A Project Report

On

‘‘BLOOD GROUP DETECTION USING

FINGERPRINT PATTERN’’
By :-

Gyanendriya Mishra (2201214037)

Under the Guidance of:-

Mr. LINGARAJ PANIGRAHY

Asst Prof. Department of Computer Science & Engineering

KONARK INSTITUTE OF SCIENCE & TECHNOLOGY

TECHNO PARK, JATANI, BHUBNESWAR – 752050, ODISHA

Department of Computer Science & Engineering

[2025-2026]

1
 KONARK INSTITUTE OF SCIENCE &TECHNOLOGY

CERTIFICATE

This is to certify that Gyanendriya Mishra, bearing registration

number 2201214037, has successfully completed the project work
entitled “BLOOD GROUP DETECTION USING FINGERPRINT
PATTERN” under the supervision, in the partial fulfilment of
Bachelor’s of Technology – Computer science & Engineering under
Biju Patnaik University of Technology, Rourkela.

Date:

Place:

Guide’s Name: Internal Examiner:

Head Department: External Examiner:

Principal:

2
 ACKNOWLEDGEMENT
I take this opportunity to express our heartfelt thanks to all those who
individually as well as collectively helped us in the successful
completion of this project.

I would like to take this opportunity to express my honour, respect, deep

gratitude, and genuine rewards to our Guide Prof.
Mr. LINGARAJ PANIGRAHY, Assistant Professor Computer
Science Engineering Department for giving us all the guidance required
for our topic, apart from being a constant source of inspiration and
motivation.

I are grateful to Prof. Dr. L. N. Padhy, Head of Department, Computer

Science and Engineering Department, for providing the necessary help
and encouragement whenever needed.
I also extend our sincere thanks to all the Computer Science
Engineering Department faculty members for their support and
encouragement.

I special thanks to our parents for their moral support and warm wishes,
and finally, We would express our appreciation to all our friends for
their support which helped me to complete our project successfully.

3
 DECLARATION
I, hereby declare that the project work entitled “BLOOD GROUP
DETECTION USING FINGERPRINT PATTERN”, is being
submitted in partial fulfillment for the award of Bachelor of Technology
in Computer Science Engineering from Konark Institute of Science
& Technology, affiliated to Biju Patnaik University of Technology, is
work carried out by us.

I further declare that no part of it has been submitted for the award,
degree, diploma, or similar title for any other Examining Body,
University, or Institute previously.

Gyanendriya Mishra (2201214037)

4
 ABSTRACT
Blood group identification is a critical requirement in medical
emergencies, blood transfusions, and healthcare services.
Traditional methods for determining blood groups involve invasive
procedures that require blood samples and laboratory analysis,
making them time-consuming and dependent on medical
infrastructure. This project proposes a non-invasive and efficient
approach for preliminary blood group detection using fingerprint
patterns.

The system utilizes fingerprint images as input and applies image

processing techniques to enhance and extract important features
such as ridge patterns and textures. These features are then
analyzed using machine learning algorithms to classify and predict
the blood group. The application is developed as a web-based
system using Python and the Flask framework, with a user-friendly
interface designed using HTML, CSS, JavaScript & Python.

The proposed system aims to provide a fast, cost-effective, and

accessible solution for blood group prediction, especially in
emergency situations and remote areas. Although the system
cannot replace conventional laboratory testing methods, it
demonstrates the potential of integrating biometric analysis with
machine learning in healthcare applications. Future improvements
can enhance accuracy and expand its real-world usability.

5
 Table of Contents:
Content Page NO.

01 Introduction 07

02 Overview related works 08

03 System analysis 09

04 Purposed feature 10-11

05 System Architecture 12-13

06 Tools and methodology used 14-15

07 Code 16-19

08 Experimental results 20-21

09 Screen Short 22-23

10 Challenges 24-25

11 Conclusion 26

12 Future enhancement 27-28

13 Reference 29

6
CHAPTER-1
INTRODUCTION :-
Blood group identification plays a vital role in medical emergencies,
blood transfusions, surgeries, and healthcare services. Accurate and
timely detection of blood groups can save lives, especially in critical
situations. Traditionally, blood group detection is performed using
laboratory methods that require collecting blood samples and
conducting serological tests. Although these methods are reliable, they
are invasive, time-consuming, and require proper medical facilities and
trained personnel.

With the advancement of technology, there is a growing need for non-

invasive and faster methods for preliminary blood group identification.
Biometric systems, particularly fingerprint recognition, have gained
significant importance due to their uniqueness, reliability, and ease of
use. Fingerprints are genetically determined and remain unchanged
throughout a person’s life. Similarly, blood groups are also inherited
traits. This has led researchers to explore a possible correlation
between fingerprint patterns and blood groups.

In this project, a system is developed to predict blood groups using

fingerprint patterns. The system uses image processing techniques to
enhance and analyze fingerprint images, extracting important features
such as ridge patterns and texture details. These features are then
provided to a machine learning model, which has been trained to
classify and predict the blood group based on the input data.

The proposed system is implemented as a web application using Python

and the Flask framework for backend processing. The frontend is
designed using HTML, CSS, and JavaScript to provide a user-friendly
interface for uploading fingerprint images and displaying results.

7
CHAPTER-2
RELATED WORK :-
Several research studies have been conducted to analyze the
relationship between fingerprint patterns and blood groups.
Fingerprints are classified into three main types: loops, whorls, and
arches. Researchers have observed that these patterns may have some
correlation with different blood groups, as both are genetically
inherited characteristics.

Earlier studies mainly focused on statistical analysis of fingerprint

patterns. Researchers collected fingerprint samples from individuals
with known blood groups and analyzed the frequency of each pattern.
The results showed that certain fingerprint types were more common
in specific blood groups. However, these studies were mostly manual
and limited in scope.

With the advancement of technology, image processing techniques

have been introduced to improve fingerprint analysis. Tools like
OpenCV are used to enhance fingerprint images, remove noise, and
extract important features such as ridge structure and texture. This has
made fingerprint analysis more accurate and efficient.

In recent years, machine learning approaches have been applied to

this problem. Algorithms such as Support Vector Machine (SVM), K-
Nearest Neighbors (KNN), and Random Forest have been used to
classify fingerprint patterns and predict blood groups. These methods
have shown better performance compared to traditional statistical
approaches, especially when trained with large datasets.

8
CHAPTER-3
SYSTEM ANALYSIS :-
System analysis is the process of understanding the existing system,
identifying its limitations, and proposing an improved system to
overcome those issues. In this project, both the existing system and
the proposed system are analyzed.
🔹 Existing System
The traditional method of blood group detection is based on
laboratory testing. It involves collecting a blood sample from the
individual and analyzing it using chemical reagents.
Limitations of Existing System:
 Requires blood sample (invasive method)

 Time-consuming process

 Needs laboratory setup and equipment

 Requires trained medical professionals

 Not suitable for quick use in emergencies

 Difficult to access in rural or remote areas

🔹 Proposed System
The proposed system introduces a non-invasive method for predicting
blood groups using fingerprint patterns. It uses image processing and
machine learning techniques to analyze fingerprint features and
predict the blood group through a web-based application.
Features of Proposed System:
 No need for blood sample

 Fast and automated prediction

 Web-based and easy to access

 Uses fingerprint image as input

 Machine learning-based classification

 User-friendly interface

9
CHAPTER-4
PROPOSED FEATURES :-
The proposed system “Blood Group Detection Using Fingerprint Pattern”
introduces several innovative features to provide a non-invasive and
efficient method for preliminary blood group prediction. The key features
of the system are as follows:

🔹 Fingerprint-Based Input
 The system accepts fingerprint images as input instead of blood

samples.
 Provides a non-invasive method for prediction.

🔹 Image Preprocessing
 Uses image processing techniques to:

o Convert images to grayscale

o Remove noise

o Enhance fingerprint clarity

 Ensures better accuracy in feature extraction.

🔹 Feature Extraction
 Extracts important fingerprint features such as:

o Ridge patterns

o Texture details

o Minutiae points

 Converts image data into a format suitable for machine learning.

🔹 Machine Learning Prediction

 Uses a trained ML model (SVM/KNN/Random Forest)

 Predicts blood group (A, B, AB, O & others)

 Provides quick and automated results

10
🔹 Web-Based Application
 Developed using Flask framework

 Accessible through any web browser

 No need for software installation

🔹 User-Friendly Interface
 Simple and interactive UI

 Easy fingerprint upload option

 Clear display of results

🔹 Fast and Real-Time Output

 Provides instant prediction results

 Saves time compared to traditional methods

🔹 Data Handling (Optional)

 Stores user inputs and prediction results

 Can be used for analysis and improvement

🔹 Scalable System Design

 Can be upgraded with:

o Deep learning models

o Mobile applications

o Cloud integration

11
CHAPTER-5
System Architecture :-
The system architecture of the project “Blood Group Detection Using
Fingerprint Pattern” defines how different components interact to
perform blood group prediction using fingerprint images. The system
is designed as a web-based application that integrates frontend,
backend, image processing, and machine learning modules.
🔹 Main Components of the System
1. User Interface (Frontend)
 Developed using HTML, CSS, and JavaScript

 Allows users to upload fingerprint images

 Displays predicted blood group results

 Provides a simple and user-friendly interaction

2. Web Server (Backend - Flask)

 Developed using Python Flask framework

 Handles user requests and responses

 Receives uploaded fingerprint images

 Sends data to image processing and ML modules

 Returns prediction results to frontend

3. Image Processing Module

 Implemented using OpenCV

 Performs:

o Grayscale conversion

o Noise removal

o Image enhancement

 Extracts important fingerprint features such as ridge patterns and

textures
4. Machine Learning Model
 Built using Scikit-learn

 Trained on fingerprint dataset with known blood groups

 Classifies input fingerprint features

 Predicts blood group (A, B, AB, O & others)

12
5. Database (Optional)
 Stores:

o Fingerprint data

o Prediction results

o Model information

 Can be implemented using SQLite/MySQL

13
CHAPTER 6:
TOOLS AND METHOLOGY USED :-
This project uses a combination of software tools and a systematic
methodology to develop a web-based system for blood group prediction
using fingerprint patterns.
🔧 Tools Used
1. Programming Language
 Python

o Used for backend development and machine learning

implementation
o Easy to integrate with image processing and web frameworks

2. Web Framework
 Flask

o Lightweight Python framework

o Handles server-side logic

o Connects frontend with backend processing

3. Frontend Technologies
 HTML

o Structure of web pages

 CSS

o Styling and layout design

 JavaScript

o Adds interactivity and dynamic behavior

4. Image Processing Library

 OpenCV

o Used for image preprocessing

o Performs grayscale conversion, noise removal, and enhancement

o Helps in feature extraction from fingerprint images

5. Machine Learning Library

 Scikit-learn

o Used to train and test machine learning models

o Implements algorithms like SVM, KNN, Random Forest

14
6. Supporting Libraries
 NumPy

o Used for numerical operations and data handling

7. Database (Optional)
 SQLite / MySQL

o Used for storing user data and prediction results

🔬 Methodology Used
1. Data Collection
 Collect fingerprint images with known blood group labels

 Dataset is used for training the machine learning model

2. Image Preprocessing
 Convert fingerprint image to grayscale

 Remove noise using filters

 Enhance image quality for better analysis

3. Feature Extraction
 Extract important fingerprint features:

o Ridge patterns

o Texture details

o Edges and contours

 Convert image into numerical data

4. Model Training
 Train machine learning model using dataset

 Algorithms used:

o SVM / KNN / Random Forest

 Save trained model

5. Prediction
 Input fingerprint image is processed

 Extracted features are given to trained model

 Model predicts blood group

6. Result Display
 Predicted blood group is displayed on web page

 User gets instant output

15
CHAPTER-7 Code:- [Link]

16
17
Css:-

18
JS:-

19
CHAPTER -8
EXPERIMENTAL RESULT :-
The experimental results demonstrate the performance and
effectiveness of the proposed system “Blood Group Detection Using
Fingerprint Pattern”. The system was tested using a set of fingerprint
images with known blood group labels to evaluate its prediction
capability.

🔹 Testing Process

 Fingerprint images were collected and used as input to the

system

 Images were preprocessed using OpenCV techniques

 Features were extracted and passed to the trained machine

learning model

 The model predicted the blood group for each input image

🔹 Observations

 The system successfully processed fingerprint images and

generated predictions

 The response time for prediction was fast

 The system performed better with clear and high-quality

fingerprint images

 Accuracy improved when trained with a larger dataset

20
🔹 Sample Results (With respective Rh factor)
Predicted
Fingerprint Image Actual Blood Group
Blood Group

Sample 1 A A

Sample 2 B B

Sample 3 O O

Sample 4 AB AB

🔹 Performance Analysis

 The system achieved moderate accuracy depending on dataset

size

 Errors occurred due to:

o Poor image quality

o Limited training data

o Similar fingerprint patterns across classes

21
CHAPTER-9
Screen short:-

22
23
CHAPTER-10
CHALLENGES :-
During the development of the project “Blood Group Detection Using
Fingerprint Pattern”, several challenges were encountered. These
challenges mainly relate to data availability, image quality, and model
performance.

🔹 Data Collection Issues

 Difficulty in obtaining a large and well-labeled dataset of

fingerprint images with corresponding blood groups

 Limited publicly available datasets for this specific problem

🔹 Image Quality Problems

 Fingerprint images may contain:

o Noise

o Blurriness

o Poor contrast

 Low-quality images affect feature extraction and prediction

accuracy

🔹 Feature Extraction Complexity

 Extracting accurate fingerprint features such as ridge patterns

and minutiae points is challenging

 Requires proper preprocessing techniques

24
🔹 Model Accuracy Limitations

 Machine learning models may not achieve high accuracy with

small datasets

 Similar fingerprint patterns can lead to incorrect predictions

🔹 Computational Requirements

 Image processing and model execution require sufficient system

resources

 Processing large datasets can be time-consuming

🔹 Lack of Direct Correlation

 Scientific correlation between fingerprint patterns and blood

groups is not strongly established

 Makes prediction less reliable compared to medical tests

🔹 Integration Challenges

 Combining frontend, backend, image processing, and ML model

smoothly

 Handling file uploads and real-time prediction

25
CHAPTER-11
CONCLUSION :-
The project “Blood Group Detection Using Fingerprint Pattern”
presents a novel and non-invasive approach for preliminary blood
group identification using biometric data. The system successfully
integrates image processing and machine learning techniques within a
web-based application to analyze fingerprint patterns and predict
blood groups.

By using tools such as Python, Flask, OpenCV, and Scikit-learn, the

system is able to process fingerprint images, extract relevant features,
and generate predictions efficiently. The web interface provides an
easy and user-friendly way for users to upload fingerprint images and
obtain results in real time.

The experimental results show that the system can perform blood
group prediction with moderate accuracy, depending on the quality
and size of the dataset. Although the system cannot replace traditional
laboratory testing methods, it demonstrates the potential of using
fingerprint analysis as a supportive tool in healthcare, especially in
emergency situations and areas with limited medical facilities.

In conclusion, this project highlights how the combination of biometric

technology and machine learning can be used to develop innovative
solutions in the healthcare domain. With further improvements in
data quality, model accuracy, and system design, the proposed system
can be enhanced for better performance and real-world applications.

26
CHAPTER-12
FUTURE ENHANCEMENT :-
The project “Blood Group Detection Using Fingerprint Pattern” has
significant scope for improvement and expansion. Although the
current system provides a basic and functional model, several
enhancements can be implemented to improve accuracy, usability,
and real-world applicability.

🔹 Use of Deep Learning Models

 Implement advanced models like Convolutional Neural

Networks (CNN)

 Improve accuracy by directly analyzing fingerprint images

 Handle complex patterns more effectively

🔹 Larger and Better Dataset

 Collect a large-scale, high-quality dataset

 Include diverse fingerprint samples

 Improve model training and prediction accuracy

🔹 Real-Time Fingerprint Scanner Integration

 Integrate hardware fingerprint scanners

 Enable real-time image capture instead of manual upload

🔹 Mobile Application Development

 Develop Android/iOS application

 Make system more accessible and portable

27
🔹 Cloud Deployment

 Deploy system on cloud platforms

 Allow remote access and scalability

 Store and manage large datasets efficiently

🔹 Multi-Biometric Integration

 Combine fingerprint with:

o Face recognition

o Iris scanning

 Improve prediction reliability

🔹 Improved Image Processing Techniques

 Use advanced preprocessing methods

 Enhance fingerprint clarity and feature extraction

🔹 Inclusion of Detailed Blood Group Factors

 Extend system to predict Rh factor (+ / –)

 Provide more detailed blood classification

🔹 AI-Based Continuous Learning

 Implement self-learning models

 Improve accuracy over time with new data

28
 REFERENCES:
◻ Jain, A.K., Ross, A., & Prabhakar, S. (2004). An Introduction to
Biometric Recognition. IEEE Transactions on Circuits and Systems
for Video Technology.
◻ Maltoni, D., Maio, D., Jain, A.K., & Prabhakar, S. (2003).
Handbook of Fingerprint Recognition. Springer.
◻ Gonzalez, R.C., & Woods, R.E. (2018). Digital Image Processing
(4th Edition). Pearson Education.
◻ Pedregosa, F., Varoquaux, G., Gramfort, A., et al. (2011). Scikit-
learn: Machine Learning in Python. Journal of Machine Learning
Research.
◻ OpenCV Documentation. Open Source Computer Vision Library.
Available at: [Link]
◻ Flask Documentation. Flask Web Development Framework.
Available at: [Link]
◻ Research papers on fingerprint pattern analysis and blood
group correlation from various international journals.
◻ NumPy Documentation. Numerical Computing in Python.
Available at: [Link]

29