Kalman filter predictions applied to glottal closure instant detection

Given the importance of the voice in our daily lives, any study focused on its pathologies and the way of detecting and treating them nimbly is of common interest. This research work focuses on the search for an image processing solution to help otolaryngologist to detect glottal closure instant by using Kalman filters based on stroboscopic videos. This solution will allow to prevent the irregularities in the images and the subsequent measuring errors, caused by different pathologies, both organic and functional. The results show that the prediction of the Kalman filter aids in the motion detection, but, above all, at the point at which very rough transitions are produced in the measurements, that is to say, when the glottal closure instant is produced. The results are encrypted and embedded into a digital image using an invisible watermarking technique (LSB) in order to ensure the privacy, confidentiality and integrity of the results when they are transmitted.