Feature extraction from LiDAR and derived rasters

RobertFord1 · ‎06-21-2016

Good morning all!

I am trying to obtain heights of man made structures and also update / create building footprints using LiDAR and the rasters I have created from them. I have gone through the this tutorial that explains how to create a DSM and DEM from LiDAR data (so I now have a DSM and DEM) and also this tutorial that explains how to get heights of trees (I just used the minus tool for the entire area and have heights for the entire area).

The issues I am running into for heights are the outliers in the LiDAR data. I have tried using the Locate Outliers tool, but it just keeps failing saying it can't locate my output table. These outliers skew data greatly when it comes to calculating average heights for buildings (I tested it on an existing polygon). I would like to reclassify these outliers as noise somehow if possible.

As for feature extraction, I really am a new guy when it comes to intense LiDAR analysis in ArcGIS and just am not sure where to begin. I know there are all kinds of paid for 3rd party extensions for LiDAR that include feature extraction, but alas we do not have the budget for such things....

Any help is much appreciated!

Rob

ArthurCrawford · ‎06-22-2016

Hello Jake,

The below works, but I have a new tools set, demo data and training material for an improved process. Here's the link to it, it requires ArcGIS Pro 2.0, Spatial Analyst and 3D Anaylst:

http://esriurl.com/3DMappingWithLidar

First question is what is the spacing of your lidar? If it's less than 3ft, you should be able to extract buildings from it. They will not be perfect. If you buildings are classified with a class code of 6 for buildings, I would suggest this process assuming your data is in feet. Some of the below is from Clayton Crawford on the 3D Team.

1. LASPointStatisticsAsRaster (LAS layer filter set to class code 6, building,
points)

2. RasterToPolygon (no simplification)

3. RegularizeBuildingFootprint:

a. Run Regularize Building Footprint as Circle with a tolerance of 8 and minimum
radius of 0.1/max radius 100000 and export those with a status of 0 and area
over 275 sq. meters to Large Circles.

b. Run Regularize Building Footprint as Circle with a tolerance of 3 and minimum
radius of 0.1/max radius 100000 and export those with a status of 0 and area
over 25 sq. meters and under 275 sq meters to Small Circles.

c. Combine those circles and use select by location to remove those polygons from
processing.

d. The rest divide into three groups: Large, Medium and Small buildings.

Large: "Shape_Area" >= 2300 Medium: "Shape_Area" >= 464 AND "Shape_Area" < 2300 Small: "Shape_Area"< 464

Large run Regularize Building Footprint as Right Angles and Diagonals with tolerance 4, densification 4,
Precision 0.25, Diagonal Penalty of 1.5.

Medium run Regularize Building Footprint as Right Angles and Diagonals with tolerance 3, densification 3, Precision 0.25, Diagonal Penalty of 1.5.

Small run Regularize Building Footprint as Right Angles with tolerance 2, densification 2, Precision 0.25

One approach with buildings not classified:

Classify LAS Ground

ClassifyLASBuilding (Pro 1.3) or the 3D Samples tools for Classify Rooftop if you don't have access to 1.3 (might want to wait).

LAS Point Statistics As Raster (LAS layer filter set to class code 6, building, points)

RasterToPolygon (no simplification)

Regularize Building Footprint (see above)

If you have parcels, you can do a union on them with the output polygons and select out those with a -1 for fid for the parcel before running Regularize Building Footprint. This will help with dividing lidar buildings to individual buildings in downtown areas and divide joined buildings (vegetation) between properties where the buildings are close. I have also found doing a negative buffer, mulitpart to singlepart and then a positive buffer to help get rid of trees next to buildings that get captured. Sometime I run a statistics function on the output of the LASPointStatisticsAsRaster and then a eliminate polygon part after the Raster to Polygon to get rid of interior parts that are not desired.

Thanks, hope this helps.

Arthur Crawford - Content Team - Living Atlas

View solution in original post

JakeSkinner · ‎06-21-2016

Hi Robert,

You can use the LAS Dataset Profile View window to reclassify the outliers. See the following help document for more information:

LAS dataset Profile View—Help | ArcGIS for Desktop

RobertFord1 · ‎06-21-2016

Thanks Jake! I am looking for a way to locate outliers on a mass scale. That's why the locate outliers tool looked like the tool I needed. My study area is county wide so going through one by one is not really an option. Although I will definitely use this for looking at buidlings in question later.

AdrianWelsh · ‎06-21-2016

Rob,

Try converting your LAS Dataset into a TIN

LAS Dataset To TIN—Help | ArcGIS for Desktop

Then you can run the Locate Outliers Tool.

Though, from looking at the tool, it looks like it will work on an LAS Dataset as well:

Locate Outliers—Help | ArcGIS for Desktop

"Summary

Identifies anomalous elevation measurements from terrain, TIN, or LAS datasets that exceed a defined range of elevation values or have slope characteristics that are inconsistent with the surrounding surface."

Have you created an LAS Dataset yet from your LAS data?

Create LAS Dataset—Help | ArcGIS for Desktop

RobertFord1 · ‎06-21-2016

Adrian,

I have been using my 13,000 .las files in a LAS Dataset then attempting to feed that into the Locat Outliers tool input. I have discovered that it completes on a small LAS Dataset but not the full one. Does anyone know a of a faster way to break up my enormous folder of .las files into small datasets and iterate through them while using locate outliers?

Thanks again guys!

JakeSkinner · ‎06-21-2016

You can use python to iterate through your LAS files. Below is an example. It first obtains a spatial reference from an existing las dataset. The code then iterates through all las files in a directory, creates a LAS Dataset, and then runs the Locate Outliers tool with a min/max of 80/325, and an outlier cap of 5,000,000. It outputs the feature classes to a single geodatabase that you can then merge together.

import arcpy, os
from arcpy import env
env.overwriteOutput = 1
arcpy.CheckOutExtension('3D')
env.workspace = r"D:\Temp\Python\UserData"

dataset = "las1.lasd"
spatial_ref = arcpy.Describe(dataset).spatialReference

x = 1
for lasFile in arcpy.ListFiles("*.las"):
    arcpy.CreateLasDataset_management(lasFile, "lasDataset.lasd", "NO_RECURSION", "", spatial_ref)
    outlierFC = env.workspace + os.sep + "Data.gdb" + os.sep + "LocateOutliers_" + str(x)
    arcpy.LocateOutliers_3d("lasDataset.lasd", outlierFC, "APPLY_HARD_LIMIT", "80", "325", "APPLY_COMPARISON_FILTER", "0", "150", "0.5", "5000000")
    x += 1

arcpy.CheckInExtension('3D')

RobertFord1 · ‎06-22-2016

Thank you 1000 times Jake! It's cranking away at it now. While it's doing that, how would I go about extracting features from the resulting rasters? Once the outliers are removed, I will then follow the steps I mentioned in my intial post to get heights (using minus tool). So now I have to extract man made features from this namely buildings.

Thanks again!

ArthurCrawford · ‎06-22-2016