Facial Recognition And Verification System For Accessing Patient Health Records

Abstract

This paper proposes an Image processing based embedded system to access patient health records by face recognition. Nowadays carrying all the medical report during every visit became a tedious process and hence the medical reports are stored in a database and accessed with a unique identification number. In this paper, face recognition is used instead of a unique identification number. In emergency situations, this helps to know the medical history of the patients. The proposed system consists of a Raspberry Pi 3 processor and a webcam which is used to capture the face. Local Binary Pattern(LBP) and HAAR Cascade algorithms are used to detect and recognize the face. The features of the face are extracted using LBP and HAAR cascade algorithm. The extracted features are given to classifier which compares these features with the trained ones and displays the data and reports stored in the database.Both algorithms are implemented in python using OpenCV. Experimental results indicate that LBP algorithm is more efficient than HAAR cascade classifier algorithm in terms of execution time, classification accuracy, confusion matrix, and F1 score.