APPLICATION OF ORIENTED FAST AND ROTATED BRIEF (ORB) AND BRUTEFORCE HAMMING IN LIBRARY OPENCV FOR CLASSIFICATION OF PLANTS

Abstract

The identification of plants is a process of recognition of plants to discover the types to disseminate knowledge, but the names of the plants can be different from each person (community, profession, language) in different geographical locations. Artificial vision is very useful for mobile devices, together with the appearance of cell phone cameras equipped with phones, new opportunities and challenges have been generated in the field of artificial intelligence (AI). The combination of ORB and BruteForce_Hamming has better computational efficiency based on the highest accuracy and identification efficiency in the implemented software and can perform accurate real-time identification. The design and implementation of the plant classification identification identification program is based on the Android operating system with a minimum level of system 21 (Lollipop) to show the key match point of the live camera to obtain type information. This application can classify plants with a correct accuracy rate of 88% and an identification error rate of 12% on pandanus leaves.

Keywords: plants, species, Android, Keypoint, ORB, BruteForce Hamming