Implication Node built Directed Acyclic Graph (INDAG) based Visual Aid
Nisha Kumari, Parth Bir, Shylaja Vinaykumar Karatangi, Prachi, Reshu Agarwal · 2020
This paper presents the design and implementation of a visual aid. It comprises of a radar unit, computer vision unit and a processing unit. The designed unit initially uses an implication graph on received data to form Implication Nodes (IN). These nodes are further used to form a Directed Acyclic Graph (DAG). The IN and DAG data is further fed to an optimization engine or Machine Learning Pipeline (MLP) for segregation and formation of the three dimensional environment. This environment is analysed by the processing unit and requisite voice commands are dictated to the user. Furthermore, this reinforced learning method ensures user specific learning and hence best performance results.