No abstract

There is a significant need to objectively evaluate layout analysis (

Braille is the most effective means of written communication between visually-impaired and sighted people. This paper describes a new system that recognizes Braille characters in scanned Braille document pages. Unlike most other approaches, an inexpensive flatbed scanner is used and the system requires minimal interaction with the user. A unique feature of this system is the use of context at different levels (from the pre-processing of the image through to the post-processing of the recognition results) to enhance robustness and, consequently, recognition results. Braille dots composing characters are identified on both single and double-sided documents of average quality with over 99% accuracy, while Braille characters are also correctly recognised in over 99% of documents of average quality (in both single and double-sided documents).

Abstract. Over the past two decades a significant number of layout analysis (page segmentation and region classification) approaches have been proposed in the literature. Each approach has been devised for and/or evaluated using (usually small) application-specific datasets. While the need for objective performance evaluation of layout analysis algorithms is evident, there does not exist a suitable dataset with ground truth that reflects the realities of everyday documents (widely varying layouts, complex entities, colour, noise etc.). The most significant impediment is the creation of accurate and flexible (in representation) ground truth, a task that is costly and must be carefully designed. This paper discusses the issues related to the design, representation and creation of ground truth in the context of a realistic dataset developed by the authors. The effectiveness of the ground truth discussed in this paper has been successfully shown in its use for two international page segmentation competitions (ICDAR2003 and ICDAR2005).