【摘要】：在線學(xué)習(xí)是學(xué)生通過網(wǎng)絡(luò)教學(xué)平臺隨時隨地進(jìn)行學(xué)習(xí)的一種全新的學(xué)習(xí)方式,這種在線學(xué)習(xí)方式是基于互聯(lián)網(wǎng)技術(shù)和信息技術(shù)構(gòu)成的開放式學(xué)習(xí)環(huán)境。與傳統(tǒng)課堂式教學(xué)相比,網(wǎng)絡(luò)教學(xué)時空分離的特性能夠?yàn)槿藗兲峁┓奖憧旖莸膶W(xué)習(xí)途徑,但其缺點(diǎn)是教師無法通過觀察學(xué)生面部表情來分析學(xué)生的學(xué)習(xí)情緒與狀態(tài),進(jìn)而不能及時調(diào)整教學(xué)策略。針對目前的網(wǎng)絡(luò)教學(xué)系統(tǒng)的缺陷,本文研究和實(shí)現(xiàn)了具有情感交互功能的網(wǎng)絡(luò)教學(xué)系統(tǒng)。本文對網(wǎng)絡(luò)教學(xué)特點(diǎn)和教學(xué)心理學(xué)的進(jìn)行了深入研究分析,設(shè)計了網(wǎng)絡(luò)在線學(xué)習(xí)情感模型,該模型從認(rèn)知度、興奮度、趨避度三個維度來描述了在線學(xué)習(xí)者的情感。認(rèn)知度和趨避度主要從學(xué)習(xí)者面部表情獲取學(xué)習(xí)者的狀態(tài)信息,興奮度主要對人眼疲勞度進(jìn)行檢測從而獲取學(xué)習(xí)者精神狀態(tài)�；谠撃Ｐ�,利用圖像處理技術(shù)實(shí)現(xiàn)了網(wǎng)絡(luò)在線學(xué)習(xí)情緒檢測系統(tǒng),論文主要完成的工作包括:設(shè)計了網(wǎng)絡(luò)在線學(xué)習(xí)情緒檢測系統(tǒng)的人臉特征提取子模塊中的檢測算法,重點(diǎn)研究了如何減少人臉特征點(diǎn)的搜索時間,以滿足在線學(xué)習(xí)系統(tǒng)的高實(shí)時性要求。構(gòu)建了學(xué)習(xí)者人臉CLM模型,使用SVM分類器對認(rèn)知度和趨避度的學(xué)習(xí)表情進(jìn)行識別;通過提取學(xué)習(xí)者眼部特征,使用P80標(biāo)準(zhǔn)的PERCLOS方法對學(xué)習(xí)者興奮度進(jìn)行檢測。基于HTML5和Java程序設(shè)計技術(shù)實(shí)現(xiàn)網(wǎng)絡(luò)在線學(xué)習(xí)情感檢測系統(tǒng),并將其用于網(wǎng)絡(luò)教學(xué)平臺。在實(shí)現(xiàn)過程中充分考慮了應(yīng)用場景、網(wǎng)絡(luò)流量、服務(wù)器負(fù)載等因素,將整個圖像處理與識別過程置于WEB前端完成,解決了采用傳統(tǒng)方法將復(fù)雜的圖像處理過程置于WEB服務(wù)器處理造成對服務(wù)器負(fù)載過大,不適合在線學(xué)習(xí)人數(shù)眾多的網(wǎng)絡(luò)學(xué)習(xí)的難題。本文通過對某課程下的25名學(xué)生做了對比檢測實(shí)驗(yàn),利用三維學(xué)習(xí)情緒模型分析得到的學(xué)生的學(xué)習(xí)狀態(tài)。實(shí)驗(yàn)結(jié)果表明網(wǎng)絡(luò)在線學(xué)習(xí)情感檢測系統(tǒng)能夠?yàn)榻處熖峁┝私饩W(wǎng)絡(luò)在線學(xué)習(xí)中學(xué)生學(xué)習(xí)情緒的可靠途徑。
[Abstract]:Online learning is a brand-new learning mode for students to study at any time and anywhere through the web-based teaching platform. This online learning mode is an open learning environment based on Internet and information technology. Compared with traditional classroom teaching, the separation of time and space in network teaching can provide a convenient and quick way for people to learn, but its disadvantage is that teachers can't analyze students' learning emotion and state by observing students' facial expressions. Therefore, the teaching strategy can not be adjusted in time. In view of the defects of the current network teaching system, this paper studies and implements the network teaching system with emotional interaction function. In this paper, the characteristics of network teaching and teaching psychology are deeply studied and analyzed, and an online learning emotion model is designed. The model describes the emotion of online learners from three dimensions: cognition, excitability and avoidance. Cognition and avoidance are mainly used to obtain learners' state information from facial expressions, and excitability is mainly used to detect human eye fatigue so as to obtain learners' mental state. Based on this model, an online learning emotion detection system based on image processing technology is implemented. The main work of this paper is as follows: the detection algorithm in the face feature extraction sub-module of the online learning emotion detection system is designed. This paper focuses on how to reduce the search time of facial feature points in order to meet the high real-time requirements of online learning system. The learner face CLM model is constructed and the SVM classifier is used to recognize the learning expression of cognitive degree and avoidance degree. The learner's eye feature is extracted and the P80 standard PERCLOS method is used to detect the learner's excitability. Based on HTML5 and Java programming technology, the online learning emotion detection system is implemented, and it is used in the network teaching platform. In the process of implementation, the application scenario, network traffic, server load and other factors are fully considered, and the whole image processing and recognition process is completed in the front end of WEB. The traditional method is used to put the complex image processing process into the WEB server processing, which results in too much load on the server and is not suitable for the online learning of a large number of people on-line. In this paper, a comparative test of 25 students in a course is carried out, and the learning state of the students is analyzed by using the three-dimensional learning emotion model. The experimental results show that the online learning emotion detection system can provide a reliable way for teachers to understand the online learning emotion of middle school students.
【學(xué)位授予單位】：北京交通大學(xué)
【學(xué)位級別】：碩士
【學(xué)位授予年份】：2017
【分類號】：TP391.41

【參考文獻(xiàn)】

相關(guān)期刊論文 前10條

1 高瑩瑩;朱維彬;;深層神經(jīng)網(wǎng)絡(luò)中間層可見化建模[J];自動化學(xué)報;2015年09期

2 田小平;喬東;吳成茂;;基于雙直方圖均衡化的彩色圖像增強(qiáng)[J];西安郵電大學(xué)學(xué)報;2015年02期

3 韓錫斌;葛文雙;周潛;程建鋼;;MOOC平臺與典型網(wǎng)絡(luò)教學(xué)平臺的比較研究[J];中國電化教育;2014年01期

4 柏宏權(quán);韓慶年;;機(jī)器學(xué)習(xí)在適應(yīng)性教學(xué)系統(tǒng)中的應(yīng)用研究[J];南京師范大學(xué)學(xué)報(工程技術(shù)版);2007年04期

5 謝曉林;余勝泉;程罡;黃燁敏;;網(wǎng)絡(luò)教學(xué)平臺的新發(fā)展[J];開放教育研究;2007年05期

6 解迎剛;王志良;永井正武;喬向杰;祝長生;平安;;基于Agent技術(shù)的人性化E-learning系統(tǒng)研究[J];計算機(jī)工程;2007年06期

7 馬希榮;劉琳;桑婧;;基于情感計算的e-Learning系統(tǒng)建模[J];計算機(jī)科學(xué);2005年08期

8 吳啟迪,凌培亮,陳其暉;基于多智能代理的協(xié)同網(wǎng)絡(luò)化教育系統(tǒng)[J];同濟(jì)大學(xué)學(xué)報(自然科學(xué)版);2004年11期

9 宋亦旭,賈培發(fā);基于人工情感的擬人機(jī)器人控制體系結(jié)構(gòu)[J];機(jī)器人;2004年06期

10 陳麗;遠(yuǎn)程學(xué)習(xí)的教學(xué)交互模型和教學(xué)交互層次塔[J];中國遠(yuǎn)程教育;2004年05期

相關(guān)碩士學(xué)位論文 前2條

1 李鵬;人臉識別中的若干算法研究[D];中國科學(xué)技術(shù)大學(xué);2009年

2 祁曼飛;基于情緒認(rèn)知的個性化網(wǎng)絡(luò)教學(xué)系統(tǒng)研究[D];首都師范大學(xué);2009年

，

本文編號：2469416