Image Processing System for Providing Sound Information as an Assistive Tool for the Visually Impaired

Blind individuals commonly use canes to help navigate their daily environments. However, this research introduces a more advanced solution by developing a smart stick that utilizes technology to differentiate between objects, such as people and vehicles, through an image processing system that delivers audio feedback via a headset. The system is powered by a Raspberry Pi microprocessor and equipped with an ultrasonic sensor to measure the distance between the user and surrounding objects. Once the sensor detects an object, the camera captures a real-time image, which is then processed using Single Shot Detector (SSD) and Cascade Classifier methods for object recognition. After successfully identifying the object, the system converts the visual information into audio signals relayed through the headset, with prompts such as “person ahead” or “car approaching.” Testing showed the system could accurately detect humans and vehicles, with average response times of 1.38 seconds and 4.244 seconds, respectively, and a margin of error in distance detection of approximately 0.281%.