Abstract
According to the World Health Organization (WHO), there are millions of
visually impaired people in the world who face a lot of difficulties in moving
independently. 1.3 billion people are living with some visual impairment problem,
while 36 million people are completely visually impaired. We proposed a system for
visually impaired people to recognize and detect objects based on a convolutional
neural network. The proposed method is implemented on Raspberry Pi. The ultrasonic
sensors detect obstacles and potholes by using a camera in any direction and generate
an audio message for feedback. The experimental results show that the Convolutional
Neural Network yielded impressive results of 99.56% accuracy.