Authors:
M.S. Minu, S.S. Subashka Ramesh, R. Aroul Canessane, Muhammad Al-Amin, Rejwan Bin Sulaiman

Addresses:
1,2Department of Computer Science and Engineering, SRM Institute of Science and Technology, Ramapuram, Chennai, India. 3Department of Computer Science, Sathyabama Institute of Science and Technology, Chennai, Tamil Nadu, India. 4Department of International Relations, Sichuan University, Chengdu, China. 5Department of Computer Science and Technology, Northumbria University, London, United Kingdom. msminu1990@gmail.com1, subashka@gmail.com2, aroulcanessane@gmail.com3, alamin2022@stu.scu.edu.cn4, rejwan.sulaiman@northumbria.ac.uk5

Abstract:

Unmanned aerial vehicles (UAVs) have recently attracted numerous regards from researchers and academics. The UAV is useful in heterogeneous applications such as transportation, disaster monitoring, surveillance, etc. Because UAVs have limited internal energy, clustering can effectively balance energy consumption and load. On the other hand, scene classification on high-quality remote sensing photos captured by UAVs is difficult for UAV networks. This paper provides an Oppositional Glowworm Swarm Optimization with Deep Learning Enabled Clustering with Classification (OGSODL-CC) scheme for UAV networks in this regard. The proposed OGSODL-CC model primarily aims to cluster UAVs for energy efficiency and classification. The OGSODL-CC algorithm learns how to maximize its fitness from residual power, trust, and distance to neighbours. The scene classification model includes NASNet feature extraction and the SoftMax classifier. Many scenarios are enacted, and the results are evaluated in several regards to demonstrate the enhanced outcomes of the OGSODL-CC algorithm. The experimental results revealed that the OGSODL-CC model outperformed the previous techniques.

Keywords: Opposition Based Learning; SoftMax Classifier; Scene Classification; Clustering of Metaheuristics; Decision Range; OGSODL-CC algorithm; Unmanned Aerial Vehicles; Deep Learning Enabled Clustering with Classification.

Received on: 15/02/2023, Revised on: 22/05/2023, Accepted on: 21/07/2023, Published on: 28/11/2023

FMDB Transactions on Sustainable Computing Systems, 2023 Vol. 1 No. 3, Pages: 124-134

• Views : 203
• Downloads : 12

Download PDF