International Research Journal of Engineering and Technology (IRJET) e-ISSN: 2395-0056

Volume: 11 Issue: 03 | Mar 2024 www.irjet.net p-ISSN: 2395-0072

Blood group detection through finger print images using image

processing (KNN)
G. Mounika1, M. Anusha2, D. Gopika3, B. Siva kumari4

1st Author -B. Tech 4th Year ,Dept. of ECE, Bapatla Women’s Engineering College , Bapatla
2nd Author- B. Tech 4th Year ,Dept. of ECE, Bapatla Women’s Engineering College, Bapatla
3rd Author- B. Tech 4th Year ,Dept. of ECE, Bapatla Women’s Engineering College, Bapatla
4th Author-Assistant professor ,Dept. of ECE, Bapatla Women’s Engneering Coleege, Bapatla
---------------------------------------------------------------------***---------------------------------------------------------------------
Abstract - The fingerprint pattern is unchangeable and the International Blood Group Reference Laboratory to
stays the same until the person passes away. The fingerprint create a professional numerical terminology based on blood
pattern is unchangeable and stays the same until the person group genetics and is essential in guaranteeing patient
passes away. Fingerprint evidence is still thought to be the safety.safety in blood transfusion. The key challenge for
most important piece of evidence in cases involving events, generating blood bunch forecast technique is non
even in legal courts. The black lines in a fingerprint image comprehensiveness of distinct examples of unique mark
are called ridges, while the lighter areas in between are modalities. Examination throws almost no light on blood
called valleys. The locations where ridges break are known bunch forecast and various infections which accompanies
as miniature points. Each human has a distinct minute maturing, especially if fingerprints are taken as a biometric
pattern, and the likelihood of any two people being methodology. There are three sorts of fingerprints plans
comparable is incredibly low—roughly one in 64 million. found in fingers are Loops, Whorls, Arches are the most
The minutiae pattern is different even for twins. Additionally generally perceived from gathered informational index it
distinct and unchanging from the moment of birth is the nearly found about 65%.
ridge pattern. Fingerprint analysis is also used to look into
the blood group. Ridge frequency estimate is used in the
fingerprint matching procedure. By using the K- Nearest
Neighbor (KNN), the accuracy of the system increases when
compared existing systems. Hence, by using this system
accuracy increases.

Key Words: Finger print pattern, human identity, unique

feature, ridge frequency, blood group, KNN (K-Nearest
Neighbour), system accuracy, biometric identification,
forensic analysis.

1.INTRODUCTION Fig1: Types of fingerprint patterns

The fingerprint pattern is the most dependable and Loop: This is the most common fingerprint pattern, where
distinctive aspect of a person' s identify. The fingerprint ridges enter from one side, form a loop, and exit from the
pattern is unchangeable and stays the same until the person same side they entered. Loop patterns can be further
passes away. Every human has a unique minute pattern, classified as ulnar loops (ridge flow towards the little finger)
with a very small probability of similarity—roughly one in or radial loops (ridge flow towards the thumb).
64 million—between them. The minutiae pattern is different
even for twins. Additionally distinct and unchanging from Whorl: In whorl the Ridges arranged in a circle or spiral
the moment of birth is the ridge pattern. The blood bunch is make up whorl patterns. They can have a number of
an innate characteristic that remains constant throughout an different subtypes, such as accidental whorls, double loops,
individual's lifetime. It's also used in the analysis cycle to central pocket loops, and plain whorls.
eradicate almost all infections. A blood test, which is
obtained by infusing blood or by squeezing a needle on a Arch: In arch Ridges that flow from one side of the
finger, is necessary to determine the type of illness is mixed fingerprint to the other, creating a pattern like waves, are
in with antibodies for the desired outcome; this may cost what define arch patterns. Unlike loops and whorls, they lack
money. The International Society of Blood Transfusion prominent deltas, or triangular ridge formations. Based on
Working Party, which was founded in 1980 in England, the orientation and arrangement of the ridges, as well as the
recognizes 43 blood group systems with 345 antigens for existence of particular characteristics like deltas and ridge
human red blood cells. The Working Party collaborates with counts, fingerprint patterns are examined and categorized.

© 2024, IRJET | Impact Factor value: 8.226 | ISO 9001:2008 Certified Journal | Page 1225
 International Research Journal of Engineering and Technology (IRJET) e-ISSN: 2395-0056
Volume: 11 Issue: 03 | Mar 2024 www.irjet.net p-ISSN: 2395-0072

These patterns are used by automated fingerprint

recognition systems to generate distinct templates for every
person, facilitating precise identification and verification in a
range of settings, such as access control, border control, and
law enforcement.

Blood group identification is essential for a number of

medical operations, including organ transplants, blood
transfusions, and forensic investigations. Blood typing has
traditionally been done via serological techniques, which can
be labor- intensive and need specific lab equipment. But as
more sophisticated technology become available, there's a
rising interest in creating automated, precise blood group
prediction systems.
Fig3: Von neumann architecture

1.2 HARDWARE ARCHITECTURE

The Harvard architecture provides distinct signal buses and
storage for data and instructions. With this architecture, the
CPU houses all of the data storage; the instruction storage is
not accessible as data. Internal data buses are used by
computers to separate memory sections for program
instructions and data, enabling simultaneous access to both.
The CPU could not start up on its own; operators were
required to load programs. It is not necessary to share
Fig2: RH Factor values of finger print patterns attributes between the two memories in Harvard
architecture.
Generally speaking, forward and reverse blood types can be
distinguished by conventional serological testing. In
particular, reverse typing is used to determine whether anti-
A, anti-B, and anti-AB antibodies are present in serum or
plasma, whereas forward typing identifies A and B antigens
based on RBC agglutination by the particular antiserum and
the intensity of agglutination. In clinical blood transfusion,
the ABO antibody titer is also very important and is typically
evaluated using several techniques such as the test tube
method and the microcolumn gel card method. But because
hemagglutination is the basis for all of these serological
techniques, standardization is challenging and accuracy
suffers as a result. It is usually required to perform an ABO
test before receiving a blood transfusion. However, no one
antiserum or method can ensure the identification of every
uncommon, feeble, or mutant antigen. Furthermore, a few
illnesses, like acute myeloid leukemia, can cause blood group Fig4: Hardware architecture
antigen levels to drop.
This project aims to develop a K-Nearest Neighbour (KNN),
1.1 VON NEUMANN ARCHITECTURE model specifically designed for blood group prediction using
fingerprint analysis. In order to accurately forecast blood
The Princeton architecture, sometimes referred to as the types, the model will be trained to recognize and extract
Von Neumann architecture, is a hardware architecture that pertinent information from fingerprint photos. Train the
supports basic components. It permits the sequential use of a system on a large dataset of labeled blood samples to ensure
single memory. Since we use a very small amount of fast that the system can accurately predict the blood group
memory (cache) that is proximate to the processor, memory samples. K-nearest neighbors (KNN) can be used in blood
access times are still far below today's processing rates. An group detection by considering the characteristics of
arithmetic unit, a program control unit, a single bus for different blood types and comparing them to determine the
memory access, and a single, shared memory for data and closest match to an unknown sample based on its features.
programs make up the Von Neumann architecture. KNN works by finding the k nearest data points (blood

© 2024, IRJET | Impact Factor value: 8.226 | ISO 9001:2008 Certified Journal | Page 1226
 International Research Journal of Engineering and Technology (IRJET) e-ISSN: 2395-0056
Volume: 11 Issue: 03 | Mar 2024 www.irjet.net p-ISSN: 2395-0072

samples with known blood groups) to the unknown sample 4. FUTURE SCOPE
and assigning the most common blood group among those
neighbors to the unknown sample. K-nearest neighbors The feature scope of this project is fingerprint scanner. It
(KNN) can be employed in blood group detection by first captures a digital image of the fingerprint. The captured
creating a dataset with features such as ABO blood type, Rh image is called a live scan. The comparison is made between
factor, and other relevant characteristics. Then, when a new the valleys and ridges. Though your whole fingerprint is
blood sample needs to be classified, KNN calculates the recorded, the computer takes only parts of the print to
distance between the features of the unknown sample and compare with others records.
those of the samples in the dataset. The blood group of the
unknown sample is then predicted based on the majority 5. CONCLUSIONS
vote of the k nearest neighbors in the dataset. This method is
particularly useful when dealing with datasets where the Utilizing image processing techniques and the K-Nearest
boundaries between blood groups are not well-defined. Neighbors (KNN) algorithm for blood group detection via
fingerprints offers a promising avenue for non-invasive and
3. RESULTS efficient identification. Through this method, accurate blood
group classification can be achieved, providing valuable
insights for medical diagnostics and forensic investigations.
Moreover, the integration of image processing with machine
learning algorithms like KNN demonstrates the potential for
innovative applications in biometric authentication and
healthcare systems. Further research and refinement of
these techniques could lead to practical implementations
with significant real-world implications

6. REFERENCES
1)“Gupta, A., Agarwal, S., & Jain, R., et.al [3] investigates the
use of machine learning algorithms, including Support
Vector Machines (SVM) and Random Forests (RF), for blood
group identification using fingerprint images” (2023)
Fig5: Process of KNN in output
2) “Verma, A., & Bhatt, V. et.al [1] provides an overview of
In above image there is a trained image and sample image. automated blood group identification methods based on
Both images are compared by the KNN algorithm. By this fingerprint analysis” use of machine learning algorithms (FEB
image we can observe the KNN algorithm process that it can 11,2022)
matching the ridges,loops,whorls,circles which are present
3) “Vijaykumar Patil N, D. R. Ingle; A Novel Approach to
in the fingerprint patterns.
Predict Blood Group using Fingerprint Map Reading use of
After matching these k-nearest neighbours in the any of two HFDU06 Scanner (Apr 02-04, 2021)”
fingerprint patterns it can detect the blood group and it gives
4) Dr. D.Siva Sundhara Raja and J. Abinaya, “A Cost-Effective
the output like below Fig6
Method for Blood Group Detection Using Fingerprints”,
International Journal of Advance Study and Research Work
,Volume 2, March 2019 .

5) A. Narkis Banu, V. Kalpana, “An Automatic System to

Detect Human Blood Group of Many Individuals in a Parallel
Manner using Image Processing”, International Journal of
Pure and Applied Mathematics, vol-118, pp. 3119-3127, 2018

6) G. Ravindran, T. Joby, M. Pravin, and P. Pandiyan,

“Determination and Classification of Blood Types using Image
Processing Techniques,” International Journal of Computer
Applications, vol. 157, no. 1, pp. 12–16, Jan. 2017.

7) Mouad. M.H.Ali, Vivek H. Mahale, Pravin Yannawar, and A.

T. Gaikwad, “Fingerprint Recognition for Person
Fig6: Output Identification and Verification Based on Minutiae Matching”,

© 2024, IRJET | Impact Factor value: 8.226 | ISO 9001:2008 Certified Journal | Page 1227
 International Research Journal of Engineering and Technology (IRJET) e-ISSN: 2395-0056
Volume: 11 Issue: 03 | Mar 2024 www.irjet.net p-ISSN: 2395-0072

2016 IEEE 6th International Conference on Advanced Third Author

Computing, Feb. 2016.

8) Tejaswini H V, M S Mallikarjuna Swamy “Determination

and Classification of Blood Types using Image Processing
Techniques”, ITSI Transactions on Electrical and Electronics
Engineering (ITSITEEE), Volume -2, pp. 2320 – 8945, 2014

9) Ferraz, Ana, Member, IEEE (2013), Automatic system for

determining of blood type using image processing technique”,
IEEE 3rd Portuguese Meeting in Bioengineering (ENBENG),
D. Gopika is a student of B. Tech in stream of
pp. 20-23, Feb. 2013.
Electronics and Communication engineering in Bapatla
10) S. Pimenta, F. O. Soares, and G. Minas, “Development of an Women’s engineering college, Bapatla , Andra Pradesh,
automatic electronic system to human blood typing,” in Proc. india. Here research in Image processing and KNN
IEEE Annu. Int. Conf. Eng. Med. Biol. Soc. (EMBC), Sep. 2012 [email protected]
BIOGRAPHIES Fourth Author
First Author

B. Siva Kumari (guide) , M. Tech (PhD)

G. Mounika is a student of B. Tech in stream of Assistant Professor, Department of ECE
Electronics and Communication engineering in Bapatla Bapatla Women’s Engineering College
Women’s engineering college, Bapatla , Andra Pradesh, Affiliated to Acharya Nagarjuna University
india.Here research in Image processing and KNN [email protected]
[email protected]

Second Author

M. Anusha is a student of B. Tech in stream of

Electronics and Communication engineering in Bapatla
Women’s engineering college, Bapatla , Andra Pradesh,
india. Here research in Image processing and KNN
[email protected]

© 2024, IRJET | Impact Factor value: 8.226 | ISO 9001:2008 Certified Journal | Page 1228