25 3 26 2 14 12350917
3 Cyclesports USBhostAPI Arduino 3 LED LED LED Cyclesports Cyclesports
1 4 1.1...................................... 4 1.2................. 5 1.3................................. 5 2 7 2.1.................. 7 2.2.................. 8 2.3................................ 8 3 10 3.1.................... 10 3.2..................... 11 3.2.1........................ 11 3.3.................................... 13 3.3.1...................... 13 3.3.2...................... 14 3.3.3....................... 14 3.3.4.................................. 15 3.4..................... 16 3.4.1................. 17 3.4.2....................... 18 3.4.3............................ 18 3.4.4............................. 20 3.4.5............................ 21 1
3.4.6....................... 22 4 25 4.1.................................. 25 4.2.................................... 25 4.2.1................................ 27 4.2.2................................ 28 4.2.3................................. 28 4.3............................ 29 4.4...................................... 30 4.5 (PCA)............................... 31 4.6................................. 31 4.6.1 K........ 32 4.6.2 K................ 33 4.6.3 SVM.......... 33 4.6.4 SVM.................. 34 4.7....................... 34 5 LED 35 5.1.................................. 35 5.2 Cyclesports............................... 35 5.3 Cyclesports..................... 36 5.4................. 37 5.5............................... 38 5.5.1............................ 39 5.6 Cyclesports............................ 41 5.6.1............................ 42 5.6.2............................ 44 5.6.3.............................. 44 6 48 2
50 51 3
1 LED 1.1 2011 3 1.1 [1] ( ) 4
1.1: (km) (%) (m/km 2 ) (m/ ) (m/ ) 1985 14,500 8.6 349 1,317 900 1985 23,100 4.7 65 660 280 2006 7,301 0.6 19 84 57 1.2 [2] LED Walmink [3] LED LED 1.3 Cyclesports 5
[4][5] Cyclesports 50hz 3 CSV 3 K Cyclesports 6
2 HASC( ) challenge [6] 2.1 Cossalter [7] 2 2 3 MotoGp2006 Condro [8] 7
[4] (ITS) 2.2 Bao [9] 20 Zhang [10] [5] wavelet 2.3 8
9
3 3.1 3 3.1 4 4 wavelet 10
3.1: 4 15 17km 3.2 3.2.1 wavelet [11] 3.1 3 Apple ipad 11
3.1: REGREX 3 3Hz CSV 3.2 130 3 50m 25m 7km 15km 2 3 2 2 12 12
3.2: 3.3 12 1 1 55 65 2 170 180 3 235 245 3.3.1 3.3 X Y Z 13
3.3: 3.3.2 3.4 3.3.3 3.5 X Y Z 14
3.4: 3.3.4 3 3.6 3.7 15
3.5: 3 3.4 3.8 45 6.5cm MINOURA MINOURA( ) [ih-100-s] 22-29mm Apple ipad Android 16
3.6: SonyEricsson Xperia arc(lt15 i) 3.2 3.2: Xperia arc LT15 i RAM 512MB ROM 1GB MSM8255 1GHz 3.4.1 X Y Z 3 GPS 50hz CSV HTTP POST 17
3.7: 3.9 3.4.2 3 3.10 2 CSV 3 CSV 3.4.3 18
3.8: 3.11 3.3 3.12 19
3.9: 2 1 3.3: 1m 5cm 5mm 3.4.4 Bao [9] 3 3.13 32 20
3.10: 50 (16 ) 1 CSV 3.4.5 21
3.11: 3.4.6 [12] [13] Zhang [10] 1 1 3 8 27 X,Y,Z A x,a y,a z t MI(t) MI(t) = A x (t) 2 + A y (t) 2 + A z (t) 2 (3.1) AI VI (3.2) (3.3) T 32 AI = 1 T T ( MI(t)) (3.2) t=1 V I = 1 T T ( (MI(t) AI) 2 ) (3.3) t=1 22
3.12: SMA SMA = 1 T T ( T T A x (t) + A y (t) + A z (t) ) (3.4) t=1 t=1 t=1 F (F (1),F (2),F (3),F (4), ) 32 FFT energy = 1 15 15 ( F ( i) 2 ) (3.5) i=2 RMS RMS = 1 T T ( A(t) 2 ) (3.6) t=1 23
3.13: 32 1 8 24
4 4 3 4.1 4 4 K 4.2 4 SUNDING SD-548C1 25
4.1: 4 15 17km 2 2 2 2 2 2 4.1 400Kps 4.1 4 4.1: 26
4.2.1 4.2 4.3 15.0 16.9km/h 4.2: 4.3: 27
4.2.2 4.4 4.4: 4.2.3 5m 4.5 4.6 5m [14] 10m 30m 3m 6m 5m 14.0 16.9km 28
2 4.5: 4.3 4 1 28 CSV 4.2 CSV 28 1 CSV 27 2 29
4.6: 4.2: CSV 28 28 1 27 2 25 25 4.4 K K WEKA R e1071 e1071 1vs1 30
cost gamma 27 Yang [15] 3 27 K 4.5 (PCA) R 3 3 3 4.7 R fa.parallel() 5 10 3 X Z X Z 11 4.6 31
4.7: 4.3: 27 K (k=11) 60 12 1 1 8 36 2 1 1 5 41 24 4 1 17 58 4.6.1 K 27 K K 1 20 K 11 71.69 4.3 32
4.4: K (k=11) 64 7 3 0 4 32 9 2 1 1 46 23 0 3 27 50 4.5: 27 SVM 67 7 0 0 6 41 0 0 1 0 44 26 0 0 17 63 4.6.2 K K 1 20 K K 11 12 70.58 4.4 K 4.6.3 SVM 27 gamma 0.001 cost 100 72.42 4.5 K 33
4.6: SVM 65 7 1 1 6 41 0 0 1 0 40 30 0 0 17 63 4.7: K 27 K 11 SVM 27 SVM 11 71.69 =11 70.58 =11 72.42 c=100, g=0.001 73.89 c=1000, g=0.001 4.6.4 SVM 11 gamma 0.01 cost 1000 73.89 4.6 K 4.7 K 4.7 27 3 X Z X Z 11 34
5 LED 5 LED 5.1 LED 3 LED LED 5.2 Cyclesports Android R e1071 LIBSVM 4 35
2 2 1 3 LED RGB LED 2 2 ( ) 2 2013 12 1 17 2 2 5.3 Cyclesports Cyclesports Chang LIBSVM cost gamma microsd 36
microsd 5.1 3 X LIBSVM 4 3 1 74.71 gamma 0.5 cost 32 5.1: Cyclesports 5.4 5.2 USBhostAPI FTDriver 3 LED Arduino Duemilanove Arduino Android 37
Arduino CONECT CONECT Arduino LED FTDriver Arduino Arduino 3 LED 2 5.3 LED LED 5.2: Arduino 5.5 Cyclesports 38
5.3: LED 1 LED Arduino 5.4 Arduino CONECT 5.5 LED LED LED 5.5.1 5.1 6 20 4 4 52 39
5.4: Arduino 5.5: 5.6 5.7 Cyclesports LED 223.5s 91.60 LED 36 69.23 69.23 5m 40
5.1: 52 36 69.23 244s 223.5s 91.60 5.6: 5.6 Cyclesports LED Cyclesports LED LED 41
5.7: Cyclesports 5.6.1 4 LED Cyclesports 2 0.5 5.8 42
Cyclesports 1 1 0.5 2 1 Cyclesports 20 14 7 LED Cyclesports 7 Cyclesports LED 5.8: 43
5.2: LED LED LED 5.6.2 5.2 1 3 4 LED 5 1 5 5.6.3 5.3 Cyclesports LED 5.9 5.4 p=0.006( 0.05) Cyclesports Cyclesports LED LED LED LED LED 3 LED 1 5 Cyclesports 44
1 LED 1 p=0.99( 0.05) LED LED 3 p 0.91( 0.05) 3 LED 4 Cyclesports 4 p=0.57( 0.05) t 1 2 5.5 5.6 LED Cyclesports 0.2 Cyclesports Cyclesports 6.84 LED 6.97 0.01( 0.05) Cyclesports LED 0.128 Cyclesports 0.128 LED 15km 50km 35km 1.2m 45
5.3: 14 Cyclesports LED 4.21 0.95 3.00 1.23 LED 1.78 1.25 1.71 0.83 LED 3.14 0.74 3.21 0.48 LED 4.00 1.38 3.78 1.41 5.9: 5.4: Cyclesports p=.006 4.5 2.5 p=.994 1.0 1.0 p=.916 3.0 3.0 p=.579 4.0 4.0 5.5: Cyclesports 6.84 0.03 - LED 6.97 0.03 +0.12 46
5.6: Cyclesports 6.971 6.842 0.034 0.036 14 14 0.622 0 13 2.937 (T =t) 0.005 1.770 (T =t) 0.011 2.160 47
6 4 32 50 27 48
3 11 K 27 11 11 Cyclesports 4 2 3 LED USBhostAPI 3 LED Cyclesports LED Cyclesports Cyclesports LED LED 49
50
[1].. http://www.mlit.go.jp/road/ir/ircouncil/bicycle environ/1pdf/3.pdf, 2006. [2],,.. IPSJ SIG Technical Report vol2011-ec-20 No.8, May 2011. [3] Wouter walmink, Florian Floyd mueller, and Alan chatham. Lumahelm - an interactive helmet. In CHI 2013 Extended Abstracts, April 2013. [4],,..,, (DICOMO2012), July 2012. [5],,,,,.. IPSJ SIG Technical Report 2006-CVIM-155, September 2006. [6],,. Hasc2010corpus.,, (DICOMO), pp. 76 82, July 2011. [7] Vittore Cossalter, Andrea Aguggiaro, Dirk Debus, Alessandro Bellati, and Andrea Ambrogi. Real cases motorcycle and rider race data investigation: Fall behavior analysis. In Proc. of 20th International Technical Conference on the Enhanced Safety of Vehicles, June 2007. 51
[8] Nowy Condro, Meng-Han Li, and Ray-I Chang. Motosafe: Active safe system for digital forensics of motorcycle rider with android. In International Journal of Information and Electronics Engineering,Vol.2,No.4, July 2012. [9] Ling Bao and Stephen S.Intille. Activity recognition from user-annotated acceleration data. In PERVASIVE 2004 LNCS 3001, pp. 1 17, 2004. [10] Mi Zhang and Alexander A.Sawchuk. Motion primitive-based human activity recognition using a bag-of-features approach. In IHI 12, January 2012. [11] Hidenobu Goto and Motoki Miura. Examination of sensor positions to detect bicycle speeding behavior. In KES-IIMSS, pp. 204 211, June 2013. [12].. 23, February 2012. [13]. 3. 24, February 2013. [14].. http://www.mlit.go.jp/road/sign/kijyun/pdf/ 19741129jitennsyadou.pdf, November. [15] Jun Yang. Toward physical activity diary: Motion recognition using simple acceleration features with mobile phones. IMCE 09, October 2009. 52