Scribd
Upload
33 views3 pages

Military Aircraft Detection with DET-YOLO

The document presents the DET-YOLO model, a high-performance deep-learning solution for detecting military aircraft in remote sensing images, addressing issues of low detection rates and complex backgrounds. It features innovative components like the C2f_DEF module for small-object detection and an optimized loss function for improved accuracy and reduced computational costs. Performance validation shows DET-YOLO achieving 94.7% mAP@0.5, outperforming existing models in accuracy, speed, and efficiency.

Uploaded by

krithikamurali1404
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
33 views3 pages

Military Aircraft Detection with DET-YOLO

The document presents the DET-YOLO model, a high-performance deep-learning solution for detecting military aircraft in remote sensing images, addressing issues of low detection rates and complex backgrounds. It features innovative components like the C2f_DEF module for small-object detection and an optimized loss function for improved accuracy and reduced computational costs. Performance validation shows DET-YOLO achieving 94.7% mAP@0.5, outperforming existing models in accuracy, speed, and efficiency.

Uploaded by

krithikamurali1404
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd

Detecting Military Aircraft in Remote Sensing Images.

2024-2025

ACKNOWLEDGEMENT

It is with the profound feeling of gratitude we would like to express our sincere thanks to
our institution Dr. T. THIMMAIAH INSTITUTE OF TECHNOLOGY, K.G.F for providing
excellent infrastructure for the successful completion of Project Phase 1.
We wish to express a whole hearted thanks to our Principal Dr. Syed Ariff for providing
lab facilities and requirements in carrying out this Project Phase 1.
We would like to extend hearty thanks to our Dean (Administration) Prof. Ruckmani
Divakaran for being a constant support of encouragement to carry out the Project Phase 1
successfully.

We would like to extend hearty thanks to our HOD Dr. Jenitha. A, HOD of ECE, for her
valuable suggestions, guidance and support in the completion of this Project Phase 1.
We would like to extend hearty thanks to our guide Prof. Jesudas J, Assistant
Professor, for his valuable suggestions, guidance and support in the completion of this Project Phase 1.
We would also like to thank Project Coordinator Prof. Mohana. C for her timely support in
completion of this Project Phase 1.
We would like to thank all teaching and non-teaching staff who was directly and indirectly
supported for carrying out this Project Phase 1 successfully.
We extend our hearty thanks to our parents, friends for all the moral support provided during
the preparation for the Project Phase 1.

DHARSHA M S [1GV22EC017]
JAYA BHARATHI V [1GV22EC020]
KRITHIKA MURALI [1GV22EC022]
ANU RANJANI [1GV22EC041]

5th Sem Dept. of ECE, Dr. T.T.I.T, K.G.F i


Detecting Military Aircraft in Remote Sensing Images. 2024-2025

ABSTRACT

“DET YOLO-AN INNOVATIVE HIGH PERFORMANCE MODEL FOR


DETECTING MILITARY AIRCRAFT IN REMOTE SENSING IMAGES”
Accurate detection of military aircraft in remote sensing images is crucial for defense and surveillance
applications. However, traditional models often struggle with low detection rates, high missed detections,
and complex background interference. To address these challenges, we introduce DET-YOLO, a high-
performance deep-learning model optimized for real-time aircraft detection. DET-YOLO incorporates
C2f_DEF, a novel feature extraction module designed to enhance small-object detection, and the DAT
module, which improves adaptability in complex environments. Additionally, an optimized loss function
reduces computational costs while boosting accuracy. The model’s lightweight architecture ensures ease
of deployment, making it ideal for resource-limited settings. Performance validation on the MAR20
dataset demonstrates DET-YOLO’s superiority, achieving 94.7% mAP@0.5 with only 80 training
epochs. It outperforms existing models in accuracy, speed, and efficiency, making it a cutting-edge
solution for real-time military aircraft detection in remote sensing applications.

5th Sem Dept. of ECE, Dr. T.T.I.T, K.G.F ii


Stimulation of single Antenna Full Duplex system 2024-25
System

3
5th Sem Dept. of ECE, Dr. T.T.I.T, K.G.F

You might also like