Image processing based body temperature estimation using thermal video sequence

Abstract

Temperature is most significant vital sign in human body regulation system if this increases at certain level it may be dangerous because temperature support whole immune system, also supporting the healing process. However, high temperature causes serious reimbursements to the human body. Therefore, in this paper, authors propose a non-contact temperature algorithm based on thermography analysis. The purpose of developing this system is to provide body temperature estimation without any discomfort. In the proposed approach viola jones algorithm has been used for face detection in a video stream. Further, as a temperature is not going to change within few seconds, thus rather than measuring the temperature of the face in each of the frame, authors have incorporated a algorithm which selects best frame that contains all the features of the face and then the temperature for the same has been measured. At the end temperature is of the face is exhibited. Simulation results of this established algorithm shows the human face detection & temperature estimation of the FLIR and IR camera videos. Results have been validated against standard temperature measuring device.