Blood Group Detection Using Infrared Hand Images and Machine Learning

This paper presents a non-invasive blood group detection system using infrared hand images and deep learning. The proposed approach leverages thermal patterns and vascular characteristics captured through infrared imaging and employs a convolutional neural network based on a pretrained VGG16 architecture. To address infrared thermal variability, the study introduces temperature-aware modeling by categorizing each blood group into low, normal, and high temperature ranges, improving robustness and generalization. Experiments conducted on 4000 infrared hand images achieved a test accuracy of 87.34%. The paper is published in the International Journal of Innovative Research in Technology (IJIRT), Volume 12, Issue 7, Pages 6590–6599, December 2025.Published URL: https://ijirt.org/article?manuscript=189616