PAAVAI COLLEGE 0F ENGINEERING

Pachal,Namakkal-637018
AIRCRAFT DETECTION CLASSIFICATION AND MECHANISM USING DEEP LEARNING

GUIDED BY: PRESENTED BY

MRS.M.MENAGA M.E (AP) KANISHKARAN E-622020243018
DEPARTMENT OF AI&DS SIVASANKAR-622020243036
PAAVAI COLLEGE OF ENGINEERING SAKTHIVEL-622020243032
NAMAKKAL – 637018 SANJAI R-622020243033

1
Introduction
• In today's dynamic military landscape, the need for advanced surveillance and rapid response capabilities is
paramount. The classification of military aircraft plays a crucial role in maintaining airspace security and
operational readiness.
• Develop a robust system using YOLOv8 to accurately identify various military aircraft types in real-time,
addressing the critical need for advanced surveillance and rapid response capabilities.
• Integrate alarm sound activation and email notification functionalities to ensure immediate alerts to
personnel upon aircraft detection, enhancing system responsiveness.
• Harness the power of deep learning and real-time communication technologies to enhance situational
awareness and enable prompt decision-making in dynamic military landscapes.
• Implement comprehensive dataset collection, model training, and rigorous testing to bolster military airspace
surveillance and operational efficiency.
• Ultimately, the innovative solution aims to strengthen military capabilities, safeguard personnel, and ensure
mission success in challenging environments by leveraging advanced technologies and methodologies.

2
Objective
• The Objective is to employ YOLOv8 for precise military aircraft classification, bolstering situational
awareness and operational effectiveness in military contexts.
• Integrate alarm sound activation for immediate auditory alerts upon aircraft detection, enhancing response
readiness.
• Implement real-time email notification functionality to inform relevant personnel promptly about detected
aircraft, improving situational awareness.
• Ensure high precision in aircraft classification to minimize false alarms, maintaining system reliability.
• Optimize the system for rapid detection and notification, effectively monitoring military airspace and
enhancing overall security.

3
Application
• YOLOv8 deployed for instant classification of military aircraft, enhancing surveillance and response
capabilities.
• Seamless integration of object detection, alarm sounds, and email notifications improves situational
awareness and enables rapid decision-making in defense operations.
• Aircraft identification and notification processes streamlined, optimizing resource allocation and mission
readiness for military personnel.
• Revolutionize airspace security with automated classification and alerting systems, leveraging advanced deep
learning technology for enhanced effectiveness.
• Ensure timely detection and response to potential threats, safeguarding personnel and assets in dynamic
operational environments.

4
Problem Statement
• Develop a system using YOLOv8 for military aircraft classification. Upon detection, trigger alarm sounds for
immediate alerts and send email notifications with detailed aircraft information.
• Train YOLOv8 to recognize various military aircraft types accurately. Optimize the model to minimize false
positives and false negatives, enhancing real-time performance metrics such as detection speed and
notification latency.
• Validate the system's effectiveness through rigorous testing and evaluation scenarios. Ensure comprehensive
dataset collection and annotation for effective YOLOv8 model training. Ultimately, aim to bolster military
security and operational efficiency through timely aircraft detection and notification mechanisms.

5
Existing system
• Limited Real-Time Capabilities: Current military aircraft classification systems often lack the real-time
capabilities of deep learning models like YOLOv8, hindering response efficiency.
• Separate Alert Mechanisms: Alarm sound activation and email notification functionalities are typically
managed separately from aircraft classification systems, potentially causing delays in response times.
• Challenges in Accuracy: Legacy systems may struggle to accurately classify military aircraft in complex
backgrounds and varying lighting conditions, compromising surveillance effectiveness.
• Manual Annotation Requirements: Traditional systems often require extensive manual annotation and
tuning for effective aircraft classification, whereas YOLOv8 automates much of this process, reducing human
intervention.

6
Disadvantages of existing system
• Traditional systems often lack the real-time capabilities necessary for swift detection and classification of
military aircraft, unlike the rapid processing capabilities of YOLOv8.
• Legacy systems may struggle to accurately distinguish between different types of military aircraft amidst
complex backgrounds and environmental conditions, leading to higher false alarm rates
• Limited innovation and advancements in traditional systems may result in stagnation and decreased
effectiveness in addressing evolving security challenges.

7
Proposed system
• Real-Time Aircraft Classification: Integration of YOLOv8 enables swift and accurate detection of military
aircraft in real-time, enhancing surveillance capabilities.
• Immediate Audible Alerts: Alarm sound activation upon aircraft detection provides instant audible alerts to
personnel, improving response readiness.
• Timely Notification Dissemination: Email notification functionality ensures prompt dissemination of
aircraft detection information to designated recipients, facilitating timely response.
• Minimized False Alarms: YOLOv8's advanced object detection capabilities minimize false alarms and
optimize classification accuracy, enhancing system reliability.
• Streamlined Response Process: Seamless integration of alarm sound activation and email notification
functionalities streamlines the response process, improving overall system efficiency and effectiveness.

8
Advantage of proposed system
• YOLOv8's advanced object detection capabilities minimize false alarms and optimize classification accuracy.

• Seamless integration of alarm sound activation and email notification functionalities streamlines the response
process.

• Automation of aircraft classification with YOLOv8 reduces manual effort and speeds up system deployment.

• The proposed system offers a comprehensive solution for military surveillance, combining deep learning, audio
alerts, and email notifications.

9
 Hardware and software used

HARDWARE REQUIREMENTS:
SOFTWARE REQUIREMENTS:

• System : Pentium IV 2.4 GHz. • Operating system: Windows XP/7.

• Hard Disk : 40 GB. • Coding Language : python

• Floppy Drive : 1.44 Mb. • IDE : anaconda navigator

• Monitor : 15 VGA Colour.

• Mouse : Logitech.

• Ram : 512 Mb.

10
Modules
• DATA COLLECTION
• DATA PREPROCESSING
• TRAINING AND TESTING
• YOLO ALGORITHM STEPS
• ALARM SOUND ACTIVATION

11
Modules Description
Data Collection:
• Acquire a diverse dataset of military aircraft images from various sources, ensuring coverage of different
aircraft types, angles, and environments.
• Annotate the dataset with bounding boxes around the aircraft to provide ground truth labels for training.
Data Preprocessing:
• Resize images to a suitable input size for the YOLOv8 model.
• Normalize pixel values to a common scale to facilitate model training.
• Augment the dataset with techniques like rotation, flipping, and cropping to increase variability and
robustness.
Training and Testing:
• Split the dataset into training, validation, and testing sets.
• Train the YOLOv8 model on the training set using stochastic gradient descent (SGD) or another optimization
algorithm.
• Validate the model's performance on the validation set, adjusting hyperparameters as needed to improve
accuracy.
• Evaluate the final trained model on the testing set to assess its performance in real-world scenarios.
12
YOLO Algorithm Steps
• Input an image into the YOLOv8 model.
• Perform convolutional operations to extract features and generate feature maps.
• Apply object detection at multiple scales using anchor boxes and predict bounding boxes, objectness
scores, and class probabilities.
• Use non-maximum suppression (NMS) to filter out redundant bounding boxes and retain only the most
confident predictions.
• Output the final set of bounding boxes along with their corresponding class labels and confidence scores.
Alarm Sound Activation
• Upon detecting a military aircraft, trigger an alarm sound to provide immediate auditory alerts.
• Implement logic to determine the intensity and duration of the alarm based on the confidence score of the
detection.
• Integrate with existing alarm systems or generate custom alarm sounds for different types of detections.
Email Notification
• Configure the system to send email notifications upon aircraft detection.
• Include detailed information about the detected aircraft, such as its class label, confidence score, and
timestamp.
• Utilize SMTP protocols for secure and reliable email transmission, specifying recipients and subject lines
dynamically. 13
Data flow diagram

14
System Architecture

Resize images
to a suitable
input size
Data collection(military
Data
aircraft within captured Normalize pixel Training and Testing
Preprocessing
images or video frames.) values

Deep learning
Augment the
Model
dataset
Development

Yolo version
Result
8 algorithm

Streamlit no
server(fornte
nd) Alarm sound
yes and email
notification
15
Result
• The YOLOv8 model accurately classifies various types of military aircraft with high confidence scores.
• Alarm sound activation provides immediate auditory alerts upon aircraft detection, enhancing situational
awareness.
• Email notifications promptly disseminate detection information to designated recipients, facilitating rapid
response.
• The integrated system improves military surveillance and response capabilities, bolstering operational
readiness and security.

16
Results

17
Conclusion
• In conclusion, the integration of YOLOv8 for the classification of military aircraft, coupled with alarm sound
activation and email notification functionalities, represents a significant advancement in defense systems. By
harnessing the power of deep learning, real-time alerting mechanisms, and modern communication
technologies, this system offers a comprehensive solution for enhancing military surveillance, situational
awareness, and response capabilities.

• Overall, the integration of YOLOv8 for military aircraft classification, along with alarm sound activation and
email notification functionalities, enhances operational readiness and security in defense operations. This
innovative solution empowers military personnel with the tools and information needed to effectively monitor
airspace, identify threats, and respond swiftly to dynamic situations, ultimately contributing to mission
success and personnel safety.

18
Reference
• [1] R. Girshick, J. Donahue, T. Darrell, and J. Malik, “Rich feature hierarchiesfor accurate object detection and
semantic segmentation,” in Proceedingsof the IEEE conference on computer vision and pattern recognition,
2014,pp. 580–587.
• [2] R. Girshick, “Fast r-cnn,” in Proceedings of the IEEE international con-ference on computer vision, 2015, pp.
1440–1448.
• [3] S. Ren, K. He, R. Girshick, and J. Sun, “Faster r-cnn: Towards real-time object detection with region proposal
networks,” Advances in neuralinformation processing systems, vol. 28, 2015.
• [4] W. Liu, D. Anguelov, D. Erhan, C. Szegedy, S. Reed, C.-Y. Fu, and A. C.Berg, “Ssd: Single shot multibox
detector,” in European conference oncomputer vision. Springer, 2016, pp. 21–37.
• [5] J. Redmon, S. Divvala, R. Girshick, and A. Farhadi, “You only look once:Unified, real-time object detection,”
in Proceedings of the IEEE conferenceon computer vision and pattern recognition, 2016, pp. 779–788.
• [6] J. Redmon and A. Farhadi, “Yolo9000: better, faster, stronger,” in Proceed-ings of the IEEE conference on
computer vision and pattern recognition,2017, pp. 7263–7271.
• [7] J. Redmon and A. Farhadi, “Yolov3: An incremental improvement,” arXivpreprint arXiv:1804.02767, 2018.
• [8] A. Bochkovskiy, C.-Y. Wang, and H.-Y. M. Liao, “Yolov4: Optimal speedand accuracy of object detection,”
arXiv preprint arXiv:2004.10934, 2020.

19
• [9] D. Xu and Y. Wu, “Improved yolo-v3 with densenet for multi-scale remotesensing target detection,”
Sensors, vol. 20, no. 15, p. 4276, 2020.
• [10] G. Huang, Z. Liu, L. Van Der Maaten, and K. Q. Weinberger, “Denselyconnected convolutional
networks,” in Proceedings of the IEEE conferenceon computer vision and pattern recognition, 2017, pp.
4700–4708

20
THANK YOU

21