A Vision-Based Approach for Blood Group Detection Using Image Processing

Abstract

Accurate is essential in blood group detection medical situations such as trauma response, organ transplantation, and blood transfusion services. Historically, blood type analysis was based on manual serological methods employing skilled technicians and visual examination, which, while successful, are time-consuming and prone to human error. These limitations pose significant risks, especially in emergency or resource-constrained environments. To tackle this problem, this project puts forward a digitized blood group detection system leveraging image processing techniques. The system utilizes high-resolution images of blood samples mixed with standard reagents (Anti-A, Anti-B, and Anti-D) to detect agglutination patterns. Using Python libraries such as OpenCV, the software processes these images to identify the ABO blood group and Rh factor with minimal human intervention. A user-friendly graphical interface enables operation by non-specialists, making the tool ideal for deployment in ambulances, rural clinics, and field blood camps. Experimental results demonstrate the system’s capability to detect blood groups with high precision and significantly reduced processing time compared to traditional methods. Its offline functionality and hardware-independent design further enhance accessibility and usability. The framework is scalable and adaptable, with potential future enhancements including integration of AI models for complex blood analysis and mobile platform deployment.

Introduction

Blood group detection is essential for safe transfusions, organ transplants, and emergency care, relying on identifying A, B, and Rh(D) antigens on red blood cells to avoid life-threatening reactions. Traditional agglutination tests, while effective, require skilled technicians and are time-consuming, making them less practical in emergencies or resource-limited settings.

This research proposes an automated blood group detection system using digital image processing techniques implemented in Python. The system captures photos of blood samples mixed with Anti-A, Anti-B, and Anti-D reagents using smartphone or digital cameras, then analyzes agglutination patterns with OpenCV and NumPy to classify blood groups (A, B, AB, O) and Rh factors (positive or negative). Key processing steps include color space conversion, noise reduction, contour detection, and pixel intensity analysis.

Designed for offline use on low-resource devices like Raspberry Pi, the system features an intuitive GUI enabling non-experts to perform accurate, rapid blood typing without internet dependency. This approach reduces human error, speeds up diagnosis, and is suited for rural clinics, mobile units, and disaster zones.

The literature review highlights prior methods ranging from basic image thresholding to deep learning, noting challenges like dependence on high-quality images, computational demands, and limited real-time applicability. This work advances the field by offering a lightweight, robust, and fully offline solution balancing accuracy and accessibility.

The methodology involves standardized image capture, preprocessing (resizing, color conversion, noise filtering), automatic extraction of reagent zones, agglutination detection via thresholding and blob analysis, and classification through decision logic. A Tkinter GUI supports image upload, displays results, and provides confidence scores.

Preliminary evaluations on a labeled dataset show promising accuracy and reliability under varied lighting and sample conditions, indicating the system’s potential for practical deployment in diverse healthcare environments.

Conclusion

Precision is defined as the genuine positive fraction of all positive forecasts, which demonstrates the blood group\'s reliability. Recall, also called as sensitivity, assesses the system\'s ability to detect real positives, such as Rh-positives.By leveraging Python-based computer vision algorithms and reagent-based agglutination pattern analysis, the system effectively addresses the limitations associated with conventional manual detection, such as human error, dependency on trained personnel, and prolonged diagnostic times. The framework follows a well-defined workflow, beginning with picture acquisition and progressing through pre-processing, partitioning, and feature extraction to agglutination detection and classification. To determine both ABO and Rh blood types, high-resolution pictures of blood samples treated with Anti-A, Anti-B, and Anti-D reagents were processed using OpenCV and proprietary thresholding logic.The implementation emphasizes both diagnostic accuracy and user simplicity, making the solution accessible in field scenarios, rural healthcare centers, and emergency vehicles. The evaluation phase supported the methodology\'s effectiveness, with the system achieving an overall accuracy of 96.3%, precision of 95.4%, and an F1-score of 95.0%. These results validate the model\'s capability to provide speedy and trustworthy outputs under varied testing settings.The system also demonstrated robustness under noisy input and lighting variations, making it viable for real-world deployment. This solution directly fulfils the problem statement presented in the abstract: enabling fast, error-free, and technician-independent blood group detection. It reduces diagnostic delays and facilitates prompt treatment decisions, especially in time-sensitive environments such as trauma centres, ambulance services, and blood donation camps.

References

[1] Varun Kumar E?B, Yashvanth H, Srinath[2025] J. Blood Group Detection Using Image Processing and Deep Learning. International Journal for Research in Applied Science & Engineering Technology (IJRASET), Vol.?13?Issue III, March 2025. DOI: 10.22214/ijraset.2025.6801 [2] P.?Padmaja et?al. [?2025]. Deep Learning and Image Processing Based Blood Group Detection Combines K Means clustering with classifiers (SVM, ANN,genetic algorithm variants). Provides GUI interface via Tkinter. [3] B?Chandrakala, K?Vyshnavi, K?Sai Kalyani, K?Saikumar, K?Chandana[2024]. Blood group detection using image processing. International Journal of Scientific Research in Science and Technology (IJSRST) [4] Shruti Dub, Aakanksha Sharma, Pankaj, Sagar Kumar Singh, et?al[2024]. Blood group detection using image processing . International Conference on Signal Processing, Communication, Power and Embedded System (SCOPES 2024) [5] Dr. M.?Lakshmi?Prasad et?al.[2024]. Blood group prediction using fingerprint International Research Journal on Advanced Engineering Hub (IRJAEH), Vol?2 No?12, Dec?2024, pp.?2835–2839 [6] Mansi?K,Hitashree?M.Chandana?L.?Hegde[2023]. Determination of Blood Group Using Image Processing International Research Journal of Modernization in Engineering Technology and Science (IRJMETS). [7] Marife?A.?Rosales & Robert?G.?de?Luna[2022]. Computer Based Blood Type Identification Using Image Processing and Machine Learning Algorithm Journal of Advanced Computational Intelligence and Intelligent Informatics (JACIII), Vol?26, No?5, pp?698–705. [8] Choudhury, P., & Chatterjee, S. (2021). Deep Learning for Biomedical Image Processing. Springer Briefs in Applied Sciences and Technology. [9] Patidar, A., & Rathi, R. (2020). Blood Group Detection using Image Processing Techniques. International Journal of Scientific Research in Engineering and Management, 4(6), 1–6. [10] Sharma, S., & Singh, D. (2018). Detection of Blood Group from Blood Sample Image Using Computer Vision. International Journal of Engineering and Technology, 7(2), 145–150.

Copyright

Copyright © 2025 Ms. Ramya A, Mrs. Subhashree D C, Dr. Girish Kumar D, Mrs. Sharvani V, Ms. MM Harshitha. This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.