UGC Approved Journal no 63975(19)

ISSN: 2349-5162 | ESTD Year : 2014
Call for Paper
Volume 11 | Issue 3 | March 2024

JETIREXPLORE- Search Thousands of research papers



WhatsApp Contact
Click Here

Published in:

Volume 9 Issue 1
January-2022
eISSN: 2349-5162

UGC and ISSN approved 7.95 impact factor UGC Approved Journal no 63975

7.95 impact factor calculated by Google scholar

Unique Identifier

Published Paper ID:
JETIR2201479


Registration ID:
319446

Page Number

e574-e584

Share This Article


Jetir RMS

Title

Optimized Deep Learning Framework for Scene Classification on UAV Images

Abstract

With the recent advancement triggered in the field of the automated territory control and monitoring, the usage of the remote sensing acquisition has reached a new level of acceptance. Furthermore, the unmanned aerial vehicles (UAV’s) based automation also gained popularity with respect to real time application in scene identification, land classification etc. The reason for the manifestation of UAV is their ability to acquire high resolution data irrespective of the geographical areas even if it is inaccessible. The size of the UAV and its flight capability with ease of use has been manifested as the reason for the consideration of the detailed image acquisition even with limited coverage zones. The adaptation of the image processing techniques on the UAV images for the application-oriented environment is found to be a surging research with the growth in the machine learning, artificial intelligence and deep learning because of the abstraction which can be achieved through diverse features of the UAV images without the assistance of human. However, the potential research field encountered variety of problem in deep learning techniques dedicated for the UAV scene classification system because of improper image processing (pre-post), computational redundancy in the deep learning framework, uncertainties in the feature extraction and GUI for the deployment. In addition to the framework issues, the hyper parameter tuning in the deep learning algorithms were also limited with its non-dominant solution. In order to overcome the aforementioned issues with UAV scene classification, in this research a deep learning framework optimized with particle swarm optimization algorithms was developed with marginal processing techniques and feature extraction paradigm. To evaluate the performance of the proposed model, two open-source UAV datasets were used. The experimental results proved in comparison to all existing methods that the proposed P-CNN framework has a maximum accuracy of 98.7%. The proposed model has been successfully deployed for usage in a GUI with the help of serialization for scene categorization in real time.

Key Words

Convolutional neural networks Deep learning, Machine learning, Particle Swarm Optimization,UAV.

Cite This Article

"Optimized Deep Learning Framework for Scene Classification on UAV Images", International Journal of Emerging Technologies and Innovative Research (www.jetir.org), ISSN:2349-5162, Vol.9, Issue 1, page no.e574-e584, January-2022, Available :http://www.jetir.org/papers/JETIR2201479.pdf

ISSN


2349-5162 | Impact Factor 7.95 Calculate by Google Scholar

An International Scholarly Open Access Journal, Peer-Reviewed, Refereed Journal Impact Factor 7.95 Calculate by Google Scholar and Semantic Scholar | AI-Powered Research Tool, Multidisciplinary, Monthly, Multilanguage Journal Indexing in All Major Database & Metadata, Citation Generator

Cite This Article

"Optimized Deep Learning Framework for Scene Classification on UAV Images", International Journal of Emerging Technologies and Innovative Research (www.jetir.org | UGC and issn Approved), ISSN:2349-5162, Vol.9, Issue 1, page no. ppe574-e584, January-2022, Available at : http://www.jetir.org/papers/JETIR2201479.pdf

Publication Details

Published Paper ID: JETIR2201479
Registration ID: 319446
Published In: Volume 9 | Issue 1 | Year January-2022
DOI (Digital Object Identifier):
Page No: e574-e584
Country: Bangalore, karnataka, India .
Area: Engineering
ISSN Number: 2349-5162
Publisher: IJ Publication


Preview This Article


Downlaod

Click here for Article Preview

Download PDF

Downloads

000378

Print This Page

Current Call For Paper

Jetir RMS