Human Blood Type Classification by ROI Based Pixel Segmentation and Neural Network

Abstract

In the modern times digital image processing technology used by the end users has been in the interest as it provides the easy solution to the complicated issues. Like face recognition, image classification etc. We have proposed the concept of blood group type detection using image processing techniques based on the input images. It will be very difficult to detect the type of the blood to any end user. The need of the exact recognition is high in disaster situation where no lab or master people are accessible to distinguish its sort. Hence we have proposed a pixel cluster based analysis of the blood type based on the pixel analysis and Region of interest (ROI) based image segmentation. Later the use of neural network will help to classify the image based on the pixel analysis features. The proposed system results were obtained by using MATLAB. Successful results were obtained and accuracy of the proposed system is most desirable.