Deep Learning Technique for Serving Visually Impaired Person
Pragati Chandankhede, Arun Kumar · 2019
The concepts like machine learning and artificial intelligence have been adopted in the real time applications due to their exceptional features and absolute algorithmic advantages. Deep learning is a one of research area in machine learning totally concerned with algorithmic techniques inspired by the structure and functions of the human neural system. It has created a niche for itself in learning processes involving multiple level of data representation and abstraction, thereby helping in making sense of data having several forms such as images, sound and text and finally revealing the feature level hierarchies. Deep learning techniques show higher level of accuracy and robustness in results and are independent of human crafted features. The Paper is divided into three main parts, first part identifies the need to develop the system for Visually Impaired person. Second part deals about various Deep learning techniques used for computer vision and object detection. And third part consist the proposed system for serving visually impaired person in doing daily chores.