In this paper, an imaginative, productive and low-cost efficient system for visually impaired people is implemented that helps them to hear the content present in the images (such as soaps, newspapers and many day-to-day products) and also helps them to stop colliding with the obstacles in their way. For hearing the content present in the image, it comprises of two modules picture handling module and voice handling module. Text extraction from shading and color images is a difficult asks in computer vision. This system extracts the content of text from the picture placed before camera using OCR (Optical Character Recognition) and then transforms the extracted text into speech using Open CV libraries in python. It detects English alphabets as well as numbers present in an image and transforms into speech. For obstacle detection, ultrasonic sensor is utilized to avoid crashing with the obstacles in their way. This system is useful for visually impaired people in two different ways: Hearing the content from pictures and by detecting obstacles. This paper portrays the structure, implementation, usage and test consequences of the system.
Volume 12 | Issue 2
Pages: 1367-1373
DOI: 10.5373/JARDCS/V12I2/S20201175