Object Recognition for Quadcopter Drone using Convolutional Neural Networks
G. Ragu1, K. Dheeraj2, M. Rama Mohan Reddy3, B. Venkata Sai4
1G.Ragu*, Department of Computer Science and Engineering, SRM Institute of Science and Technology, Chennai.
2K. Dheeraj, Department of Computer Science and Engineering, SRM Institute of Science and Technology, Chennai
3M. Rama Mohan Reddy, Department of Computer Science and Engineering, SRM Institute of Science and Technology, Chennai
4B. Venkata Sai, Department of Computer Science and Engineering, SRM Institute of Science and Technology, Chennai
Manuscript received on April 20, 2020. | Revised Manuscript received on April 25, 2020. | Manuscript published on May 10, 2020. | PP: 224-227 | Volume-9 Issue-7, May 2020. | Retrieval Number: E3152039520/2020©BEIESP | DOI: 10.35940/ijitee.E3152.059720
Open Access | Ethics and Policies | Cite | Mendeley
© The Authors. Blue Eyes Intelligence Engineering and Sciences Publication (BEIESP). This is an open access article under the CC-BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/)
Abstract: Object detection is as of now generally utilized in industry. It is the strategy for location and design of genuine items. Models incorporate intermittent scaffold examinations, debacle the executives, power line observation and traffic examinations. As UAV applications become progressively broad, more significant levels of self-sufficiency and free dynamic procedures are expected to improve the security, proficiency and exactness of the gadgets. This article exhibits in detail the method and parameters important for the preparation of convolutional neural systems (CNN) in the programmed acknowledgment of items. The potential areas of utilization in the vehicle division are additionally featured. The precision and unwavering quality of the CNNs rely upon the arrangement of the system and the determination of working parameters. The impact of article recognition shows that by picking a parameter setting course of action, a CNN can recognize and gather objects with a noteworthy degree of accuracy (97.5%) and computational profitability. Moreover, utilizing a convolutional neural system actualized in the YOLO stage (V3), items can be followed, distinguished and characterized progressively.
Keywords: Convolution Neural Networks, Object Recognition and Detection, YOLO(V3).
Scope of the Article: Convolutional Neural Networks