A Cognitive Framework on Object Recognition and Localization for Robotic Vision

摘要

The main aim of this work is to develop an algorithm for object recognition and localization with application to automatons. The algorithm development is done in three phases: Initially, the algorithm for object detection is framed which in result gives us segmented object matrices. Secondly, the object recognition algorithm is framed which in result gives the object label of recognized object. Thirdly, the position of recognized object is found out from object localization algorithm. Vision is vital for the functioning of robots in amorphous surroundings. In an aim of simplifying the interactions between human and robots as well as to carry out the more complex operations, cognitive abilities such as visual perception are appended to the systems for the sake of intelligent operations. The consolidated developed algorithm will recognize the desired object and also gives its location in the world coordinate system. The algorithm developed in this paper proves to be good and fast as we achieve the accuracy of 100% in object detection as well as in recognition algorithm and 98.03% to 99.99% accuracy in object localization algorithm, while considering the unvarying light conditions.