text blocks from the document images. In this paper, we present simple top-down approach for document page segmentation. We have tested the proposed method on unstructured documents like newspaper which is having complex structures having no fixed structure. Newspaper also has multiple titles and multiple columns. In the proposed method, white gap area which separates titles, columns of text, line of text and words in lines have been identified to separate document into various segments. The proposed algorithm has been successfully implemented and applied over a large number of Indian newspapers and the results have been evaluated by number of blocks detected and taking their correct ordering information into account. Keywords— Document Layout Analysis, data extraction, document page segmentation, unstructured document.

View PDFchevron_right
Automatic Page Analysis for the Creation of a Digital Library from Newspaper Archives
Ανδρέας Τσίγκρης

Digital preservation of newspaper archives aims both at the salvation of endangered material (paper) and at the creation of digital library services that will allow full utilization of the archives by all interested parties. In this paper, we address a series of issues pertaining to the retro-conversion of newspapers, i.e., the conversion of newspaper pages into digital resources. An integrated approach is presented that provides solutions to problems related to newspaper page image enhancement, segmentation of pages into various items (titles, text, images etc), article identification and reconstruction, and, finally, recognition of the textual components. Emphasis is placed on the most difficult intermediate stages of page segmentation and article identification and reconstruction. Detailed experimental results, obtained from a large testbed of old newspaper issues, are presented which clearly demonstrate the applicability of our methodology to the successful retro-conversion of newspaper material.

View PDFchevron_right
Arabic Newspaper Page Segmentation
Karim Hadjar

2003

The aim of layout analysis is to extract the geometric structure from a document image. It consists of labeling homogenous regions of a document image. This paper describes the performance of segmentation algorithms and their adaptation in order to treat complex structured Arabic documents such as newspapers. Experimental tests have been carried out on four different phases of newspaper image analysis: thread recognition, frame recognition, image text separation, text line recognition, and line merging into blocks. Some promising experimental results are reported.

View PDFchevron_right
Document page decomposition by the bounding-box project
Robert Haralick

Proceedings of 3rd International Conference on Document Analysis and Recognition, 1995

This paper describes a method for extracting words, textlines and text blocks by analyzing the spatial configuration of bounding boxes of connected components on a given document image. The basic idea is that connected components of black pixels can be used as computational units in document image analysis. In this paper, the problem of extracting words, textlines and text blocks is viewed as a clustering problem in the &dimensional discrete domain. Our main strategy is that profiling analysis is utilized to measure horizontal or vertical gaps of (groups of) components during the process of image segmentation. For this purpose, we compute the smallest rectangular box, called the bounding box, which circumscribes a connected component. Those boxes are projected horizontally and/or vertically, and local and global projection profiles are analyzed for word, textline and text-block segmentation. In the last step of segmentation, the document decomposition hierarchy is produced from these segmented objects.

View PDFchevron_right

Related topics

  • Publishing
    • 580 California St., Suite 400
    San Francisco, CA, 94104
    © 2025 Academia. All rights reserved