Cognitive learning enabled real time object search robot

Object Tracking is usually performed in the context of higher-level applications that require the location and/or shape of the object in every frame. Most works are focused on a specific application, such as tracking human, car, or pre-learned objects. All these require database and considerable amount of training time to detect the current object and to track it. In this paper we propose a method to track objects where a pre-stored database is not a requirement. The proposed method uses a combination of Scale Invariant Feature Transform (SIFT) based feature extraction, Kalman filter and Cognitive learning. The algorithm has the ability to make its own database of the objects in the due course of time by interacting with the user through text based communication. This algorithm is deployed on a search robot which does the operation of searching an object in real time upon a command from the user. The search operation of robot is made more flexible using Bluetooth wireless communication protocol.