Aquisição e processamento de imagens relativas ao movimento pupilar utilizando plataforma desenvolvida para ESP e MATLAB

Abstract

The pupillary movement acts with the objective of ensuring precision in vision by controlling the amount of light entering the eyes, resulting from the photo-motor reflex, and is composed of pupil dilation and retraction, movements called miosis and mydriasis. By collecting information from this movement it is possible to attenuate several types of complications and neurological deficits, this collection and analysis is called pupillometry. The device developed using the ESP32 microcontroller, has the function of capturing the image and transmitting it to another device provided with greater computing Power in order to process this information. The image processing has the function of filtering possible noise present in the capture and graphically display the pupil diameter in relation to time, so the capture device will be controlled by the processing device using a wireless network where the images will be transmitted, being also possible to change certain image filtering parameters, such as the type of light that will be used in the capture, through an IoT application. Each frame obtained by the image capture Will be submitted to a sequence of filtering with the final objective of identifying the pupil area, using interpolation methods for the frames in which it is not possible to identify the pupil diameter, such as during blinking. The tests performed showed the efficiency of the method used, where it was possible to clearly obtain the temporal variation of the pupil movement, where such information is used for the elaboration of the patient’s diagnosis. The possible future improvements in the capture device foresee practicable tests in human patients, where all the systems created can be implemented in real tests, since the observed results allow the conclusion that all the developed systems have the necessary aptitude.