Real Time Camera-Based Reading Assistant Using OpenCV

Abstract

While science and technology have made great strides in making life easier for people, these advancements have not been made equally in all spheres of life. Little has been done to give visual aids to help blind and visually impaired people achieve academic brilliance and better prospects, even though instruments are available to help them navigate their environment. By creating a visual learning tool for the blind and visually handicapped, our project seeks to close this gap. The created model uses a camera to first take pictures of printed text, which it then transforms into digital text using Optical Character Recognition (OCR) and various OpenCV pre-processing methods. Using Text-to-Speech (TTS) technology, the text is then converted into speech output and transmitted to the user via earphones. The system is composed. The system is composed of a power supply, camera, headphones, and Raspberry Pi processor. The device provides a useful and effective solution to the academic and educational needs of the visually impaired, improving their quality of life and prospects for the future. Tesseract OCR and Mimic3 TTS technology allowed for accurate and efficient recognition and synthesis of text, making this model useful in real-time to access and read printed materials. This technology can help provide equal opportunities and promote the academic and scholastic success of people who are blind or visually impaired with continued growth and improvement.