Abstract: Ensuring accurate blood classification is imperative prior to administering a transfer of blood from one individual to another during emergency scenarios. Currently,performing these assessments conducted conducting this task laboratory specialists, and when handling a large volume of tests, it becomes tedious and mayresult in errors attributable to humans. This paper proposes the replacement of manual labor in clinical laboratories for the identification of blood groups . The proposed system aims to develop an embedded system utilizing image processing algorithms to conduct blood tests based on blood typing systems. Through a review of various existing methods and their performance evaluation, this paper aims to assist researchers in their endeavors

Keywords: Antigen, Blood Samples, GPU, Histogram, LBP (nearby paired example),Nearest Neighbor Classifier, Image Processing, Pattern Matching.

| DOI: 10.17148/IARJSET.2024.11716