基于Intel SR300的网页视觉兴趣区域检测

Detection of Visual Area of Interest on Webpages Based on Intel SR300 Camera

以人眼眼球运动和视线数据为出发点,利用Intel SR300摄像头,采集RGB-D彩色深度图像,研究并提出了一种网页视觉兴趣区域检测的设计思路和实现方法。首先,通过校准程序建立人眼视线和屏幕上像素点的对应关系。其次,结合闭眼检测接口的数据,对SDK输出的眼动数据进行预处理以提高其稳定性和准确性。同时,搭建Web Socket服务器,通过localhost回环接口实现摄像头调用进程和本机前端Chrome浏览器进程间的消息传输。最后,编写浏览器插件,利用热点图对人眼视线数据进行可视化。经实验测试,该系统可以较准确地检测出用户在网页上的视觉兴趣区域。

Based on the eye movement and gaze information,a design idea and realization method of detecting visual area of interest on webpages is proposed in this paper. The Intel SR300 camera is used to acquire the RGB image and depth image. Firstly,a relationship between the screen pixel point and gaze information is established through the calibration program. Then,with the intention of improving data stability and accuracy,gaze information from the SDK program is preprocessed combined with blink detection. A Web Socket Server is erected to send data between the SDK process and local front-end Chrome browser through the localhost network loopback interface. Finally,by programming the Chrome extension,gaze information is visualized using heatmap. Experiments show that this system can detect the visual area of interest on webpages accurately.