レーザースキャナによる 幾何補正手法の精度向上
Contents
1.2.
3. X cm Y cm
4. p 1 0 0 cost sing COSK Pa ps cost sinw 0 1 0 0 64 Po ps sint cost 5 coso 0 0 6 2 6 4
5. 21 O00000 4 2 1 Xi Po pi pa Xo Yi ps pa Ds Vi Yo 21 2 Po Ps Zo Yi Zi ui Vi Wi Poll Ps Yo Zo Xi Yi Zi Uj Vi Wi Poll Ps U Yo Zo Poll Ps Yo Zo ay gt uj uj gt a gt v v gt 25 gt w gt gt gt uj gt gt wi n gt X a4 gt uj uj gt as
6. X cm Y cm cm cm 1 2 38 2 44 5 39 6 34 3 15 2 1 08 2 01 1 39 2 46 1 89 3 3 51 0 88 1 63 3 84 3 49 4 1 43 0 97 2 13 2 59 1 58 5 2 78 1 70 3 63 4 69 3 05 6 3 19 1 29 2 50 3 69 2 66 HRD b 7075 cm 2 3 5 6 2 102 2 53 29cm 4 3cm 10cm 2 59 63 000000000000000 Poll Ps x lt 20
7. 61 lcm 10 6 1 4 106 19 8 10 1 6 lcm 6 2 6 1 X cm Y cm Z cm 1 11 89 1 70 6 75 2 6 61 13 03 7 67 3 1 54 5 92 3 42 4 3 47 26 89 14 43 5 6 23 14 76 8 84 6 12 98 19 95 11 31 6 2 6 X cm cm 1 13 00 14 93 13 32 1 2 6 50 7 46 6 66 1 4 3 25 3 73 3 33 1 6 2 16 2 49 2 22 1 8 1 63 1 87 1 67 1 10 1 30 1 49 1 33 1 12 1 16 1 33 1 17 1 14 0 93 1 06 0 94 14 1 17
8. 00 GPS 10cm 10cm 1 3 0000
9. 4 1
10. 2 2005 3 4 LMS Z210 32 2007 DD Oo 33
11. 22 2 4 Poll Ps Yo Zo 22 000 2 2 1 APS AP 2 1 2 1 300 58 2 1 10 u 1 54cm 5 92cm 10cm 2 1 u m v m wm m scanl 56 177 16 194 0 050 58 464 scan2 56 189 16 087 0 048 58 447 scan3 56 214 16 095 0 031 58 472 scan4 56 190 16 139 0 060 58 461 scan5 56 190 16 165 0 076 58 469 56 201 16 073 0 031 58 455 scan7 56 155 16 201 0 100 58 446 scang 56 178 16 219 0 073 58 472 scan9 56 194 16 156 0 073 58 471 scanl0 56 202 16 029 0 029 58 443 0 015 0 059 0 034 0 011 2 2 2 0 00000000000000000000000000000000000000000000
12. 101 8 THEMES al 10 15 20 25 30 7 12 6 14 22 41 159 1 15 14 30 102 26cm 31
13. RS70N 0000 2 20 00 01 00 00 00 OU UU 00 UU 0700 UU UU 000 000 UU 0000 2 200 0000000 19000000000000000000000000 07emr u O 0 86cm 1 70cm 0 22 0000 UU u m v m m scanl 29 848 49 610 6 276 57 843 scan2 29 854 49 588 6 302 57 827 scan3 29 866 49 583 6 304 57 829 scan4 29 829 49 600 6 293 57 824 scan5 29 861 49 572 6 285 57 817 29 880 49 574 6 297 57 828 scan7 29 864 49 584 6 298 57 828 scan8 29 860 49 577 6 299 57 820 scan9 29 875 49 577 6 305 57 828 scanlO 29 864 49 573 6 301 57 834 0017 0 012 0 009 0 007 000 20070 120 250 00000000000000000000000000000000000
14. 71 7 1 7 1 7 1 3 7 1 7 12 THEME 6 4 10 15 gt 24 7 2 00000000000 200000 00000000 7300000000000 300000 00000000 25 7400000000000 400000 00000000 75 00000000000 500000 00000000 26 12 44 10 B 4 2 z 5 3 07600000000000 600000 00000000 12 0000 10 8 000000 e 0000 4 2 0 i i T x 5 10 15 20 25 30 0000 7700000000000 100000 000000000000 27 78 00000000000 200000 000000000000 79 00000000000 300000 0000000000000 28 0 7 10 00000000000 400000 000000000000 7 11 00000000000 500000 0000000000600 29
15. 6 2 2 6 5 6 10 6 2 6 4 X cm Y cm Z cm cm cm 1 8 77 3 54 10 48 13 86 8 39 2 38 67 15 70 35 81 53 29 39 25 3 4 65 1 90 6 14 7 81 4 51 4 27 44 11 15 30 26 41 47 26 75 5 7 18 2 93 6 07 9 41 7 34 6 3 82 1 61 2 66 4 33 3 58 Btr2Zn7354 PER IR cm 2 3 4 5 mur 18 6 1 6 5
16. or OA 09 a b c d 4 1 X Xo 4 8 b un Vo X E 1 10 4 9 un Y Yo 4 8 4 9 a gt gt uivi T gt u X A 10 b gt uivi gt v 2 vi X S Z uw gt u Y Yo 4 11 Z Zv ow Y Yo 4 11 2 Xx Ay X po EME UX E un 419 ow
17. 62 0 000000000000000000 6 1 27cm 100 150 100 100 ujk 6 1 6 1 U U viki 6 2 rg 6 2 V V wik 6 3 rd 6 3 W 6 2 1 6 3 6 3 X cm 1 3 87 0 70 1 37 2 0 58 1 00 0 64 5 1 22 2 07 2 05 4 1 07 2 97 2 14 5 2 18 3 60 4 34 6 1 42 2 61 1 68 18 14
18. 4 12 OO 4 6 4 7 a 4 13 2 gt Xj ViXo M gt ul gt vy _ gt gt 2 ujX X X ujvi X vixi X 2 ve 2vi X ouis Xx 02 m wp 2 ui X pa Xo vi Xi ES Xo 2 uivi Xi XoM gt gt vy gt urd ve gt wyP X uiX X ujiXo X vixi X vixo QviE 2u X IE ov 4 13 5 wr E ud b c d 4 8 4 9 0 5 0504 4 2 4 3 X Y ed V X E 4 14 ed Xy X Y Y bos X Ys Y XO Y 4 20 4 3 4 15 c2 k 0 YE x Q6 Xx 6 x 00 xo 4 15 42 0000000000000000000000 B 4 2 4 30 4 14
19. 12 5 0000 51 5 1 5 4 5 1 X cm Y cm Z cm cm cm 1 16 61 8 61 14 62 23 37 17 67 2 81 46 36 63 58 79 102 26 80 43 3 8 31 5 44 8 57 12 63 8 96 4 53 30 25 16 43 11 71 95 55 47 5 15 43 7 55 11 42 19 30 15 15 6 8 59 5 75 7 33 10 57 8 40 B 0 4 RAM FARE 5 1
20. X Y b Yo 4 10 o 2 2 2 2 4 2 X 0 0 u v OaOp a b o 2 co 2 2 2 4 3 06 0 d a b a b c d u v u v rsin 4 4 8 rsin sin 4 5 r m x 10 1 4 sin cos 20 2 cos x sin g 2 1 4 6 o Af sin sing 20 2 rcos sing 20 I rsin cos Go x 4 7
21. 10 5 31 35 10 57 70m 300 1 54cm 26 89cm 0 58cm 4 34cm 4 33cm 4 69 4 4cm Abstract The laser scanner is used for various purposes The laser scanner can acquire three dimensional data in a short time with wide area The acquired data by the laser scanner has own coordinates The coordinate is called laser coordinate in this study The laser coordinate must be transformed into the ground coordinates using control point The prism 1 usually used as control point Control points of ground coordinate can be precisely measured by total station The laser coordinates of control points cannot be precisely acquired Because the laser scanner doesn t have the telescope for collimation The coordinates of control point are automatically acquired by image processing software of laser scanner However center coordinates of prism cannot be acquired in this method When the prism is used for control point the acquisition accuracy shows about 10cm or more in standard deviation There
22. 7 1 iii 24 24 31 32 33 2 1 1 1 2 LMS Z210 Trimble 5601 2 1 2 2 3 1 5 1 5 2 5 3 5 4 6 1 6 2 6 3 6 4 7 1 7 2 7 3 7 4 7 5 7 6 7 7 7 8 7 9 7 10 7 11 7 12 C C N ER Susah 2 Poe URBES ee eee a e Otm a s Bee 4 5 A 6 amp 00m 2 2 PA 6 09 28 0
23. RIGL LMS Z210 ks AN 1 1 LMS Z210 LMS Z210 T A 6 LMS Z210 LMS Z210 LMS Z210 3D RiSCAN LMS 210Z 1 1 1 2 1 1 LMS
24. 3 4 6 mi G OBS GN 6 4 UU UU 5 95cm 23 7 0000000000000
25. b 0 3100000000000000 000000000000 0 00 000000000000000000000000000000000 0 00000000000000 60000000000000000 7000000000 1900000 0 0 000000000000000000000000000000000000000000000 000000 0000000000000000 500000000 1000000000000 000001000 00000000 50000000000000000000000000000 D0000 0000000000000000000 80000000000000000000000 0 0000000000 150000000000000 0000000000000 1900000000 4 00000000000000000000000000000000 41 00000000000000000
26. 5 2 Z em 000 cm l 10 03 807 838 14 90 12 35 2 2533 1520 10 19 28 47 26 00 3 9 01 11 17 508 14 38 13 50 4 34 26 4720 19 36 59 56 56 62 5 28 93 3256 11 32 43 20 41 38 6 2034 23 19 997 31 35 29 88 2 2 3 300838 5 2 OU 14 5 3 X em Z em 000 em 1 10 31 6 57 7 07 12 96 11 21 2 14 75 39 66 17 47 35 90 33 46 3 8 37 12 21 4 68 13 95 13 15 4 28 29 60 34 21 84 66 57 61 34 5 277 96 45 20 21 64 55 80 50 79 6 18 57 31 83 14 42 37 67 34 49 15 0 3 RRA FMRE co 5 3 00 UU BO 15 5 4
27. 6 6 6 7 2 4 5 6 7 8 9 10 28 00 A 11 8 00 2 01m 4 03m 6 01m 8 00m 16 00m 28 01m 4 01m 6 00m 16 00m vi
28. 10 4 1 X cm Y cm cm 1 21 77 2144 21 62 2 3619 41 48 40 66 3 8 42 978 9 78 4 3498 40 71 40 66 5 42 66 45 89 43 31 6 2700 30 04 27 50 4 2 A X cm Y cm cm 1 1772 15 19 15 81 9 3417 36 95 34 38 3 9 17 10 06 9 95 4 37 06 4099 40 99 5 42 61 43 25 42 97 6 27 84 2945 29 41 11 4 3 X cm Y cm cm 1 17 84 15 46 17 08 2 30 64 38 79 36 68 5 9 88 10 57 10 54 4 40 03 43 11 42 87 5 44 82 45 08 44 90 6 29 65 30 83 30 47 4 4 Xm Y m cm 1 18 43 17 18 17 66 2 31 28 35 75 34 00 5 8 11 9 43 9 48 4 33 66 39 21 39 20 5 39 19 41 89 39 27 6 25 56 28 45 25 89 301 4 5
29. cm cm 1 6 55 5 97 6 68 10 67 8 34 2 5 65 9 97 5 64 12 04 10 62 3 4 97 5 60 4 38 8 09 6 85 4 5 66 8 84 5 95 11 41 9 62 5 6 41 3 65 4 91 8 20 6 49 6 5 26 3 10 5 57 6 71 4 73 1 15 0 4 RAM FARE on 5 4 3 12 63c 102 63cm 2 X 80 43 3 14 38c 4 59 56c B 10 12 96 57 3 A B 5 6 A 2 1 10 00000
30. v X gt T dg gt X vw w gt gt Yo gt gt v gt wi n gt gt gt gt T gt Ziti ag gt v v gt Zivi gt vw w gt gt Ziwi Lo gt u gt gt Wi n gt Z
31. 6 3 1 6 6 6 7 6 3 6 4 6 6 X cm Y cm cm cm 1 6 45 4 93 9 75 12 59 7 09 2 1 84 4 85 3 59 5 95 4 82 3 3 31 1 39 5 96 6 90 3 50 4 2 19 3 16 2 20 4 09 3 44 5 4 43 3 31 4 37 6 30 4 88 6 3 57 2 47 2 82 4 69 3 88 21 IEMIA 3E RRBOTI4 EM 2 4 5 6 iE 50S UR 6 3 Uu UU UU Du 00 6 7 UU UU UU UU UU UU UU 000 000 Uu X cm Z em 000 em DD cm 1 3 52 4 52 9 52 11 08 5 08 2 2 14 3 73 3 69 5 51 4 01 3 3 31 1 39 5 96 6 90 3 28 4 2 24 3 96 1 92 4 61 4 13 5 4 97 3 86 5 65 775 5 65 6 3 07 2 94 2 29 440 3 82 22 EMR EM
32. 00000000000 00000000000 Accuracy Improvement of Geometric Transformation with Laser Scanner 0000000000000000000 00000000000 1115120 00 0000 0000 00 0000 00 00 00 00
33. Z210 350m 80 150m 10 0 9 H 2m 2 5 0 1 2 LMS z2100 000000000 2000000000000 10000000000000 1110 1106 4630 4621 00000000 40 80 total 0 333 00000000 0000000 50009000 Sun onon Bub uut Trimble5601 120 3mm 1 2 Trimble 5601 2 00080600
34. fore geometric transform cannot be executed accurately In this study doesn t acquire the control point with 1mage processing software The control point coordinates are acquired in self made program Moreover not linear affine transformation formula but nonlinear affine transformation formula is used for the geometric transformation formula And accuracy will be improved In this study The required accuracy is 10mm of average distance error margin for land slide monitoring Error margins of image processing software and self made program were compared from the measurement data The prism and the reflector sheet were used for the control point When the validation point in landslide ground is targeted the 5 prism showed 10 57cm average distance error the 5 reflector sheet showed 31 35cm average distance error When prism was used as a control point average distance error has become small more than reflector sheet From an existing study if a lot of reflector sheets are used a highly accurate geometric transformation can be done But reflector sheet can be set up only within the range of 70m from laser scanner The prism can be set up to the range of over 300m from laser scanner If the acquisition accuracy of the prism was improved geometric transformation of the wide area and high accuracy can be expected The control point coordinates were acquired by self made program Standard deviation was minimum 1 54cm and maximum 26 89 in the case of i
35. mage processing software Standard deviation was minimum 0 58cm and maximum 4 34cm in the case of self made program The self made program was able to be acquired in high accuracy The geometric transformation was done by using highly accurate control point Average distance error was 4 33 The geometric transformation was done by using a nonlinear affine transformation type Average distance error was 4 69cm Accuracy has not improved When two or more reference points of the prism and the reflector were used the average distance error margin was 4 4cm But accuracy has improved in other validation point The precision aimed at was not able to be achieved However accuracy has improved Therefore this geometric transformation method will be effective ii DD Abstract 1 1 1 2 1 3 2 1 2 2 2 2 2 2 2 4 1 4 2 5 1 6 1 6 2 6 2 1 6 2 2 6 3 6 3 1
36. tmi koe BE umu tr 7 A 8 A 9 A 10 1 1 1 2 LMS Z210 LMS Z210 2 1 2 2 4 1 4 2 4 3 4 4 5 1 5 2 5 3 5 4 6 1 6 2 6 3 6 4 6 5
Download Pdf Manuals
Related Search