Presentation Overview

1
Presentation Overview

• PAMAP LiDAR Program
• PAMAP LiDAR overview
• LiDAR Basics
• Acquisition and sensors
• LiDAR processing
• Contour generation

2
PAMAP LiDAR Program

• Supports companion PAMAP orthophoto program
• Requirements for detailed terrain data
• Flooding, geologic mapping, slope analysis, etc.
• LiDAR terrain data will provide higher definition
  accuracy
• Wider range of topographic data products can/will
  be developed
• Cost schedule advantages as compared to
  traditional collection techniques
• Mission collection parameters designed to
  generally meet FEMA LiDAR Guideline Specs (A.8)
  and PAMAP Standards
• Post processing and filtering of LiDAR point
  cloud to bare earth (removal of artifacts,
  outliers, buildings and vegetation)
• 18.5 cm vertical RMSE for bare earth elevation
  surface
• Contours 2-foot interval

3
LiDAR Deliverables

• Reports/Documents
• LiDAR Pre-Flight Plan, Sensor Settings, GPS
  Control Plan
• LiDAR system data report
• Post Mission Flight report
• Ground control report
• Data processing procedures
• Systems calibration report
• Data Deliverables (delivered in PAMAP 10K x 10K
  tile format)
• Raw LiDAR datasets including all original point
  cloud data in LAS format (unclassified)
• Classified LiDAR dataset in LAS format including
  point classification, intensity, etc.
• Raster surface at 3.2 grid of bare earth surface
  in GeoTIFF format
• 2 LiDAR contours and breaklines in ArcGIS 3-D
  shape file format

4
PAMAP LiDAR Program

• 2006 21 counties, 4929 tiles (15K square miles)
• 2007 20 counties, 4745 tiles (17K square miles)
• 2008 24 counties, 4278 tiles (15K square miles)

5
LiDAR Basics

• Airborne Light Detection And Ranging
• Very fast, accurate and cost effective technology
  to measure and quantify reflective surfaces
  (elevations)
• Systems Components
• Aircraft
• Crew (Pilot Instrument Operator)
• Laser w/ mirror
• Uses its own energy source (NIR red laser)
• Direct (active) acquisition of terrain
• Allowing day or night operation
• GPS Receivers (Aircraft Ground)
• Provides aircraft position
• Inertial Measurement Unit
• Provides aircraft orientation direction
• Post Processing Software Specialized Technicians

6
LiDAR Basics

• Mirror sweeps laser beam across the ground.
• Range to target is determined by measuring time
  interval between transmission and return of
  reflected laser pulse.
• Aircraft position is determined using GPS phase
  differencing techniques.
• Pointing direction of laser determined with
  Inertial Measuring Unit (IMU) and recording of
  mirror position.
• Data streams recorded and synchronized for post
  processing.

7
LIDAR Measures Objects From Line of Sight
8
LiDAR Basics

• LiDAR is indiscriminate
• Target must be visible- its not an all weather
  sensor
• Does not see through trees but around them
• It is not imagery but can be shaded to look like
  imagery
• High absorption rate of laser energy in water and
  new asphalt (no reflection)
• Newer systems can collect 150,000 elevation
  points per second!
• Capable of vertical accuracies of 15 cm RMSE or
  greater
• Does not automatically identify break lines
• Ability to collect multiple returns (4 or more)
  with both range (elevation) and intensity
• Multiple returns allows reflective surface
  mapping as well as creation of bare earth models

9
Multiple Returns Intensity

• Systems today have the ability to measure
  multiple returns and the intensity of the
  returned signal for each.
• This enables specialized applications using the
  LiDAR data.

10
LIDAR 1st and Last Return
1ST stop from canopy
Last stop from ground
11
Intensity Signal
Aerial Photo
Intensity Image
12
Value of Noise in LiDAR Data
13
Data Processing Steps

• Compute Aircraft Position Trajectory from
  GPS/IMU
• Calculate XYZ of laser points
• Merging and tiling of swaths
• Remove vegetation and other features
• Classification, editing thinning
• Integration of other terrain features (i.e.
  breaklines)
• Surface Development
• Data Formatting- LAS, ASCII, DTM, Contours
• Validation and inspection of results

14
(No Transcript)
15
Data Thinning

• Typically, LiDAR datasets are many times the size
  of a traditional DTM collection of gt 3
  Gigabytes per hour of flight.
• LiDAR data density constant regardless of
  topographic change.
• Some software may not be able to handle huge
  amounts of data from LiDAR.
• Data thinning may be necessary.

16
Bare Earth Model

• Significant editing must be employed to create a
  Bare Earth Model which models the natural
  ground.
• Some automated procedures may be used but also
  requires interactive editing (70/30). Imagery
  backdrop may be necessary.
• In some cases, traditional photogrammetry or
  LiDARgrammetry may be necessary to add
  breaklines.

17
(No Transcript)
18
LiDAR Mission Planning

• Project Requirements
• Quality of the hardware and software
• Knowledge of the planners, operators, office
  staff
• Flying height
• Scan angle (also important for vegetation
  building penetration)
• Laser power
• Laser rep rate
• GPS configuration
• Distance from base station to aerial platform
• Sensor Calibration

19
LiDAR Applications

• LiDAR has significant fixed cost to get the
  aircraft sensor to the project site but can be
  very cost effective for large area projects and
  corridor projects.
• Required accuracy use cases must be carefully
  evaluated.
• Appropriate for a wide range of project including
  bio mass/forestry, corridor studies, obstruction
  mapping, flood studies, city/county mapping
  projects.
• Local Users
• State of Pennsylvania
• PA Turnpike
• Allentown
• Reading
• Lancaster County
• FEMA Region 3
• USACE

20
LiDAR Accuracy

• Accuracy of elevation in range of 10 to 30
  centimeters (0.33 to 0.98 feet.)
• Accuracy of XY position in range of 11 to 46
  centimeters (0.36 to 1.51 feet.)
• Accuracy depends on
• pulse rate
• flying height
• GPS configuration
• location of ground stations
• position of the scanner with respect to nadir

21
LiDAR TestingCat 1-bare earth Cat
2-weeds/cropsCat 3-scrub Cat 4-forested
Cat 1 To 6 tall
Cat 2 To 3 tall
Cat 3 To 6 tall
Cat 4 gt6 tall
22
LiDAR Advancements

• Multi-Pulse Systems
• Huge Advancement in LiDAR
• Allows Higher Altitude Flights
• Terrain Changes Cause Many Problems
• Waveform Technology
• Provides opportunity for more accurate data
• Provides ability to resolve closely spaced
  targets
• Sensor Fusion

23
LiDAR Resources

• USGS- Center for LiDAR Information Knowledge
  (CLICK)
• http//lidar.cr.usgs.gov/
• ASPRS- LAS Data Standards-
• http//www.asprs.org/society/divisions/ppd/standar
  ds/lidar_exchange_format.htm
• ASPRS- DEM Users Manual
• https//eserv.asprs.org/eseries/source/Orders/inde
  x.cfm?activesectionorders

24
Processing LiDAR Data
25
(No Transcript)
26
LiDAR Processing Steps

• Perform initial classification on LiDAR LAS file
  ground vs. non-ground
• Generate Break Lines
• Integrate LiDAR Bare Earth and Breakline Data
• LiDAR data is removed from streets and hydro
• Further classify and refine LAS file
• Perform Model Key thinning on Bare Earth and
  then produce final files
• Produce Tin and filter with breaklines
• Produce Contour files and grids

27
LiDAR LAS File- Fields Key

• LAS file contains the following fields for each
  point
• Class
• Northing
• Easting
• Elevation
• Intensity
• Flight Line
• Time Stamp
• Echo Number 
• Echo Type

Class 1 Unclassified (default) Class 2
Ground Class 8 Model Key (thinned) Class 9
Hydro Class 12 Non-Ground Class 15 Road
edges
28
LiDAR LAS File - Class Key

• Class 1 (Default) - These are the points that are
  a mixture of the remaining points after the
  ground classification.  These would contain cars,
  buildings, parts of vegetation, possible ground
  (that doesn't add to the bare earth surface),
  etc.
• Class 2 (Ground) - These are points on the bare
  earth surface.  They are from the automated
  processing, as well as the manual surface review.
• Class 12 (Non-Ground) - These are points that are
  identified as first of many return or
  intermediate of many returns from the LIDAR
  pulse.  These are points that are most likely
  vegetation returns or points identified to be not
  on the ground surface.

29
LiDAR LAS File- Class Key

• Class 8 (Model Key) - These are the subset of
  Class 2 points that have been filtered using
  educated thinning process.
• Class 9 (Hydro) - These are points that fall
  within hydro features. 
• Class 15 (Road Edges) These are points that
  fall within 1.5 of road break lines. 

30
Raw LiDAR all points are on default class
31
Bare Earth processing ground is orange,
non-ground is green, white is default
32
Manually Generate Breaklines

• Stereo digitizing of relevant features
• Edge of pavement, double line streams, rivers,
  etc. (gt20 in width), ponds and lakes
• Directionality, bridges, intermediate points
  along features
• Integrate LiDAR Bare Earth and Breakline Data
• Add vertices at 25 intervals to all break lines
  and populate with LiDAR z value

33
LiDAR data is removed from street edges and
hydro

• LAS file is classified further and refined
• Remove LiDAR data falling
• within hydro area features (double line drains,
  ponds, lakes, rivers)
• within 1.5swaths on either side of break lines
  (road edges)
• on bridge decks

34
Perform Model Key thinning on Bare Earth and
then produce final files

• Model Key Class Development of Class 8
• Greatly reduces size of file by removing
  unnecessary points (actually leaves points on
  Class 2) and creates a Class of necessary
  points on Class 8
• Educated Thinning of Bare earth dataset for use
  in Contour and Grid Surface Production
• Thinning process results in the development of a
  Model Key bare earth subset
• Amount of thinning ultimately determines density
  of TIN smoothness of contours

35
Model Key filtering (keep red, lose orange)
36
Post Model Key filtering Class 8
37
Generate output files

• Bare Earth Surface Development
• Generate TIN surface from model key class and
  LiDAR processed break line data
• Use TIN surface to generate 2 contours for each
  PaMAP tile
• Use TIN surface to generate 3.2 raster grid
  format for each PaMAP tile and then convert into
  final GeoTIFF Raster format

38
Tin and Raster Surface Files
39
Preliminary TIN Surface Elevations are color
coded
40
Breaklines over preliminary surface
41
Contours generated from TIN displayed over color
surface
42
Contours only
43
(No Transcript)
44
Contour Smoothing Option No. 1
45
Contour SmoothingOption No. 2
46
Contour SmoothingOption No. 3
47
Contour Comparison
48
Anomalies in LiDAR Spikes
49
Left Clean of Artifacts, Poor DrainageRight
Clean Drainage, Poor Artifacts
Same area, 3 years later
50
Artifacts
51
Over-Smoothing
52
PAMAP Sample LiDAR Data
53
(No Transcript)
54
(No Transcript)
55
LiDAR Reflective SurfacePNC Park - Pittsburgh
56
City of Philadelphia- LiDAR Pilot
57
Statistics for Sample PAMAP 10,000 x 10,000 tile

• LAS File ( 155 MB)
• 5,675,866 - Total Points
• 2,392,159 - Bare Earth Points (42)
• 349,135 Model Key Points (6)
• ASCII File of LAS data (263 MB)
• LiDAR TIN (372 MB)
• 2 topographic contour shape file (29 MB)
• Break Line shape file (lt1 MB)