ORIGINAL ARTICLE Speed-up ellipse enclosing character detection approach for large-size document images by parallel scanning and Hough transform H. Waruna H. Premachandra • Chinthaka Premachandra • Chandana Dinesh Parape • Hiroharu Kawanaka Received: 7 April 2014 / Accepted: 11 January 2015 Ó Springer-Verlag Berlin Heidelberg 2015 Abstract This paper presents a speed-up ellipse enclos- ing character detection algorithm that uses parallel image scanning and the Hough transform (HT) for large-size document images. Objects in images are generally detected based on geometrical information obtained via raster scanning. In raster scanning, all pixels of an image are scanned starting from the upper-left point and ending with the lower-right point. In the case of large-size images, considerable time is needed for processing an image by scanning all pixels. In this paper, an object detection approach for large-size images is proposed which does not require scanning all pixels in the images. In this speed-up ellipse enclosing character detection approach for large- size document images, pixels are scanned on constantly spaced vertical parallel lines. If an object larger than a certain size is identified while scanning, the presence of an ellipse enclosing character is assumed and ellipse detection is conducted by applying HT only in a defined local image area over the found object. With this approach, processing time can be dramatically reduced by disregarding some objects and reducing the total image area used for ellipse detection. Keywords Speed-up image processing  Document image  Ellipse detection  Parallel pixel scanning  Hough transform 1 Introduction Document image processing is an interesting research area in computer vision, and many studies can be found in the literature concerning document scanning, document struc- ture understanding, document computerization, and so on [1–16]. Document images should be of a sufficient size for analyzing information in the document effectively. How- ever, considerable time is required to process a large-size image. In this paper, an image larger than 2,480 9 3508 pixels is considered a large-size image. Depending on the processing, dozens of minutes may be required to process such a large-size document image for a given objective. Image size is one of the main factors that increase pro- cessing time. Therefore, shortening processing times for large-size images is a desirable goal. In this paper, short- ening processing times for ellipse enclosing character detection is addressed. We accomplish this by skipping some objects, which can be disregarded, in the pixel scanning stage. In addition, targeted ellipse enclosing characters are detected only by processing limited image domains. Raster scanning is used in digital image processing. In raster scanning, an image is scanned starting from the upper-left pixel and ending with the lower-right pixel. In the case of large-size images, considerable processing time is required to scan all pixels. In the approach presented in H. W. H. Premachandra Wayamba University of Srilanka, Makadura, Srilanka e-mail: warunaprema@yahoo.com C. Premachandra (&) Tokyo University of Science, 6-3-1 Niijuku, Katsushikaku, Tokyo 125-8585, Japan e-mail: chinthaka@ee.kagu.tus.ac.jp C. D. Parape Kyoto University, C1-1-209 Kyoto University Katsura, Nishikyo-Ku, Kyoto 615-8540, Japan H. Kawanaka Mie University, 577 Kurimamachiya, Tsu, Mie 514-8507, Japan 123 Int. J. Mach. Learn. & Cyber. DOI 10.1007/s13042-015-0330-0