Research And Implementation Of Pupil Dynamic Parameter Detection System

Pupil recognition and dynamic feature detection technology is an important research topic of modern computer technology.It mainly studies the detection of pupil characteristics,and restores the pupil dynamic process according to the test results.The dynamic characteristics of pupils not only reflect the functional status of pupils,but also indicate the status of certain systems and organs inside the body.It has a wide application prospect in neuroscience,psychology and clinical diagnosis.According to the clinical requirements of pupil detection,a quantitative and accurate method for detecting the pupil dynamic characteristics is proposed based on the existing pupil detection technology,which is integrated use of optoelectronic technology and image processing technology.The main contents are as follows:Designing the pupil image acquisition system with light stimulation,its core is the light stimulator.It mainly includes the controller and the stimulating light source.The main task of the controller is to establish communication with PC through the serial communication module.And then it drives the light source by analyzing the received control instructions.Controller is the core of the bottom of the hardware to ensure the normal operation of the entire light stimulator.Each LED in the stimulating light source is distributed in a circular shape to achieve uniform scattering of light.This paper presents a pupil dynamic characteristic parameter detection algorithm,which can be used in the light stimulation environment.First,the preprocessing about filtering and image normalization was necessary.An improved Otsu algorithm is used for image segmentation to roughly determine the region where the target is located.Then,the edge region is extracted and refined.Finally,The Hough transform is used to locate the pupil and obtain the radius and position of the pupil.In addition,Kalman filter and mean-shift algorithm are used to track pupil target tracking in video stream,which not only improves the accuracy of pupil tracking,but also improves the tracking speed.Finally,according to the acquired characteristic parameters,the change tendency of the pupil is drawn,and the parameters which can be used to characterize the pupil function are obtained by statistical analysis.