Cloud Enabled Text Reader for Individuals with Vision Impairment

摘要

The paper describes the development of a text reader for people with vision impairments. The system is designed to extract the content of written documents or commercially printed materials. In terms of hardware, it utilizes a camera, a small embedded processor board, and an Alexa Echo Dot. The software involves an open source text detection library called Tesseract along with Leptonica and OpenCV. The system in its current version can only work with English text. By using the Amazon cloud web services, a skill set was deployed, which would read aloud the detected text utilizing a OpenCV program via the Alexa Echo Dot. For this development, a Raspberry Pi was utilized as the embedded processor system.