Title :
Automated pagination and layout analysis of handwritten notes
Author :
Mandalapu, D. ; Madhvanath, S. ; Babu, B. Ramesh ; Bellam, H.
Author_Institution :
Hewlett Packard Labs., Bangalore, India
Abstract :
Simple “electronic clipboards” such as Pegasus NotesTaker and ACECAD DigiMemo offer a cost effective solution for capturing handwritten notes in electronic form, in many settings such as meetings, schools, and so on. However they do not have a way of knowing when the writer is done with the current physical page and has moved to a new page. They address this problem by providing a physical "new page" button on the device. However writers frequently forget to press this button, resulting in multiple physical pages of ink being captured in a single digital page. In this paper, we describe an algorithm to automatically separate (paginate) multiple pages of ink into the constituent pages, thereby doing away with the need to explicitly indicate the new page. We then perform layout analysis on each of the digital pages. Text strokes and graphic strokes in a page are separated, and the ink is grouped into logical segments such as figures and paragraphs. We also describe a tool that integrates the algorithms for automatic pagination and layout analysis, and provides a convenient user interface to move digital ink segments across pages, and reorganize handwritten notes.
Keywords :
handwritten character recognition; notebook computers; text analysis; user interfaces; automated pagination; electronic clipboards; graphic strokes; handwritten notes; layout analysis; logical segments; text strokes; user interface; Algorithm design and analysis; Clustering algorithms; Ink; Layout; Signal processing algorithms; Writing; Layout analysis; Pattern clustering; Pattern recognition; Text analysis;
Conference_Titel :
Intelligent Systems and Signal Processing (ISSP), 2013 International Conference on
Conference_Location :
Gujarat
Print_ISBN :
978-1-4799-0316-0
DOI :
10.1109/ISSP.2013.6526940