Portable Camera Based Assistive Text and Product Label Reading from Hand-Held Objects for Blind People Using Raspberry Pi

Abstract

This project involves designingObstacles detection and text reading aid for blind people using Raspberry Pi 3, ultrasonic sensor etc. Raspberry Pi 3 is the controller used in this device. The obstacle is detected by the ultrasonic sensor and the distance is sent  to the blind user as audio output. GPS is used to find the exact location of the person in terms of longitude and latitude, this information is sent to the care taker through e-mail. The caretaker can use the latitude and longitude to find the address on Google Maps. Camera is used to identify the obstacle or object ahead and the output is send to the blind user in audio form. The camera also identifies objects with words on them, using image processing these images are converted to text and using Tesseract the text is converted into audio, thus giving the audio output to the blind about what is written on the object. The buzzer goes ON to identify the stick location. A threshold value for distance between the user and the stick is set, when the distance is less than the threshold value, the buzzer sound increases.