Vigle: A Visual Graphical Learning Module on Optical Character Recognition

Abstract

Students of the Arts and Humanities use OCR to convert scanned images of old text (pre-1800 AD). They need to know how digital text is extracted from the scanned image. Thanks to cell phones and images captured with them, understanding this is useful for everybody. The processing steps employed by a typical OCR software are, in order, Binarization, Deskew, Segmentation, Character Segmentation and Character Recognition. In this research project, a standalone asynchronous visual graphical learning environment (VIGLE) on Optical Character Recognition (OCR) was developed. Constructivist learning strategy was employed. The learning module was integrated into a website that works on mobile. The project attempts to generalize the instruction so that it is useful for everybody. Latest web technology was used for the implementation to achieve one stop interface, browser compatibility, responsive window sizing and interactive visual content. Binarization, Deskew and Segmentation modules were implemented in the time available. The VIGLE consists of a graphical representation and a visual interface to the lessons. Results show that the participants found both the graphical representation and the visual interface helpful. They found the incomplete learning module on OCR at best moderately useful in helping them digitize text.