Project

Overview

Blind Assistance System is a program which could help blind people in their daily lives. This Blind assistance system could detect 80 types of objects using tensorflow with SSD mobilenet V1 and when an object is detected, an audio is played which tells the user what kind of object is present.

This project uses python with TensorFlow. TensorFLow is a powerful open-source deep learning framework that provides the necessary tools and functions to build, train, and deploy machine learning models, especailly deep neural networks. Tensorflow offers a wide range of pre-trained object detection models like SSD that is Single Shot Multibox Detector, Faster RCNN (Region-based Convolutional Neural networks), and YOLO (You Only Look Once ). Here SSD mobilenet V1 is used.

Coco dataset, which is a widely used dataset for object detection, is used in this program.