自然光源照明眼動儀系統設計
No Thumbnail Available
Date
2015
Authors
Journal Title
Journal ISSN
Volume Title
Publisher
Abstract
眼動儀系統是藉由量測受測者眼睛的凝視點位置以實現追蹤眼球移動的技術，目前已普遍應用於教育、商業、安全以及醫療等方面。目前眼動儀系統以紅外線照明瞳孔追蹤為大宗。由於陽光光線包含了紅外線波段，在白天戶外操作時會因為影像過曝造成眼球追蹤上之困難。本篇論文提出自然光源照明眼動儀系統，僅藉由消費型網路攝影機即可進行眼動追蹤。自然光源照明眼動儀亦因外界環境光源的多變，影像對比度不如紅外線影像高，有許多挑戰需要克服。
本論文提出的系統架構共有三大部分，凝視方向辨識訓練、九點校正處理以及即時眼動追蹤。藉由支持向量機(SVM)辨識人眼凝視方向，以及適用於虹膜追蹤的特徵基底凝視點估測技術找出凝視點座標。此系統已經能夠達到13fps的即時處理應用，凝視點的水平方向誤差角度的總體平均值為0.248度，垂直方向誤差角度的總體平均值為0.306度。
Eye tracking is the process of measuring the point of gaze. An eye tracker is a device for measuring eye positions and eye movement. Eye trackers are used in education, business, security and medical treatment. Most modern eye trackers use the center of the pupil to track the gaze direction by infrared light. It is not suitable to use outdoor because of the image overexposure. We propose an eye tracking system with natural light. The proposed system can measure eye positions and eye movement with webcam. However, it faces several new challenges such as light changes in the environment and lower contrast images. The system contains three parts: training gaze direction recognition, process of nine point calibration, and real-time gaze tracking. Using Support Vector Machine (SVM) to recognize gaze direction and find gaze points using featured-based gaze estimation. The system can immediately process under 13 frames/s. The system can achieve the accuracy of 0.248 degree and 0.306 degree for the horizontal and vertical coordinates of the detected gaze center, respectively.
Eye tracking is the process of measuring the point of gaze. An eye tracker is a device for measuring eye positions and eye movement. Eye trackers are used in education, business, security and medical treatment. Most modern eye trackers use the center of the pupil to track the gaze direction by infrared light. It is not suitable to use outdoor because of the image overexposure. We propose an eye tracking system with natural light. The proposed system can measure eye positions and eye movement with webcam. However, it faces several new challenges such as light changes in the environment and lower contrast images. The system contains three parts: training gaze direction recognition, process of nine point calibration, and real-time gaze tracking. Using Support Vector Machine (SVM) to recognize gaze direction and find gaze points using featured-based gaze estimation. The system can immediately process under 13 frames/s. The system can achieve the accuracy of 0.248 degree and 0.306 degree for the horizontal and vertical coordinates of the detected gaze center, respectively.
Description
Keywords
眼動儀系統, 自然光源, 支持向量機, 特徵基底凝視點估測技術, Gaze tracker, Natural light, SVM, Featured-based gaze estimation