Nasal Expiration Image Detection on Mirror Surface Using Alternating Optimization

Abstract

ENT (ear, nose, and throat) doctors examine an infant nasal patency by using a nasal breath test. In nasal breath test, ENT doctors observe the fogging of mirror (the nasal expiration image) caused by the nasal expiration of the patient. Due to an immediate disappearance of the nasal expiration image, it is desirable to record the intensity of the nasal expiration image. The nasal expiration image is recorded by a video camera. The captured video is a mixture of video of the intensity of the nasal expiration image and video of background. The purpose of this paper is to separate video of the intensity of the nasal expiration image from captured video. The intensity of the nasal expiration image is estimated using an alternating optimization of nasal expiration image and background. Experimental results show that the proposed method can record the intensity of the nasal expiration image in a simulation environment.