a Record control number 000259670300086

a ISBN 978--0--7695-3122--9

a Original cataloging agency NEU GrandLibrary

a Language code of text/sound track or separate title eng

a Classification number QA76

b Item number .G56 2008

a Personal name Khashman, Adnan,

h Medium Electronic Document

a Title Global Binarization of Document Images Using a Neural Network

c Statement of responsibility, etc Adnan Khashman, Boran Sekeroglu.

b Name of publisher, distributor, etc IEEE,

c Date of publication, distribution, etc 2008.

a Place of publication, distribution, etc Los Alamitos :

a Summary, etc In degraded scanned documents, where considerable background noise or variation in contrast and illumination exists, pixels may not be easily classified as foreground or background pixels. Thus, the need to perform document binarization in order to enhance the document image by separating foregrounds (text)from backgrounds. A new approach that combines a global thresholding method and a supervised neural network classifier is proposed to enhance scanned documents and to separate foreground and background layers. Thresholding is first applied using Mass-Difference thresholding to obtain various local optimum threshold values in an image. The neural network is then trained using these values at its input and a single global optimum threshold value for the entire image at its output. Compared with other methods, experimental results show that this combined approach is computationally cost-effective and is capable of enhancing degraded documents with superior foreground and background separation results.

a Topical term or geographic name as entry element Computer science

a Topical term or geographic name as entry element Near East University Article

a Topical term or geographic name as entry element Yakın Doğu Üniversitesi Makale

a Topical term or geographic name as entry element Electrical & Electronic

a Main entry heading SITIS 2007: Proceedings Of The International Conference On Signal Image Technologies & Internet Based

g Relationship information 2008, p665-672

u Uniform Resource Identifier http://library.neu.edu.tr:2048/login?url=http://dx.doi.org/10.1109/SITIS.2007.58