Professional Documents
Culture Documents
Gnana Padmesh Internship Report
Gnana Padmesh Internship Report
Gnana Padmesh Internship Report
AN INTERNSHIP REPORT
Submitted by
of
BACHELOR OF TECHNOLOGY
in
INFORMATION TECHNOLOGY
APRIL 2024
SRI SAI RAM ENGINEERING COLLEGE
(An Autonomous Institution; Affiliated to Anna University, Chennai -600 025)
BONAFIDE CERTIFICATE
Our sincere gratitude to our beloved Founder Chairman Shri. MJF. Ln.
LEO MUTHU for his great endeavors in establishing this institution and
standing as a figure of guidance.
Our heartfelt thanks to our Chairman and CEO Dr. Sai Prakash Leo Muthu
to provide industrial interaction to the faculty members and students.
We wish to express our gratefulness and gratitude to our beloved Head of the
Department, Dr. Sheela T for her encouragement, support, and guidance.
A Special thanks to all the heads of the departments, HR, and all employees
of Department of Information Technology for their support for imparting
knowledge about the industry and various processes involved in the industry
throughout the training.
DETAILS OF INTERNSHIP
ZEBU
01.06.2023
INTELLIGENT
I Internship OFFLINE TO III / IV
SYSTEMS PVT.
15.07.2023
LTD.
CHAPTER NO TITLE PG NO
1 INTRODUCTION 1
2 OBJECTIVES 2
3 RESPONSIBILITIES – SKYEYE 3
During my internship at Zebu Intelligent Systems Pvt. Ltd, I was delved in a dynamic
learning environment aimed at meeting industry demands while fostering innovation
and practical knowledge application. Leading the "SkyEye" project, I delved deep
into the implementation of the YOLOv7 algorithm for custom object detection of
ships and drones. This involved rigorous exploration of computer vision techniques
and algorithm optimization, honing my skills in machine learning and image
processing.
The 45-day intensive training, spanning from June 1,2023 to July 15, 2023, was
pivotal in deepening my understanding of object detection algorithms and refining
my programming skills. Collaborating on the "SkyEye" project allowed me to
leverage my strengths in problem-solving and teamwork, while also expanding my
knowledge of project management methodologies. My internship experience has
been instrumental in shaping my professional growth and preparing me for future
endeavors in the field of computer vision and AI. By contributing to the development
of the "SkyEye" project, I have gained practical expertise in custom object detection
and made meaningful strides towards advancing sustainable solutions for maritime
and land surveillance.
1
2. OBJECTIVES
During the 45-day tenure at Zebu Intelligent Systems Pvt. Ltd. company, as part of
the Innovative4Defence (i4D) program under the Innovation for Defence Excellence
(iDEX), Defense Innovation Organization (DIO), the primary objective is to
immerse interns in adynamic learning environment focused on defense innovation.
The internship is designedto provide hands-on experience and practical exposure to
cutting-edge technologies within the defense sector.
Interns will undergo rigorous skill development exercises, honing expertise in areas
such as artificial intelligence, computer vision, robotics, and sensor technologies.
Through active participation in projects related to defense innovation, interns will
apply their newly acquired knowledge to tackle real-world challenges.
2
3. RESPONSIBILITIES
The project aims to develop a robust object detection system using the YOLOv7
algorithm to detect and classify drones and ships in aerial images or video footage. The
system will be capable of real-time detection, enabling applications such as maritime
surveillance, border control, and security.
The following are the steps to be carried out in the project:
1. Data Collection
2. Data Annotation
3. YOLOv7 Model Selection
4. Model Training
5. Evaluating the model
6. Testing the model
7. Performance Analysis
8. Deployment
The project aims to provide a reliable and efficient solution for detecting and
classifying drones and ships in aerial imagery or video footage. By leveraging the
YOLOv7 algorithm's capabilities, the system can enhance situational awareness, support
decision- making processes, and contribute to various applications related to surveillance,
security, and monitoring.
3.2 PROPOSED SOLUTION
The proposed system for drone and ship detection using the YOLOv7 algorithm focuses
on real-time detection and includes a robust alert system to notify relevant authorities or
3
personnel. The system comprises the following key components:
Data Acquisition: Gather aerial images or video footage from various sources, such as
drones, surveillance cameras, or satellite imagery. Ensure the dataset contains adverse
range of scenarios, lighting conditions, and backgrounds.
Data Annotation: Annotate the collected data by manually labelling bounding boxes
aroundthe drones and ships in the images or frames. Assign appropriate class labels to
the annotated objects.
YOLOv7 Model Training: Configure and train the YOLOv7 model using the annotated
dataset. Fine-tune the model to optimize its performance for drone and ship detection.
Experiment with different hyperparameters to improve accuracy and reduce false
positivesand false negatives.
Real-Time Object Detection: Implement the trained YOLOv7 model for real-time object
detection on live video feeds or aerial imagery. Utilize efficient techniques such as video
stream processing and parallel computing to achieve real-time performance.
Visual Display and Tracking: Provide a visual display interface that overlays the detected
drones and ships on the live video feed or aerial imagery in real-time. Track the movement
of the objects and display trajectory information to aid in situational awareness and
monitoring.
Deployment and Integration: Deploy the system in the desired application environment,
considering hardware requirements and compatibility. Integrate the system with existing
surveillance infrastructure or other relevant systems, such as command centres or security
management platforms.
The proposed system aims to provide an efficient, accurate, and real-time drone and ship
detection solution using the YO LOv7 algorithm.
By incorporating a real-time alert system and visual display interface, the system
enablesprompt response to potential threats, enhances situational awareness.
4
3.3 MODULAR DESCRIPTION
The project of drone and ship detection using the YOLOv7 algorithm can be organized
into several modular components, each serving a specific purpose within the system.
Hereis a modular description of the project:
MODULE 1:
Data Acquisition:
Module Description: This module focuses on acquiring the necessary aerial imagery or
live video feeds containing drones and ships.
Functionality: Capture aerial images or video streams using cameras, drones, etc.,
Input: Aerial imagery or live video feeds.
Output: Raw imagery or video data for further processing.
MODULE 2:
Data Preprocessing:
Module Description: This module prepares the acquired data for subsequent processing
and analysis.
Functionality: Perform necessary preprocessing steps such as resizing, normalization,
and enhancement to optimize the data for detection.
Input: Raw imagery or video data.
Output: Pre-processed data ready for detection.
MODULE 3:
YOLOv7 Detection:
Module Description: This module implements the YOLOv7 algorithm for drone and
ship detection.
Functionality: Utilize the YOLOv7 model to perform object detection on the pre-
processed data. Identify and localize drones and ships in the imagery or video frames.
Input: Pre-processed data.
5
Output: Bounding box coordinates and class labels of detected drones and ships.
MODULE 4:
Tracking and Monitoring:
Module Description: This module tracks the detected drones and ships over time to
monitor their movements and trajectories.
Functionality: Apply tracking algorithms to associate detected objects across
consecutive frames and estimate their paths.
Input: Detected objects from the YOLOv7 module.
Output: Tracked trajectories of drones and ships.
6
4 LEARNING OUTCOMES
7
Fig 3 - Annotation
Fig 4 - Labelling
8
Fig 6: Training
Fig 7: Training
9
Fig 7 – Testing
10
5. CONCLUSION
Our internship experience has not only sharpened our technical skills but also
reinforced the importance of collaboration and innovation in software development.
By actively contributing to the design, coordination, and ideation processes, we
have demonstrated our ability to deliver impactful solutions that address real-world
challenges. Moving forward, we are confident that the skills and knowledge gained
during this internship will serve as a strong foundation for our continued growth and
success in the field of software development.
11
INTERNSHIP CERTIFICATE
12
DEPARTMENT OF INFORMATION TECHNOLOGY
Proforma for Evaluation of Internship
4 Presentation Presentation 10
Technical 10
Content
Answering 10
Questions
Quality of 5
Report
Evaluator 1 Evaluator 2
Evaluator 3 Evaluator 4
Chairman
13
STUDENT INTERNSHIP FEEDBACK FORM
Student Name : Gnana Padmesh C K
Student ID No : SEC20IT020
Register No 412520205040
Department / Class / Section : IT / 4 / A
Academic Year IV
Company Name : ZEBU INTELLIGENT SYSTEMS PVT. LTD,
CIN/TIN/PAN : U72900TG2021PTC147559
GST No : NIL
Contact Person Name : Mr. JAGADEESHWARAN THANGARAJAN
Mobile No : NIL
Area (Key Words) : AI/ML
Internship Mode : Offline
Internship (Free / Paid) : Free
Projects Completed during
Internship : Yes - SKYEYE
Beginning of Internship : 01.06.2023
End of Internship : 15.07.2023
🗸
My supervisor was helpful and 5
2 1 2 3 4
🗸
provided task feedback
Student Signature
14
INTERNSHIP II
15
CONTENETS
CHAPTER NO TITLE PG NO
1 INTRODUCTION 17
2 OBJECTIVES 18
16
1. INTRODUCTION
17
2. OBJECTIVES
18
3. PROJECT – CISCO PACKET TRACER
One of the standout features of Cisco Packet Tracer is its user-friendly interface,
making it accessible to individuals at various skill levels. The drag-and-drop
functionality, coupled with a rich library of Cisco devices and components,
empowers users to construct intricate network setups effortlessly. This intuitive
design promotes active learning, enabling users to experiment with different
configurations and scenarios to reinforce theoretical knowledge.
Cisco Packet Tracer's versatility extends beyond the educational realm, proving
invaluable for professionals seeking to test and validate network designs before
implementation. The software supports a wide array of Cisco devices, making it a
comprehensive tool for modelling different network architectures. This aids in
minimizing the risk of errors and downtime in real-world deployments, ultimately
saving time and resources.
19
PACKET TRACER CONFIGURATION
20
4. LEARNING OUTCOMES
21
5. CONCLUSION
22
INTERNSHIP CERTIFICATE
23
DEPARTMENT OF INFORMATION TECHNOLOGY
Proforma for Evaluation of Internship
YEAR / SECTION: IV / A
4 Presentation Presentation 10
Technical 10
Content
Answering 10
Questions
Quality of 5
Report
Evaluator 1 Evaluator 2
Evaluator 3 Evaluator 4
Chairman
24
STUDENT INTERNSHIP FEEDBACK FORM
Student Name : GNANA PADMESH C K
Student ID No : SEC20IT020
Register No 412520205040
Department / Class / Section : IT / 4 / A
Academic Year IV
Company Name : CISCO – AICTE VIRTUAL INTERNSHIP
CIN/TIN/PAN : U72200KA2011FTC056691
GST No : 33AABCC0258Q1ZL
Contact Person Name :-
Mobile No :-
Area (Key Words) : Networking
Internship Mode : Online
Internship (Free / Paid) : Free
(If Paid Internship, Specify the Amount)
Projects Completed during
Internship : Yes – Design and Configuration of Packet Tracer
(If Yes Kindly Specify the Project Title)
Beginning of Internship : 01.08.2023
End of Internship : 30.08.2023