An Android Application for Visually Impaired Person Using Deep learning.
Hozefa Abbas Arielwala, Salik Zulfiqar Ahmed Ansari, Mohd Eklaque Shah, Shaikh Mohd Ashfaque · Journal of Emerging Technologies and Innovative Research · 2021
Presently, visually impaired individuals carry sticks with them which helps them to scan their surroundings for obstacles or orientation marks and to measure the distance between them and objects in their surroundings. If the visually impaired people are provided with an audio description of their surroundings, it would have a significant effect on their lives and help them understand their surroundings better. The recent advances in Deep Learning and Computer Vision have led to excellent Image Captioning models using advanced techniques like Deep Reinforcement Learning. These captions can then be read out loud to the visually impaired so that they can have a better understanding of their surroundings. Our model uses a convolutional neural network (CNN) to extract features from an image which is then converted into a sentence, describing the image invalid English by feeding it to a recurrent neural network or a Long Short-Term Memory network. We believe that our model with greatly improve the life of visually impaired people by allowing them to understand their surroundings.