21 e-learning Development of Real-time Learner Detection System for e-learning 1100349 2010 3 1
e-learning WBT (Web Based training) e-learning LMS (Learning Management System) LMS WBT e-learning e-learning WBT LMS i
Abstract Development of Real-time Learner Detection System for e-learning Naoki MATSUMOTO In recent years, WBT (Web Based Training) and other e-learning are introducing at higher education institutions such as distance learning. Distance learning mainly apply the internet. Distance lecture is one of distance learning style. Higher education institutions is conducting distance lecture using the internet. accredit credit based on university establishment standards. Distance lecture can Each higher education institutions are addressing the distance lecture. It is necessary that distance lecture secure learning time based on university establishment standards to accredit credit. Currently, Learner s progress figure out from recorded information on LMS (Learning Management System) such as learning time and result of test. However, administrator of distance lecture can not figure out learner s situation from recorded information on LMS at directly and real-time. Therefore, administrator of distance lecture can not figure out the Learner s situation such as learn and not learn. In this paper, we developed real-time learner detection system for WBT and other e-learning. Moreover, we evaluated this system to confirm correctly detect the presence of learner. key words e-learning, WBT, LMS, Distance Lecture, Certification Unit, University Establishment Standard ii
1 1 2 2 2.1 e-learning....................... 2 2.1.1 WBT (Web Based Training).................... 3 2.1.2 LMS (Learning Management System)............... 3 2.1.3 ( )..................... 4 2.1.4....................... 4 2.2................... 5 2.3................................. 6 3 8 3.1..................... 8 3.2............................ 9 3.2.1 2................................. 9 3.2.2............................... 10 3.2.3................................ 10 3.2.4 ( )........................... 11 3.2.5 ( )...................... 11 3.2.6............................... 12 3.3............................ 13 3.3.1 2............... 13 3.3.2.................... 13 3.3.3..................... 14 iii
3.3.4................ 14 3.4............................. 15 3.5...................... 16 4 18 4.1................................ 18 4.2................................. 18 4.3.................................... 20 4.4............................. 21 4.4.1.......................... 21 4.4.2.................... 22 4.4.3.......................... 23 5 24 5.1.................................. 24 5.2................................. 24 5.3................................ 29 5.3.1 A B............... 29 5.3.2 C.................. 30............. 30.................... 30................ 31 6 33 6.1................................ 33 6.2........................... 33 6.3.......................... 35 iv
6.4........................... 36 6.5................................ 38 6.5.1.................... 38..................... 38. 39 6.5.2............................. 39 7 41 42 43 v
3.1................... 16 4.1......................... 19 4.2 ( )....................... 19 4.3.................................... 20 4.4 ( ).............. 22 4.5 ( )............. 23 4.6............................. 23 5.1 A....................... 26 5.2 A.......................... 26 5.3 B....................... 27 5.4 B.......................... 27 5.5 C....................... 28 5.6 C.......................... 28 6.1............................ 38 vi
3.1.............................. 15 5.1................................. 25 5.2 Logicool 1.3-MP Webcam C300................... 25 6.1................................ 34 6.2 Logicool 1.3-MP Webcam C200................... 35 6.3 Microsoft LifeCam VX-3000..................... 36 6.4......................... 37 6.5.......................... 37 6.6.................. 37 vii
1 WBT e-learning Web 1999 [1] WBT e-learning LMS (Learning Management System) LMS [2] WBT e-learning Web Web 1
2 e-learning e-learning 2.1 e-learning (ICT : Information Communication Technology) e-learning e-learning 18 IT [3] 2 WBT (Web Based Training) e-learning (Distance Learning) SCS 2
2.1 e-learning (Space Collaboration System) WBT 2001 3 [1] 2.1.1 WBT (Web Based Training) WBT Web e-learning e-learning CD-ROM DVD CBT (Comuputer Based Training) CBT WBT CBT Web Web 2.1.2 LMS (Learning Management System) LMS (Learning Management System) CMS (Course Management System) WBT e-learning LMS LMS 3
2.1 e-learning 2.1.3 ( ) 2 VOD (Video On Demand) VOD VOD Web 2.1.4 [1] [1] 1999 3 60 4
2.2 2000 11 2001 3 [4] [1] 124 60 [5] 62 30 [6] 167 30 e-learning [7] e-learning 2.2 WBT WBT e-learning LMS (Learning Management System) LMS [2] LMS 5
2.3 e-learning 2.3 [1] LMS WBT e-learning Web Web Web USB (Universal Serial Bus) USB Web Web LMS Web Web 6
2.3 7
3 Web PC Web USB PC USB Web WBT e-learning 3.1 WBT Web 8
3.2 WBT e-learning Web Web WBT e-learning PC 3.2 2 ( ) ( ) 3.2.1 2 2 2 2 2 2 9
3.2 2 3.2.2 3.2.3 10
3.2 3.2.4 ( ) ( ) 2 3.2.5 ( ) 11
3.2 3.2.6 12
3.3 3.3 WBT e-learning 3.3.1 2 2 2 3.2.1 3.2.2 2 2 3.3.2 3.2.6 WBT 13
3.3 e-learning ( ) 3.3.3 3.3.4 3.2.3 14
3.4 3.1 2 3.2.4 3.4 3.1 2 15
3.5 3.5 3.1 3.1 16
3.5 4 17
4 3.1 4.1 WBT e-learning Web ActionScript3.0 Web Apache Web Web 1 FPS (Frames Per Second) 4.2 18
4.2 4.1 4.1 2 4.2 4.2 ( ) 4.4.1 19
4.3 4.3, Haar-like Haar-like Haar-like Haar-like OpenCV ObjectDetection ( ) ActionScript3.0 Marilena OpenCV haarcascade frontalface alt.xml 4.3 4.3 20
4.4 4.4 4.4.1 RGB RGB R ( : Red) G ( : Green) B ( : Blue) RGB RGB RGB R G B HSV HSV H (Hue : ) S (Saturation : ) V (Value : ) 3 RGB HSV RGB H ( ), HSV RGB HSV HSV H ( ) 0 30 [8] HSV 21
4.4 3.3.4 0.1 4.4.2 haarcascade frontalface alt.xml ( ) 4.4 4.5 4.4 4.5 4.4 ( ) 22
4.4 4.5 ( ) 4.4.3 4.6 3 3 4.6 23
5 5.1 3 (A B C) 5.1 4.2 Web Logicool 1.3-MP Webcam C300 Logicool 1.3-MP Webcam C300 5.2 5.2 3 (A B C) A B A 5.1 5.2 B 5.3 5.4 24
5.2 5.1 ( ) ( ) ( ) ( ) 5.2 Logicool 1.3-MP Webcam C300 2cm - ( ) ( ) OS CPU 60 130 30 / WindowsXP(SP2 ) WindowsVista Windows7 1GHz CPU(1.6GHz ) 512MB USB 2 C 3 C 5.5 5.6 5.5 5.6 96 426 555 25
5.2 5.1 A 5.2 A 26
5.2 5.3 B 5.4 B 27
5.2 5.5 C 5.6 C 28
5.3 5.3 A B C A B 5.3.1 A B A B 2 A 456 A A A 29
5.3 5.3.2 C C 3 96 3 5 426 30
5.3 90 cm 3 3 3 555 31
5.3 20 cm 20 cm 32
6 6.1 6.1 160 120 pixel 6.2 Logicool 1.3-MP Webcam C300 Logicool 1.3-MP 33
6.2 6.1 Logicool 1.3-MP Webcam C300 ( ) ( ) ( ) ( ) Webcam C200 Microsoft LifeCam VX-3000 3 5.2 6.2 6.3 6.4 Logicool C300 Microsoft LifeCam VX-300 1000px Logicool 1.3-MP Webcam C300 Microsoft LifeCam VX-300 Logicool 1.3-MP WebCam C200 2088px 90cm 34
6.3 6.2 Logicool 1.3-MP Webcam C200 ( ) OS CPU ( )mm 2cm - ( ) 60 30 30 / WindowsXP(SP2 ) WindowsVista Windows7 1GHz CPU(1.6GHz ) 512MB USB 48 114 73 ( ) 6.3 ( ) ( ) ( ) 3 6.5 Logicool 1.3-MP Webcam C300 6.2 ( ) 1808px ( ) 2473px 35
6.4 6.3 Microsoft LifeCam VX-3000 ( ) OS CPU 2cm - ( ) 55 30 30 / WindowsXP SP2 WindowsVista Windows7 Windows Vista : Pentium4 2.8GHz (Pentium Dual Core 1.8 GHz ) Windows XP SP2 : Pentium4 1.8GHz (Pentium4 3GHz ) WindowsVista : 1G RAM (2GB ) Windows XP SP2 : 256MB RAM (500MB ) USB ( )mm 57 57 68 ( ) 535px 6.4 36
6.4 6.4 ( ) (Logicool C300) 1454px 1808px 354px 390px Logicool C200 2041px 2844px 803px 2088px Microsoft LifeCam VX-300 2390px 2707px 317px 693px 6.5 ( ) ( ) 1454px 1808px 354px 390px ( ) 2129px 2473px 344px 289px ( ) 328px 535px 207px 296px 6.6 90cm 6.6 ( ) 1332px 1654px 322px 380px 0px 939px 939px 339px 37
6.5 比較検証の考察 6.5 比較検証の考察 比較検証の結果に対して考察を行った 本節では なぜカメラの違いや 照明環境の違い 装飾品の違いから検出結果および顔検出 肌色抽出結果に違いがでたのか理由を明らかにす る そして これらの検出結果の違い 抽出結果の違いから今後どのような対策が必要とな るか考察を行う 6.5.1 カメラの違いと照明環境の違い 肌色情報の抽出結果について 本検証から使用するカメラによって色の発色が違うということがわかった 実際に使用し たカメラで撮影した画像を図 6.1 に示す Logicool 1.3-MP WebCam C200 で撮影した画像 図 6.1 Logicool 1.3-MP WebCam C300 で撮影した画像 Microsoft LifeCam VX-3000 で撮影した画像 各カメラで撮影した画像 これらの画像からわかるように 使用するカメラによって明るさが異なり 肌色にも違い が見られる 本研究では HSV 表色系の H (色相) の値が 0 から 30 のときに肌色として抽 出するようにしているが このように 使用するカメラによって発色が違うため H の値が 変わってしまい 肌色情報の抽出結果に差が現れたと考えられる また 照明環境の違いにより検証を行った 照明環境の違いにより抽出できる肌色情報に 差がでることを確認した とくに 暗い (蛍光灯ひとつ点灯) 環境であれば 安定して肌色 情報を抽出することが難しいことが示唆された これは 学習を行う環境が暗くなってし まったため 設定しておいたしきい値では肌色情報を抽出できなかったことが原因だと考え られる 38
6.5 H ( ) 6.4 Logicool 1.3-MP WebCam C200 2088px H ( ) Web 6.5.2 39
6.5 40
7 e-learning LMS LMS 41
2 1 4 3 42
[1] http://law.e-gov.go.jp/cgi-bin/idxsearch.cgi [2] e 25 pp501-502 2009 [3] IT http://www.kantei.go.jp/jp/singi/it2/kettei/060119honbun.pdf [4] http://www.mext.go.jp/b menu/shingi/12/daigaku/toushin/971202.htm [5] http://law.e-gov.go.jp/cgi-bin/idxsearch.cgi [6] http://law.e-gov.go.jp/cgi-bin/idxsearch.cgi [7] http://law.e-gov.go.jp/cgi-bin/idxsearch.cgi [8] Sherrah, J. & Gong, S (2001). Skin Color Analysis. CvOnline. http://homepages.inf.ed.ac.uk/cgi/rbf/cvonline/entries.pl?tag288, 2009 2 5 43