18
LiDAR and GIS: Classification and Feature Extraction Nicholas M. Giner – Esri

LiDAR and GIS: Classification and Feature Extraction · 2020. 7. 2. · Agenda •LiDAR basics •Data formats and management •The LAS dataset •Classifying LiDAR •Feature extraction

  • Upload
    others

  • View
    12

  • Download
    0

Embed Size (px)

Citation preview

Page 1: LiDAR and GIS: Classification and Feature Extraction · 2020. 7. 2. · Agenda •LiDAR basics •Data formats and management •The LAS dataset •Classifying LiDAR •Feature extraction

LiDAR and GIS:Classification and Feature ExtractionNicholas M. Giner – Esri

Page 2: LiDAR and GIS: Classification and Feature Extraction · 2020. 7. 2. · Agenda •LiDAR basics •Data formats and management •The LAS dataset •Classifying LiDAR •Feature extraction

Agenda

• LiDAR basics

• Data formats and management

• The LAS dataset

• Classifying LiDAR

• Feature extraction from LiDAR

- Building footprints / 3D buildings

- Trees

- Traffic lights using machine learning

Page 3: LiDAR and GIS: Classification and Feature Extraction · 2020. 7. 2. · Agenda •LiDAR basics •Data formats and management •The LAS dataset •Classifying LiDAR •Feature extraction

LiDAR basics• LiDAR – Light Detection and Ranging

- Optical remote sensing technique using laser light to densely sample the Earth’s surface, producing a point cloud of highly accurate x,y,zmeasurements

• Types of point clouds- Airborne scanned-based LiDAR

- Mobile / Tripod-based LiDAR

- Photogrammetric / Drone-based LiDAR

• LiDAR point cloud attributes- x,y,z measurements

- Intensity

- Return number

- Class code

- RGB

- Overlaps

Page 4: LiDAR and GIS: Classification and Feature Extraction · 2020. 7. 2. · Agenda •LiDAR basics •Data formats and management •The LAS dataset •Classifying LiDAR •Feature extraction

Data formats and management• Formats

- LAS / zLAS / LAZ

• Management

LAS Dataset

Mosaic Dataset

Terrain Dataset

Point Cloud Scene Layer

Page 5: LiDAR and GIS: Classification and Feature Extraction · 2020. 7. 2. · Agenda •LiDAR basics •Data formats and management •The LAS dataset •Classifying LiDAR •Feature extraction

Demo #1Explore a LAS file, Create a LAS Dataset

Page 6: LiDAR and GIS: Classification and Feature Extraction · 2020. 7. 2. · Agenda •LiDAR basics •Data formats and management •The LAS dataset •Classifying LiDAR •Feature extraction

The LAS Dataset• LAS Dataset

- “Container” for storing reference to many LAS/zLAS files on disk

- Pointer to the original LAS/zLAS files

- Quick to create, small in file size, easy to update with additional LAS/zLAS files

- Can reference surface constraints (breaklines / boundaries)

- Quick display of LAS/zLAS data as point clouds or a dynamic TIN in 2D or 3D

- Excellent for QA/QC of LiDAR coverage (point density/spatial extent)

- Basis for generated products such as DEMs and DSMs or TINs

Page 7: LiDAR and GIS: Classification and Feature Extraction · 2020. 7. 2. · Agenda •LiDAR basics •Data formats and management •The LAS dataset •Classifying LiDAR •Feature extraction

Working with LiDAR - GeoprocessingData Management (6) 3D Analyst (18) Conversion (2)

Page 8: LiDAR and GIS: Classification and Feature Extraction · 2020. 7. 2. · Agenda •LiDAR basics •Data formats and management •The LAS dataset •Classifying LiDAR •Feature extraction

Classifying LiDAR• Automated classification (Geoprocessing)

- Change Class Codes

- Classify Buildings

- Classify Ground

- Classify Noise

- Classify Overlap

- Classify by Height

- Classify by 3D Proximity

- Classify by 2D Proximity (using features)

- Classify using raster

• Interactive classification (Manual)

• Editing considerations

Page 9: LiDAR and GIS: Classification and Feature Extraction · 2020. 7. 2. · Agenda •LiDAR basics •Data formats and management •The LAS dataset •Classifying LiDAR •Feature extraction

Workflow: Classifying LiDAR

Unclassified LAS

Classify LASOverlap

Create LASDataset

Classify LASNoise

Classify LASGround

Create DEMClassify LASBuildings

Classify LASby Height (Vegetation)

Colorize LAS

ClassifiedLAS Dataset

Interactivelyedit class codes

Page 10: LiDAR and GIS: Classification and Feature Extraction · 2020. 7. 2. · Agenda •LiDAR basics •Data formats and management •The LAS dataset •Classifying LiDAR •Feature extraction

Demo #2Classify LAS data

Page 11: LiDAR and GIS: Classification and Feature Extraction · 2020. 7. 2. · Agenda •LiDAR basics •Data formats and management •The LAS dataset •Classifying LiDAR •Feature extraction

Other tools for classification• Change LAS Class Codes

- Reassigns classification codes from one class to another

Page 12: LiDAR and GIS: Classification and Feature Extraction · 2020. 7. 2. · Agenda •LiDAR basics •Data formats and management •The LAS dataset •Classifying LiDAR •Feature extraction

Other tools for classification• Classify by 2D and 3D Proximity (vector)

Set LAS Class Codes using Features (2D) Locate LAS Points by Proximity (3D)

Page 13: LiDAR and GIS: Classification and Feature Extraction · 2020. 7. 2. · Agenda •LiDAR basics •Data formats and management •The LAS dataset •Classifying LiDAR •Feature extraction

Other tools for classification• Classify using raster

- Requires integer raster with pixel values representing ASPRS LiDAR class codes

Page 14: LiDAR and GIS: Classification and Feature Extraction · 2020. 7. 2. · Agenda •LiDAR basics •Data formats and management •The LAS dataset •Classifying LiDAR •Feature extraction

Workflow: Extract building footprints

Classified LAS Dataset

Make LAS Dataset LayerClass Code = 6

LAS Point Statistics as RasterMost Frequent Class

Elevation VoidFill (Optional)

Raster to PolygonNo Simplify

SQL(Remove

small polygons)

EliminatePolygon Part(Fill holes)

Raw buildingpolygons

RegularizeBuildingFootprints

Clean up(Optional)

Building Footprints

Page 15: LiDAR and GIS: Classification and Feature Extraction · 2020. 7. 2. · Agenda •LiDAR basics •Data formats and management •The LAS dataset •Classifying LiDAR •Feature extraction

Workflow: Extract traffic lights

Point CNNClassified LAS Dataset

LAS to Multipoint

Explode multipoints

Reduce datasetsize

DBSCANCluster boundariesAdd fields

Remove noisepoints

Calculatefields

Clean up noise polygons

Final outputpoints

Page 16: LiDAR and GIS: Classification and Feature Extraction · 2020. 7. 2. · Agenda •LiDAR basics •Data formats and management •The LAS dataset •Classifying LiDAR •Feature extraction

Demo #3Feature extraction: Building footprints, trees, traffic lights

Page 17: LiDAR and GIS: Classification and Feature Extraction · 2020. 7. 2. · Agenda •LiDAR basics •Data formats and management •The LAS dataset •Classifying LiDAR •Feature extraction

Print Your Certificate of Attendance

Print Stations Located in 150 Concourse Lobby

Tuesday12:30 pm – 6:30 pmExpoHall B

5:15 pm – 6:30 pmExpo SocialHall B

Wednesday10:45 am – 5:15 pmExpoHall B

6:30 pm – 9:30 pmNetworking ReceptionSmithsonian National Museumof Natural History

Page 18: LiDAR and GIS: Classification and Feature Extraction · 2020. 7. 2. · Agenda •LiDAR basics •Data formats and management •The LAS dataset •Classifying LiDAR •Feature extraction

Download the EsriEvents app and find

your event

Select the session you attended

Scroll down to “Survey”Log in to access the

surveyComplete the survey and select “Submit”

Please Share Your Feedback in the App