Visvesvaraya Technological University,

Belagavi, Karnataka

BLDEA's V.P. Dr. P. G. Halakatti

College of Engineering & Technology
Ashram Road, Adarsh Nagar, Vijayapura, Karnataka-586103

Department of Computer Science and Engineering

Computer Graphics &

Fundamentals of IP [21CS63]

MINI PROJECT REPORT

ON :
“IRIS SEGMENTATION”

Submitted By

1.Shravan Shivayogi [2BL21CS141]

2.Shreyas Benur [2BL21CS144]
3.Shreyas Keshapur [2BL21CS145]
4.Sujitkumar Chakrapani [2BL21CS159]

Under the Guidance of

Dr. Anand Jumnal

During

6th Semester 2023-2024

B.L.D.E.A.’s V.P. Dr. P.G. HALAKATTI COLLEGE OF ENGINEERING
AND TECHNOLOGY, VIJAYAPURA – 586 103
DEPARTMENT OF COMPUTER SCIENCE AND ENGINEERING

CERTIFICATE
This to certify that the mini project work of Computer Graphics &
Fundamentals of Image Processing [21CS63] entitled “IRIS
SEGMENTATION” is a bonafied work carried out by 1. Shravan Shivayogi
[2BL21CS141], 2.Shreyas Benur [2BL21CS144], 3.Shreyas Keshapur

[2BL21CS145], 4.Sujitkumar Chakrapani [2BL21CS159] during 6th

Semester of BE program. This report has been approved as it satisfies the
academic requirement in respect of mini project prescribed for Computer
Science and Engineering from Visvesvaraya Technological University,
Belagavi for the year 2023-24.

Dr. Anand Jumnal Dr. Leena Ragha Dr. V.G.Sangam

Guide HOD Principal

Examiners Signature with date

1.
2.
 DECLARATION

We, by 1. Shravan Shivayogi [2BL21CS141], 2. Shreyas Benur

[2BL21CS144], 3. Shreyas Keshapur [2BL21CS145], 4. Sujitkumar
Chakrapani [2BL21CS159] students of 6th semester, Department of
Computer Science and Engineering, B.L.D.E.A.’s V.P. Dr P.G
Halakatti College Of Engineering and Technology, Vijayapura declare
that the mini project titled “IRIS SEGMENTATION” has been
carried out by me and submitted for partial fulfillment of the 6th
Semester BE program Computer Science and Engineering from
Visvesvaraya Technological University, Belagavi for the year 2023-
24.
Abstract :
Iris segmentation is a crucial task in the field of computer graphics and image processing,
aiming to accurately delineate the iris region from eye images. This project leverages
advanced image processing techniques to enhance the precision and robustness of iris
segmentation, essential for applications such as biometric identification and ocular
disease diagnosis. By integrating state-of-the-art algorithms, including deep learning
models and traditional edge-detection methods, the system ensures high accuracy in
varying lighting conditions and image qualities.

The methodology involves preprocessing steps like noise reduction and contrast
enhancement, followed by segmentation using convolutional neural networks (CNNs).
These networks are trained on extensive datasets to recognize and isolate the iris region
effectively. Post-segmentation, the system performs morphological operations to refine
the boundaries and eliminate artifacts, ensuring the segmented iris is suitable for
subsequent analysis and identification processes.

This project also includes a performance evaluation component, where the segmented iris
images are compared against manually annotated ground truths. Metrics such as the Dice
coefficient and Intersection over Union (IoU) are used to quantify the accuracy and
reliability of the segmentation. The outcomes demonstrate significant improvements over
traditional methods, showcasing the potential of combining deep learning with
conventional image processing techniques in iris segmentation tasks.
 CONTENTS

Chapter No. Name Page No.

1. INTRODUCTION 1-2

2. FLOW CHART & DESIGN 3-7

ARCHITECTURE

3. IMPLEMENTATION & 8-9

DISCUSSIONS

4. RESULTS 10-12

5. CONCLUSION 13

6. REFERENCES 14
 Chapter 1:Introduction

Introduction

Introduction to Computer Graphics

Computer Graphics is a field of computer science that deals with generating images
with the aid of computers. It encompasses both the creation and manipulation of visual
content, spanning a wide range of applications including video games, movies,
simulations, and more. The key areas of computer graphics include:

 2D and 3D Modeling: Creating representations of objects in two or three

dimensions.

 Rendering: Converting models into images, which involves lighting, shading,

texture mapping, and other processes to achieve realism.

 Animation: Adding movement to models to create sequences of images.

 Human-Computer Interaction: Designing interfaces that allow users to interact

with graphic content.

OpenGL

OpenGL (Open Graphics Library) is a cross-language, cross-platform API for

rendering 2D and 3D vector graphics. It is widely used in applications where high-
performance rendering is necessary, such as video games, simulations, and CAD
(Computer-Aided Design) software. Key features of OpenGL include:

 Portability: Works across various platforms including Windows, macOS, and

Linux.

 Extensibility: Allows developers to add custom features through extensions.

 Efficiency: Optimized for high performance in real-time rendering applications.

Fundamentals of Image Processing

Image Processing involves performing operations on images to enhance them or extract

useful information. The fundamentals include:
  Image Acquisition: Capturing images from a source, such as a camera.

 Image Enhancement: Improving the visual appearance of an image, for example,

through contrast adjustment or noise reduction.

 Image Restoration: Correcting defects in an image, such as blurring.

 Image Compression: Reducing the size of image files for storage or transmission.

 Feature Extraction: Identifying and extracting important features from an image,

such as edges, textures, or shapes.

Iris Segmentation and Image Processing

Iris segmentation is a crucial technique in image processing, involving the precise

extraction of the iris region from eye images for various applications.

 Image Acquisition:

 High-resolution cameras capture detailed eye images, converted to digital format

for processing.

 Image Enhancement:

 Filters reduce noise, and contrast adjustment techniques like histogram

equalization improve clarity.

 Segmentation Techniques:

 Edge detection (e.g., Canny, Sobel) and thresholding methods separate the iris
from the rest of the eye.

 Feature Extraction and Analysis:

 Unique iris patterns are extracted for biometric identification and medical
diagnosis.

 Applications:

 Used in biometric identification for secure access, medical imaging for disease
detection, and security systems for enhanced recognition.
 Chapter 2: Flow Chart & Design Architecture
Flow Chart & Architecture Design for Iris Segmentation

Figure 1: Flow chart of 3D- Helicopter Game

Flowchart
This flowchart outlines the steps involved in an iris-based authentication system:

1. System Loads Iris Image from the File:

- The process begins with the system loading an iris image from a file. This is the input
stage where the iris image is captured and fed into the system for processing.

2. Image Processing:
- The loaded iris image undergoes image processing to enhance its quality. This
includes steps like noise reduction, contrast adjustment, and segmentation to isolate the
iris from other parts of the eye.

3. User Iris Code is Extracted from the Image:

- After processing, the system extracts the unique iris code from the image. This code
represents the distinct patterns and features of the iris, which are used for identification.

4. System Matches User Iris Code with Information Saved in Database:

- The extracted iris code is then compared with existing codes stored in a database. This
step involves matching the input iris code with the saved iris codes to find a match.

5. Decision:
- Based on the comparison, a decision is made. If the iris code matches an entry in the
database, the process proceeds to the next step. If not, the test fails.

6. Test Success:
- If the iris code matches successfully, the test is marked as successful. This indicates
that the user has been correctly identified.

7. Access is Granted:
- Once the test is successful, access is granted to the user. This is the final step where
the user is authenticated and allowed to proceed.
8. Test Failed:
- If the iris code does not match any entry in the database, the test fails, and access is
denied. The process loops back to the initial stage for another attempt.

Design Architecture Points

1. Input Stage:
- Captures and loads the iris image into the system.

2. Processing Stage:
- Enhances and processes the image to prepare it for feature extraction.

3. Feature Extraction Stage:

- Extracts unique iris patterns to create an iris code.

4. Matching Stage:
- Compares the extracted iris code with codes stored in the database to find a match.

5. Decision-Making Stage:
- Determines whether the user is authenticated based on the matching result.

This architecture ensures a systematic and reliable approach to iris-based authentication,

combining image processing and pattern recognition to verify user identity accurately.
Chapter 3: Implementation Modules

Implementation Modules for Iris Segmentation

 Image Acquisition Module:

 Camera Interface: Connects to high-resolution cameras to capture eye images.

 Image Loading: Reads and loads images from files or live camera feeds.

 Preprocessing Module:

 Noise Reduction: Applies filters like Gaussian or median to remove noise.

 Contrast Enhancement: Uses techniques such as histogram equalization to
improve image contrast.
 Normalization: Standardizes the intensity values to a uniform scale.

 Segmentation Module:

 Iris Localization: Identifies and isolates the iris region using methods like
circular Hough Transform.
 Edge Detection: Employs algorithms such as Canny or Sobel to detect edges
around the iris.
 Thresholding: Separates the iris from the rest of the eye using binary thresholding
techniques.

 Feature Extraction Module:

 Pattern Recognition: Extracts unique patterns from the segmented iris, focusing
on textures and ridges.
 Feature Vector Generation: Converts the extracted patterns into a feature vector
representing the iris code.
 Post-Processing Module:

 Morphological Operations: Refines the segmented iris region using operations

like dilation and erosion to remove small artifacts.
 Boundary Smoothing: Smoothens the edges of the segmented iris to enhance
accuracy.

 Database Module:

 Storage: Saves the extracted iris codes and corresponding user information.
 Retrieval: Provides fast and efficient access to stored iris codes for matching.

 Matching and Decision Module:

 Comparison Algorithm: Compares the extracted iris code with codes in the
database using similarity measures like Hamming distance.
 Decision Logic: Determines the authentication result based on the comparison,
deciding whether to grant or deny access.

 Evaluation Module:

 Performance Metrics: Uses metrics such as Dice coefficient, Intersection over

Union (IoU), and accuracy to evaluate segmentation performance.
 Testing and Validation: Validates the system using a dataset of annotated eye
images to ensure reliability and accuracy.
Chapter 4: Results

The 3D helicopter game achieves high visual fidelity through detailed modeling and
texturing, dynamic lighting, and realistic shading. The rendering quality is complemented by
smooth performance, maintaining a consistent frame rate of 60 FPS, and efficient resource
management that minimizes load times. User experience is enhanced by an intuitive and
responsive interface, providing immediate feedback for user actions and seamless transitions
between game states. Advanced graphical techniques, including shaders and physics-based
rendering, create realistic effects such as dynamic lighting, shadows, and environmental
interactions. The game’s physics engine ensures fluid and accurate helicopter movements,
adding to the realism and challenge. Overall, the game demonstrates successful integration of
computer graphics principles, resulting in an immersive and engaging gameplay experience.

Figure 2: Output-1 of 3D- Helicopter Game

The image depicts a screenshot of a simple 3D helicopter game. The game interface shows a
brightly colored helicopter flying through a landscape divided into a blue sky and a green
ground. The helicopter is positioned in the center, with propellers depicted as white and the
body in orange and blue colors. In the upper right corner of the screen, the current score is
displayed as "0," and the game level is shown as "02." The background is minimalistic, with
the sky and ground being solid colors, and there are circular obstacles ahead of the helicopter,
indicating the player needs to navigate through them. The game appears to be in an early
stage, given the score and level information. The bottom of the image shows the Windows
taskbar with various application icons and the time and date.

Figure 2: Output-2 of 3D- Helicopter Game

The image shows an updated screenshot of the 2D helicopter game. The game
interface is largely the same, featuring a helicopter navigating through a landscape divided
into a blue sky and green ground. This time, the helicopter is about to pass through a large
purple ring, indicating a game obstacle or goal. The score displayed in the upper right corner
has increased to "1," suggesting that the player has successfully navigated through one or
more obstacles. The level remains at "02." The simplistic graphics continue with the
helicopter's body in orange and blue, and its propellers in white. The Windows taskbar is still
visible at the bottom, with various application icons and the date and time showing. The
visual elements of the game suggest it is in progress and that the player is making progress by
passing through the rings.
 Figure 2: Keys used for 3D- Helicopter Game

The image shows the title screen of the helicopter game with instructions and credits. The
background is a solid blue color. At the top, the title "HELICOPTER GAME" is
displayed. Below the title, the credits list the names of the developers: Shruti Jadhav,
Sindhu Bardiger, Suma Borannavar, and Sunayana Deshpande, each followed by their
respective identification numbers. The instructions for controlling the game are listed at
the bottom: pressing "SPACE" to continue, "8" for up, "2" for down, "4" for left, "6" for
right, and "ESC" to exit. The simplicity of the screen indicates it is designed to provide
clear and direct information to the player before starting or continuing the game. The
Windows taskbar is visible at the bottom, indicating the application is running on a
Windows operating system.
 Chapter 5: Conclusion

Iris segmentation in image processing is a pivotal technology that underpins a variety of

critical applications, particularly in biometric identification and medical diagnostics. By
isolating the iris region accurately, the technology ensures high precision and reliability in
identifying individuals based on their unique iris patterns. Advanced preprocessing
techniques, such as noise reduction and contrast enhancement, set the foundation for
effective segmentation. Sophisticated algorithms for edge detection and thresholding
facilitate precise delineation of the iris, while feature extraction and post-processing steps
further refine the results. The integration of robust comparison algorithms and decision
logic ensures that the extracted iris codes can be matched accurately against stored data,
enabling secure authentication. Evaluation metrics help in fine-tuning the system for
optimal performance. Overall, iris segmentation stands out as a robust and efficient
solution in the field of image processing, enhancing security and providing valuable
insights for medical applications.
 Chapter 6: References

BIBILIOGRAPH
Y
1. www.w3schools.com
2. in.php.net
3. en.wikipedia.org/wiki/PHP
4.
www.hotscripts.com/category/
php/
5. www.apache.org/
6. www.mysql.com/click.php?
e=35050
BIBILIOGRAPH
Y
1. www.w3schools.com
2. in.php.net
3. en.wikipedia.org/wiki/PHP
4.
www.hotscripts.com/category/
php/
5. www.apache.org/
6. www.mysql.com/click.php?
e=35050
BIBILIOGRAPH
Y
1. www.w3schools.com
2. in.php.net
3. en.wikipedia.org/wiki/PHP
4.
www.hotscripts.com/category/
php/
5. www.apache.org/
6. www.mysql.com/click.php?
e=35050
1. www.w3schools.com

2.https://drive.google.com/drive/folders/1JT9cw6qAxzbB5xazU5oyEM2
MGXSMB48b