Real Time Web-based System to Detect Military Aircraft Using RESNET-50 Algorithm

Author:

Venkata Sudhakar C1,Deekshitha Limbakar1,Kummari Charan1,Pintu Sah Rauniyar1,Kessamsetty Mahathi1

Affiliation:

1. Sree Vidyanikethan Engineering College, Tirupati, Andhra Pradesh, India.

Abstract

As target detection in remote sensing imaging depends on aircraft type recognition, it is essential in both civil and military applications. The job is made more difficult by the existence of fine-grained features, which can result in significant intra-class changes due to variations in size, posture, and angle, as well as modest inter class changes due to very similar subcategories. This kind of system can be helpful for military security as recognition of the type of aircraft is very critical to the decisions being made. There are several existing ways which uses methods like Radar System and Radio footprints, Speed etc., to detect type of Aircraft. Although these methods are massively costly and still cannot detect the type of Aircraft accurately. In this paper aircraft is detected using ResNet-50, Advance State of Art Object Detection Algorithm implementing in Anaconda tool with train accuracy is 98% & validate accuracy is 75%. A crucial area of artificial intelligence is object detection, which enables computer systems to perceive their surroundings by identifying things in visual pictures or movies. In case of any dangerous Aircraft, the system will have capability to raise alarm and Alert using Audio Sirens. The software requirement for this project is python, 3.6/anaconda, or newer and necessary python modules.

Publisher

REST Publisher

Subject

General Medicine,Materials Chemistry,General Medicine,General Medicine,General Materials Science,General Medicine,General Medicine,Aerospace Engineering,General Medicine

同舟云学术

1.学者识别学者识别

2.学术分析学术分析

3.人才评估人才评估

"同舟云学术"是以全球学者为主线,采集、加工和组织学术论文而形成的新型学术文献查询和分析系统,可以对全球学者进行文献检索和人才价值评估。用户可以通过关注某些学科领域的顶尖人物而持续追踪该领域的学科进展和研究前沿。经过近期的数据扩容,当前同舟云学术共收录了国内外主流学术期刊6万余种,收集的期刊论文及会议论文总量共计约1.5亿篇,并以每天添加12000余篇中外论文的速度递增。我们也可以为用户提供个性化、定制化的学者数据。欢迎来电咨询!咨询电话:010-8811{复制后删除}0370

www.globalauthorid.com

TOP

Copyright © 2019-2024 北京同舟云网络信息技术有限公司
京公网安备11010802033243号  京ICP备18003416号-3