Research And Realization Of Monocular Distance Measurement System Based On Human Pupil Positioning

Human eye positioning as the current popular and important research direction has been developed rapidly and widely used in practice,such as video surveillance,fatigue driving detection and other fields.In this paper,a method of distance measurement based on human eye pupil location will be introduced,target to stop the vision decline caused long time naked eyes contact with mobile devices such as computers and mobile phones.The system can measure the distance between the human eyes and the device screen to protect the role of the eye.The main work of this paper is as follows:1)The calibration method based on camera board panel is proposed,and the calibration process and calibration result is realized by using the OpenCV(Open Source Computer Vision Library)tool on the mobile platform,which avoids the complicated process by using Matlab.2)A face tracking based on Adaptive Boosting(AdaBoost)face detection and Mean Shift is designed,and the corresponding system is implemented on the mobile device.Scale-adaptive target tracking algorithm based on the Mean Shift template matching will be applied in the face tracking,which eliminates the influence of the distance between the human eyes and the screen when we use the mobile device and improves the accuracy of the distance monitoring effectively.3)A priori knowledge of the three-dimensional approach will be used to roughly split the human eye in the face area image.The method of eyes precise positioning based on the gradient integral projection and the method of edge detection combined with the least squares method are used to locate the pupil center.The experimental results show that the method has high pupil positioning speed and high accuracy,and satisfy the requirements of the system.4)The principle of using monocular distance measurement is put forward by using the pinhole imaging model.Lastly,it calculates the distance between the human eyes and the screen to achieve the purpose of distance by combining with the camera matrix obtained above,the human eye pupil pixel information and the actual distance of pupils.The results obtained from the final system show that the method can effectively detect the distance between the human eye and the screen,and has wide practical value.