Introduction to Lidar Background of Technology and Accuracy
Introduction to Lidar Background of Technology and Accuracy Assessment 27. 02. 2021 MACBIO
27. 02. 2021 1064 nm 532 nm Lidar on the Electromagnetic Spectrum MACBIO
Creation of lidar data
Quantization of a wave form into a point cloud
RGB Coloured Point Cloud
Point Cloud Classification Value 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 -63 64 -255 Meaning Created, never classified Unclassified Ground Low Vegetation Medium Vegetation High Vegetation Building Low Point (noise) Reserved Water Rail Road Surface Reserved Wire – Guard (Shield) Wire – Conductor (Phase) Transmission Tower Wire-structure Connector (e. g. Insulator) Bridge Deck High Noise Reserved User definable
Ground Classification Algorithms Multiscale Curvature Classification (MCC) is the most simple http: //sourceforge. net/projects/mcclidar/ • Variables • • SD : scale domain, integer in [1, 3] t. SD: curvature tolerance for scale domain SD t 1 = specified by user t 2 = t 1 + 0. 1 t 3 = t 2 + 0. 1 CRSD : cell resolution for scale domain SD CR 1 = 0. 5 * CR 2 = specified by user (nominal post-spacing of input Li. DAR data) CR 3 = 1. 5 * CR 2 f : tension parameter (invariant across scale domains) f = 1. 5 U : vector of points that remain unclassified U = (P 1 , P 2 , P 3 , . . . Pn) n : # of points in U (at the start of each loop pass) Pj : single Li. DAR point Pj = (xj , yj , zj) U 0 = initial point-cloud data specified by user • • • Process 1. U = filter out higher points at same x, y in U 0 and classify them as non-ground 2. for scale domain (SD) = 1 to 3: 3. . . repeat 4. . . S = interpolate new raster surface using Thin Plate Spline 5. . . S' = surface resulting from passing 3 x 3 averaging kernel over S 6. . . for each point Pj in U: 7. . . . if zj > S'(xj , yj) + t. SD then 8. . classify Pj as non-ground and remove it from U 9. . . n. C : # of points classified and removed from U during current iteration through inner loop 10. . . until n. C < 0. 1% * n 11. classify all the points remaining in U as ground •
RGB Coloured Point Cloud
Elevation Coloured Triangular Irregular Network (TIN) Introduction to Lidar
Elevation Coloured TIN: Vegetation Removed
Elevation Coloured TIN: Vegetation/Buildings Removed
Elevation Coloured TIN: Veg/Buildings/Bridges Removed
Assessing Vertical Accuracy Easting Northing Known. Z Laser. Z d. Z 430312. 0 5442736. 0 14. 649 14. 57 -0. 079 430314. 3 5442753. 0 14. 492 14. 39 -0. 102 430317. 4 5442771. 0 14. 309 14. 22 -0. 089 430320. 5 5442789. 0 14. 137 14. 06 -0. 077 430323. 6 5442807. 0 13. 915 13. 84 -0. 075 430326. 4 5442825. 0 13. 651 13. 59 -0. 061 430329. 4 5442842. 0 13. 434 13. 36 -0. 074 430332. 5 5442860. 0 13. 219 13. 11 -0. 109 430335. 7 5442878. 0 13. 021 12. 92 -0. 101 430338. 7 5442896. 0 12. 856 12. 78 -0. 076 430342. 0 5442915. 0 12. 629 12. 55 -0. 079 513001. 9 5413384. 0 3. 324 3. 27 -0. 054 512980. 4 5413384. 0 3. 102 3. 05 -0. 052 512962. 3 5413384. 0 2. 923 2. 86 -0. 063 512943. 8 5413383. 0 2. 724 2. 62 -0. 104 512925. 5 5413382. 0 2. 63 2. 56 -0. 07 512906. 1 5413382. 0 2. 544 2. 48 -0. 064 512887. 7 5413381. 0 2. 587 2. 53 -0. 057 512868. 9 5413381. 0 2. 695 2. 63 -0. 065 512849. 8 5413380. 0 3. 068 3. 01 -0. 058 512830. 7 5413380. 0 3. 467 3. 4 -0. 067 528551. 1 5260013. 0 49. 865 49. 82 -0. 045 528568. 1 5260007. 0 49. 95 49. 92 -0. 03 528585. 2 5260000. 0 50. 035 50. 01 -0. 025 528602. 3 5259994. 0 50. 152 50. 13 -0. 022 528619. 4 5259988. 0 50. 147 50. 13 -0. 017 528637. 5 5259984. 0 49. 733 49. 73 -0. 003 528655. 2 5259984. 0 48. 882 48. 86 -0. 022 528672. 9 5259983. 0 47. 719 47. 71 -0. 009 528690. 8 5259982. 0 46. 527 46. 5 -0. 027 528708. 7 5259981. 0 45. 082 45. 08 -0. 002 521199. 7 5232055. 8 3. 557 3. 59 0. 033 Vertical Error Histogram -0. 30 -0. 26 -0. 22 -0. 18 -0. 14 -0. 10 -0. 06 -0. 02 0. 06 0. 10 0. 14 0. 18 0. 22 0. 26 0. 30 Control Number 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32
Assessing Vertical Accuracy (RMSE)
Assessing Vertical Accuracy (FVA) Vertical Error Histogram RMSE = ± 0. 067 m 0. 25 RMSE = 1σ = 68% of population FVA = 2σ = 95% of population 0. 15 2σ 1σ 0. 1 Normal Distribution Control Points 0. 05 Vertical Error (m) 0. 30 0. 24 0. 18 0. 12 0. 06 0. 00 -0. 06 -0. 12 -0. 18 -0. 24 0 -0. 30 Therefore: 68% of the values are within ± 07 cm 95% of the values are within ± 13 cm Frequency (%) FVA = RMSE * 1. 96 FVA = ± 0. 13 m 0. 2
Thank You Jonah Sullivan Marine and Coastal Biodiversity Management in Pacific Island Countries (MACBIO) Senior GIS Officer Oceania Regional Office (ORO) | The International Union for Conservation of Nature (IUCN) 5 Ma’afu Street, Private Mail Bag, Suva, Fiji Islands. tel: +679 3319084 | fax: +679 310 0128 | mobile: +679 763 8137 | skype: Jonah. Sullivan
- Slides: 17