摘要
为弥补传统在线教育模式的不足,设计了基于图像识别在线视频教育平台系统,由视频播放前端和服务器后端组成,视频播放前端基于浏览器进行教学视频的播放,服务器后端主要由识别模块、控制模块和WebSocket通信框架组成。该系统在用户观看网上教学视频过程中,运用霍夫圆变换技术识别并追踪眼球,利用微软表情识别模块API识别表情,经过系统后台分析推断用户是否认真听讲或是否听懂等学习状态,并根据用户不同的学习状态调控视频的播放。该系统同时解决了人脸表情识别与异常状态检测这两大难点。
In order to make up for the deficiency of traditional online education mode, the online video education system based on image recognition is used. The platform consists of a video playback front end and a background server. The front end is based on a browser for teaching video playback, and the background is mainly composed of an identification control module and a WebSocket communication framework. It uses Hough Circle transform to identify and track the eyeball, uses expression recognition API of Microsoft to identify the expression. Whether the user listens attentively or understands can be inferred from analysis results of system background and the playback speed of videos can be adjusted based on different learning states. The system solves the two difficulties of facial expression recognition and abnormal state detection.
作者
陈环宇
高莺
李晓珂
余永华
周宇哲
CHEN Huan-yu;GAO Ying;LI Xiao-ke;YU Yong-hua;ZHOU Yu-zhe(School of Information Management,Jiangxi University of Finance and Economics,Nanchang 330013 China)
出处
《科技创新与生产力》
2018年第12期24-25,31,共3页
Sci-tech Innovation and Productivity
基金
2018年度江西财经大学创新创业项目之创新训练项目(201810421101)
