Ergina Kavallieratou, Laurence Likforman-Sulem · 2018
The Special Issue "Document Image Processing" in the Journal of Imaging aims at presenting approaches which contribute to access the content of document images. These approaches are related to low level tasks such as image preprocessing, skew/slant corrections, binarization and document segmentation, as well as high level tasks such as OCR, handwriting recognition, word spotting or script identification. This special issue brings together 12 papers that discuss such approaches. The first three articles deal with historical document preprocessing. The work by Hanif et al. [1] aims at removing bleed-through using a non-linear model, and at reconstructing the background by an inpainting approach based on non-local patch similarity. The paper by Almeida et al. [2] proposes a new binarization approach that includes a decision-based process for finding the best threshold for each RGB channel. In the paper by Kavallieratou et al. [3], a segmentation-free approach based on the Wigner-Ville distribution is used to detect the slant of a document and correct it.
Cited by 3Open access
Kuan Rong Chan, D.B.H. Tay, Farook Sattar · DR-NTU (Nanyang Technological University) · 2008
This project aims to develop image processing techniques to enhance the readiblity of document images.
Cited by 0
Qian Guo · DR-NTU (Nanyang Technological University) · 2009
In an office environment, a large amount of information is manipulated in the form of business documents. Currently, two types of documents are commonly used for business communication. They are paper documents (for example letters, memos, reports, invoices, etc.) and electronic documents (e.g. electronic mail, fax messages, etc.). Although documents can be transmitted in electronic form, paper documents are still the favorite medium for business communication in the office environment. With the advent of computer technologies, the concept of the "paperless office" evolved and information can be processed and transmitted via computers. To achieve this, it is necessary to handle both the electronic and paper documents in an efficient and integrated way.
Cited by 0
Barni, Mauro · 2006
Cited by 0
Jerry P. Skelton, Anthony P. Cavallo, Julie Peternick · Proceedings of SPIE, the International Society for Optical Engineering/Proceedings of SPIE · 1990
The nineteen nineties will usher in a new era in image processing systems. The maturation of the technologies required to capture, store, process, retrieve, display and distribute documents electronically will enable government and industry to implement large scale, highly productive and cost effective, document image processing (DIP) systems. This new information processing technology will greatly enhance our ability to deal with the ever increasing volume of paper documents that have become the hallmark of our knowledge based societies. This paper briefly discusses the forces driving the evolution of DIP systems and the DIP marketplace. Technical considerations and technology trends, involved in determining a realistic and successful approach to implementing a DIP system, are also presented.
Cited by 0
CERN Document Server (European Organization for Nuclear Research) · 1989
Cited by 0
V. Jiroušek · Digital Repository (National Repository of Grey Literature) · 2024
The aim of the methodology is to offer libraries and other memory institutions in the Czech Republic recommended procedures for facilitating access to graphic and other non-textual objects contained on the pages of digitised documents, and for facilitating the public's orientation to these non-textual elements and the possibilities of their further use. The sub-objectives leading to easier traceability are in particular: 1) automatic detection of non-textual elements in documents contained in digital libraries, 2) their type categorization and supplementation with necessary contextual and other descriptive metadata using machine learning methods, and 3) writing the obtained information through international metadata standards into metadata in accordance with the national NDK Standard.
Cited by 0Open access
2014
Cited by 0
Frank L. Walker, George R. Thoma · 2000
Increasing numbers of research libraries are turning to the Internet for electronic interlibrary loan and for document delivery to patrons. This has been made possible through the widespread adoption of software such as Ariel and DocView. Ariel, a product of the Research Libraries Group, converts paper-based documents to monochrome bitmapped images, and delivers them over the Internet. The National Library of Medicine’s DocView is primarily designed for library patrons to receive, display and manage documents received from Ariel systems. While libraries and their patrons are beginning to reap the benefits of this new technology, barriers exist, e.g., differences in image file format, that lead to difficulties in the use of library document information. To research how to overcome such barriers, the Communications Engineering Branch of the Lister Hill National Center for Biomedical Communications, an R&D division of NLM, has developed a web site called the DocMorph Server. This is part of an ongoing intramural R&D program in document imaging that has spanned many aspects of electronic document conversion and preservation, Internet document transmission and document usage. The DocMorph Server web site is designed to fill two roles. First, in a role that will benefit both libraries and their patrons, it allows Internet users to upload scanned image files for conversion to alternative formats, thereby enabling wider delivery and easier usage of library document information. Second, the DocMorph Server provides the design team an active test bed for evaluating the effectiveness and utility of new document image processing algorithms and functions, so that they may be evaluated for possible inclusion in other image processing software products being developed at NLM or elsewhere. This paper describes the design of the prototype DocMorph Server and the image processing functions being implemented on it.
Cited by 0
John E. Graves, Jacqueline Justice, Marge Cataldo · 1998
Cited by 1
Prepared Josh King · 2004
Many problems can arise when scanning an image containing text into a computer. Examples of these problems, taken from the course website are shown below: • The edge of a book binding may appear along one side* • The text may be cut off on one side
Cited by 0
Andrew P. Cooper, W. Kahari, R. Such · IEE Proceedings E Computers and Digital Techniques · 1988
Reos is a document image processing system which stores images of paper documents as a bit map on optical discs. These images are displayed on a CRT at a lower resolution compared to the stored data. The conversion and display of the stored information must be performed quickly to achieve an acceptable system response time; since even highly optimised software could not process more than a few thousand pixels per second, a hardware solution capable of processing many millions of pixels per second was developed. The mesh offset algorithm which combines image resampling with convolution, was selected for its performance and ease of mapping onto silicon. The hardware of this algorithm option is suitable for more general purpose image processing tasks such as filtering to reduce noise, or contrast enhancement. A high level description of the system has been mapped onto silicon using a number of design options to provide the best balance between speed of design, performance and die area. The resolution converter in 2.5 μm CMOS double layer metal technology supports scan path test and signature analysis for production test purposes.
Cited by 0
Lin Yi · Scholarship at UWindsor (University of Windsor) · 2003
Image thresholding and page segmentation are necessary components of any image understanding and recognition system. In order for an OCR to function properly, texts in a document image has to be isolated and then fed to the OCR for recognition. This requires development of a robust and accurate page segmentation technique. In any page segmentation technique, a preprocessing step in terms of image restoration and thresholding is needed. This thesis therefore concentrates on the development of efficient and robust image thresholding and page segmentation algorithms. In this thesis, three efficient contrast enhancement techniques are proposed that in conjunction with the thresholding techniques of Ridler and Calvard constitute the preprocessing step for the image segmentation algorithm. This thesis also provides a survey of the pertinent page segmentation techniques in the literature, and proposes a new block labeling technique based on smearing algorithm. An exhaustive experimentation is conducted in this thesis to demonstrate the efficiency of the proposed techniques.Dept. of Electrical and Computer Engineering. Paper copy at Leddy Library: Theses & Major Papers - Basement, West Bldg. / Call Number: Thesis2003 .L56. Source: Masters Abstracts International, Volume: 42-03, page: 1015. Thesis (M.A.Sc.)--University of Windsor (Canada), 2003.
Cited by 0Open access
Donald L. Kohn · PubMed · 1992
Cited by 0
Charles Dickens · 1991
Cited by 0
Robert E. Wiggins · International Journal of Information Management · 1990
Cited by 2
Mr.Shivraj Badiger, Dr.Raju Hiremath · JOURNAL OF ADVANCE AND FUTURE RESEARCH · 2025
This study presents the development of a Document Forgery Detection System using a Flask-based web application integrated with the Roboflow API and a YOLO-based detection model. The primary objective of this system is to accurately identify and classify forged regions within document images through automated analysis. The model leverages the Roboflow cloud inference API for prediction and retrieves bounding box data representing potential forgeries(seal,stamp,signature) while the Flask backend ensures seamless communication between the client interface and the detection engine. In addition, the application includes a performance metrics module that reads model training results from CSV outputs and visualizes key evaluation indicators such as Precision, Recall, F1-Score, and mAP (Mean Average Precision). The system further facilitates visualization of evaluation plots, including the confusion matrix and precision-recall curves, providing comprehensive insights into model behavior. The implementation employs Python Flask for RESTful API construction, CSV parsing for metrics extraction, and CORS configuration for broad client compatibility. This architecture establishes a robust and modular pipeline for real-time document integrity verification, with potential applications in legal, financial, and administrative domains where document authenticity is critical
Cited by 0Open access
Singh, Prashant, Vats, Ekta, Hast, Anders · arXiv (Cornell University) · 2017
Computation of document image quality metrics often depends upon the availability of a ground truth image corresponding to the document. This limits the applicability of quality metrics in applications such as hyperparameter optimization of image processing algorithms that operate on-the-fly on unseen documents. This work proposes the use of surrogate models to learn the behavior of a given document quality metric on existing datasets where ground truth images are available. The trained surrogate model can later be used to predict the metric value on previously unseen document images without requiring access to ground truth images. The surrogate model is empirically evaluated on the Document Image Binarization Competition (DIBCO) and the Handwritten Document Image Binarization Competition (H-DIBCO) datasets.
Cited by 0Open access
Itaru Nagayama · IEEJ Transactions on Industry Applications · 2008
This paper describes a new approach for high-performance image searching method named interlaced fast sequential matching and its application to document image processing for OA system. Conventional image retrieval method for a huge image such as Correlation Matching and SSDA have a shortcoming with their performance. Thus, we introduce the interlaced fast sequential matching in order to avoid the difficulty in the conventional methods. Generally speaking, image retrieval in many kind of documents continues to be one of the most common application for document prcessing. Therefore, if we can find an efficient method to retrieve any character objects without OCR process, the more efficient document management can be performed. We first discuss some essential issues to be considered in document processing with large document image. Experimental results by using the proposed method are also described. The good performance of the interlaced fast sequential matching approach is shown.
Cited by 2Open access
LOO POH KOK · National University of Singapore · 2005
Ph.D
Cited by 0