click here to download project abstract
ABSTRACT
There is a ton of classified and earnest data trade here today which is converted into.
a desired form with the convolutional neural network to make the blind feel things
surrounding him. Efficient Implementation of this Model depends upon the
compatibility with python and library installation hurdles. The proposed Blind
Assistance ideas were developed to help those visually impaired people around the
world. The proposed system has two levels. It works on the SSD algorithm and
TensorFlow, in which it recognizes the object, not only the recognition but also
localization of an object. It also provides the distance between the person and the
object. The Laptop Based Server will be using a pre-trained SSD detection model
trained on COCO DATASETS. It will then test and the output class will get detected
with an accuracy metrics. In this manner, we have proposed a framework that utilizes
AI, which makes the framework more straightforward to use for individuals with visual
impedances. After testing with the help of voice modules the class of the object will
be converted into a default voice notes which will then be sent to the blind victims for
their assistance. Main key aspect of the proposed system is identifying or naming the
object detected, prints accurate distance between the user and the objects and voice
over using Audio Commands.
