Lidar 101: Intro to Lidar Jason Stoker USGS EROS / SAIC
Lidar Light Detection and Ranging Laser altimetry ALTM (Airborne laser terrain mapping) Airborne laser scanning
Lidar Laser IMU (INS) GPS Scanning Mirror On board computer
Laser Pulse laser Records distance to target Time * c / 2 Laser wavelengths can differ 1064 nm 100-150 150 khz systems available today
Lidar Laser L A S E R L A S E R
IMU Inertial Measurement Unit Gyroscopes and accelerometer Roll Pitch Records roll, pitch, yaw of aircraft Yaw Courtesy of NASA
GPS Global positioning system Differentially corrected Provides cm accuracy of aircraft Allows cm accuracy of laser pulse
Scanning Lidar Courtesy of Dodson & Associates
Records data Photon detector (intensity) Laser timing IMU info GPS info Mirror scan angle Converts into XYZ Millions of points On-board display On-board Computer
X r X r G o r X G LIDAR Equation r r = X o + R φ, κ RΔ ω, Δφ, Δκ PG + Rω, φ, κ RΔ ω, Δφ, Δ 0 ρ ω, κ Rα, β 0 ground coordinates of object point under consideration ground coordinates of GPS antenna phase center P r G offset between laser unit and phase center w.r.t. the laser unit coordinate system R ω, φ, κ rotation matrix that needs to be applied to the ground coordinate system until it is parallel to the IMU coordinate system R rotation matrix that needs to be applied to the IMU coordinate system until it is parallel to the Δ ω, Δφ, Δκ laser unit coordinate system R α,β rotation matrix that needs to be applied to the laser unit coordinate system until it is parallel to the laser beam coordinate system Ayman Habib
Error Sources Bore-sighting offset error (P G ) Bore-sighting angular error (Δω, Δφ, and Δκ) Laser range error (ρ) Laser beam angular error (α, β) Laser range noise (ρ) GPS noise (X o ) IMU noise (ω, φ, and κ) Laser beam angular noise (α, β) Ayman Habib
Error Budget Detector Bias and gain Diff b/w electronical and mechanical sensor origin Fluctuations in pulse caused by atmosphere Variance caused by SNR, variance of pulse length and variance of sampling frequency Pointing Jitter Horizontal position dependent on stability of mirror (tan of terrain slope- high relief, higher error) INS (IMU) errors Misalignment and time-dependent gyro-drift
Error Budget (cont) GPS Errors Orbit Errors Ionospheric and tropospheric delays, Phase ambiguities Multipath returns Atmospheric influences Propagation delays Defraction, absorption and scattering Dependent on flying height, moisture
Slope and flight line effects: FLIGHT LINE dz dz dz
Error Budget (cont) Reflectivity of target Influences SNR and consequently point precision GPS and INS integration When use the same clock, errors are minimal When don t, asychronation in the clocks cause positioning errors in the order of the velocity of the aircraft time the sync error
Courtesy of Amar Nayegandhi
Platform type Lidar Differences Profile or scanning Single, multiple, or waveform returns Footprint Size Posting density Atmospheric / terrestrial / bathymetric
Space- based Platforms Atmospheric Airborne Ground
Courtesy of Robert Kayen
Pulses vs Returns
Courtesy of Dave Harding
Returns Single Return Multiple returns Waveform Returns
Returns Single Return Multiple returns Waveform Returns 1 st return 2 nd return 3 rd return 4 th return
Courtesy of Amar Nayegandhi
Intensity Intensity = amount of energy reflected for each return Different surfaces reflect differently Hard to quantify DNs
Returns Single Return Multiple returns H E I G H T Waveform Returns Energy Returned
Returns Single Return Multiple returns H E I G H T Waveform Returns Energy Returned
Footprints Different types of systems: Large footprint
Footprints Different types of systems: Large footprint Small footprint
Beam Divergence Light tends to spread out Laser is coherent light, but spreads too Beam divergence Measured in mrads Higher up, the larger the footprint will be
Beam Divergence Light tends to spread out Laser is coherent light, but spreads too Beam divergence Measured in mrads Higher up, the larger the footprint will be
Posting Density Returns called postings Function of: Laser pulse rate Hz or khz Flying ht/speed Scan angle Not regular interval 0.5m 0.5m 0.5m 0.5m 0.5m 0.5m 0.5m 0.5m 0.5m 0.5m 0.5m 0.5m 0.5m 0.5m 0.5m 0.5m 1.5m 0.5m 0.5m 0.5m 0.5m 0.5m 0.5m 0.5m 0.5m 0.5m
Courtesy of Amar Nayegandhi
Raw lidar returns
Introduction Most commercial systems today are: Small footprint Multiple return Collecting imagery simultaneously Large footprint continuous waveform operated by NASA Some systems sample in the waveform
Current Lidar Representations Point Cloud of 3D data Triangulated Irregular Network (TIN) Raster
Current Lidar Representations Point Cloud of 3D data Triangulated Irregular Network (TIN) Raster
Converting Points to TIN
TIN Without Breaklines Not Hydro-Enforced Courtesy of Dave Maune
TIN With Breaklines Hydro Hydro- Enforced Courtesy of Dave Maune
Current Lidar Representations Point Cloud of 3D data Triangulated Irregular Network (TIN) Raster (Grid)
Interpolating a Raster- IDW
Research and Applications
A wealth of information derived from lidar Volcano monitoring Vegetation / Biomass Land Cover Earthquake faults Hydrologic / Hydraulic Urban / Suburban Response Coastal Studies Carbon studies
Lidar Applications Visualization Vendor Data Dissemination Bare Earth Analyses Vegetative Analyses Structural Analyses
Bare Earth Analysis Primary focus of commercial sector $$$$$ Technology is become widely accepted More cost effective / accurate than photogrammetry
1-arc-second resolution 1/3-arc-second resolution 1/9-arc-second resolution
Multi-resolution NED: 1-arc-second 1/3-arc-second 1/9-arc-second
Multi-Resolution NED 1 arc second 1/3 arc second 1/9 arc second
Manual Post-Processing
Northern San Andreas LIDAR: fault geomorphology Courtesy of Carol Prentice
Vegetation
Bare Earth Filtering
Vegetation Identified
Canopy Height
Canopy Height
Canopy Height Model Leon County, FL
Canopy Height
Crown Base Height Distance from the ground to the lowest needle-bearing branch Important in fire modeling Surface fire / crown fire Crown Base Height
Length Length of of live live crown crown Crown base height
Multi-temporal LIDAR As LIDAR collections increase, there will be more availability to use LIDAR for monitoring change detection purposes
Monitoring Growth with LIDAR Individual Tree LIDAR Datasets 2003 1999 1998 3 m height growth Courtesy of Hans-Erik Andersen
Mount Saint Helens Courtesy of Dave Harding
Bare earth difference
Bare earth difference
Lidar / Spectral Fusions
Intensity information Trees Roof types Grass Water
Lidar point cloud of Structures
Feature Extraction using Lidar
National Lidar Initiative Meeting February 14-16 16 th 2007, USGS headquarters
http://lidar.cr.usgs.gov http://lidarbb.cr.usgs.gov