Towards Automatic Interpretation Of Statistical Graphs For The Visually Impaired

Abstract

Statistical graphs, as visual representations of quantitative data, are usually not able to be processed by assistive technologies. Thus, when a user with visual impairment intends to access the information contained in these visual representations, he has to resort to the use of screen readers. But such tools only transmit the adjacent information of the image, without performing image processing to generate an interpretation of the data contained. To tackle such problem, a conceptual and functional architecture is proposed considering the use of image processing techniques to extract in an automated way the information of a statistical graphs under an image format, generating a description accessible to users with visual impairment. To support it, it is carried out research on related work in image processing to identify relevant elements in a chart, as well as advantages and disadvantages of assistive technologies. With the results obtained, it is observed that there is an increase in the use of assistive technologies based on the sonification of textual and graphic elements to transmit information to users with visual disabilities through the use of Web platforms.