Converting Optically Scanned Regular or Irregular Tables to a Standardised Markup Format to be Accessible to Vision-Impaired

Abstract

Documents use tables to communicate multidimensional information clearly, summarise and present data in an easy-to-interpret way. Tabular information in scanned PDF due to its nature without further processing is not accessible for vision-impaired people who use assistive technology such as screen readers. The lack of access to table contents limits educational and workplace opportunities for people with vision impairment. They require a complete equivalent to access table. This paper describes techniques which apply to scanned PDF document for table detection, extraction and cell segmentation to retrieve cell contents and represent them in a navigable manner to vision-impaired. The output is in mark-up format and provides navigation ability to access content of a table.