自适应视线检测与跟踪算法研究  

Research on Adaptive Line of Sight Detection and Tracking Algorithm

作　　者：朱远知 陆汝华[2] 李佳 李亚兰 ZHU Yuan-zhi;LU Ru-hua;LI Jia;LI Ya-lan(School of Electronic Information and Electrical Engineering,Xiangnan University,Chenzhou 423000,Hunan;School of Software Communication Engineering,Xiangnan University,Chenzhou 423000,Hunan)

机构地区：[1]湘南学院电子信息与电气工程学院,湖南郴州423000 [2]湘南学院软件通信工程学院,湖南郴州423000

出　　处：《电脑与电信》2020年第10期9-12,24,共5页Computer & Telecommunication

基　　金：湖南省大学生创新创业训练计划项目(湘教通[2020]191号-3661);湖南省教育厅科研课题,项目编号:19C1716;湘南学院校级重点科研项目,项目编号:2018XJ31;郴州市科技计划项目郴财教指[2018]102号,项目编号:zdyf201810。

摘　　要：视线检测与跟踪是实时、准确地记录人们观察场景信息时眼球的运动情况,在心理、教育等领域有广大应用前景。大多追踪技术需要佩戴眼动仪等专业设备才能实现,成本高、操作不便。采用基于HOG人脸检测并采用ERT算法回归树的人脸对齐算法标记人脸特征,建立掩膜分离眼睛,并自适应检测瞳孔轮廓,计算瞳孔的位置与视线方向。该追踪系统降低硬件设备的要求,促进视线追踪系统的应用。Line of sight detection and tracking is a real-time and accurate recording of the movement of the eyeball when people observe scene information,and has a wide application prospect in the fields of psychology,education and so on.Most tracking technology needs to wear eye movement instrument and other professional equipment to achieve with high cost and inconvenient operation.This paper uses face alignment algorithm based on HOG face detection and ERT algorithm regression tree to mark face features,establish mask to separate eyes,detect pupil contour adaptively,and calculate pupil position and line of sight direction.The tracking system reduces the requirement of hardware equipment and promotes the application of line of sight tracking system.

关 键 词：人脸检测 特征标记 瞳孔检测 视线跟踪 

分 类 号：TP391.41[自动化与计算机技术—计算机应用技术]