A Novel Approach for Assisting Blind People Using a Smart Wearable Device

As per the records of world blindness and visual impairment, globally there are millions of blind people, facing several problems in leading an independent life. The aim of the proposed model is to design and develop an intelligent voice-based guidance system using various ML and DL algorithms and t...

Full description

Saved in:
Bibliographic Details
Main Authors: Peraka, Shyam, Ali, SK Irfan, Sudheer, Reddy, Kumar, Pilli Praveen, Kondala, Goutham, Samal, Dimple
Format: Conference Proceeding
Language:English
Subjects:
Azure Maps
Blindness
Deep neural networks
Dimensional ratio similarity
Eigen faces
Face recognition
Haar cascades
Octa-polar segmentation
ONNX model
OpenCV
Optical character recognition
Paddle OCR
Performance evaluation
Principal Component Analysis
Prototypes
Py-Tesseract
Py-YOLOV5 algorithm
Viola Jones algorithm
Visual impairment
Wearable computers
Online Access:Request full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:As per the records of world blindness and visual impairment, globally there are millions of blind people, facing several problems in leading an independent life. The aim of the proposed model is to design and develop an intelligent voice-based guidance system using various ML and DL algorithms and to enhance their lifestyle through our smart solution. The proposed prototype consists of five (05) unique functional blocks in which each block assists the user in performing the corresponding task through voice commands from the headset. 1.Navigation and vehicle avoidance block: This takes the destination name from the user and guides the user along the shortest path to the destination. Meanwhile directs the user to avoid the vehicles along the path of the journey. 2.Face recognition and face-saving block: Recognizes the known persons to the user (from saved faces of the user's database) and informs the user of their name. In the case of an unknown person, it facilitates the user to save his/her face. 3.Object detection and obstacle avoidance block: Helps the user to detect various objects and clusters them into obstacles and non-obstacles. In case of obstacles, it warns the user to deviate in a sophisticated direction. 4.Text extraction and text-to-speech conversion block: Helps the user to study the detected text present on the books, hoardings, etc without the dependency on Braille script. It is mainly helpful for blind students. 5.Currency recognition block: Helps the user to identify the currencies of several countries including their values.
ISSN:2380-6923
DOI:10.1109/VLSID57277.2023.00082