Auto-Classify Noise
Global Mapper has several tools that allow for automated classification
and feature extraction. For each analysis tool, there are default recommended
input values, along with the option to customize input values for optimizing
either processing speed or accuracy.
Auto-Classify
Non-Ground Points
Global Mapper's lidar auto classification tools allow you to identify and classify noise, ground and non-ground points from unclassified point clouds. When you run auto classifications on multiple input files, the same classification parameters are applied to each input file. If ground conditions and quality of lidar points vary by file, then files should be classified separately. The accuracy of auto-classification results will depend on the the level of detail and the quality of point cloud data, in addition to the operator's knowledge of ground/terrain conditions. Post-processing and Quality Assurance steps can be used to optimize classification settings when defaults do not yield satisfactory results. Global Mapper's Path/Profile tool will have the Lidar Toolbar available, so that manual classification of points in a profile view may be conducted.
Outlying elevation values can lead to spurious or inaccurate classification results. Laser pulse returns from obstructions such as haze, air craft, or birds or multiple reflections from tree canopies can result in outlying points with very high elevation values. Ground structures and trees or tree canopies can create multiple reflections, leading to excessively long travel times back to the lidar sensor and creating points with outlying low elevation values. These points with anomalous high and low elevation values are noise points, and can be most easily and quickly seen by coloring or drawing point clouds by elevation. The noise points will expand the expected elevation range for the area. If your lidar data has noise points, and you plan on using the Auto-Classify Ground Points and Auto-Classify Non-Ground Points tools, you will want to Auto-Classify Noise points first. You can also filter out extreme spikes by using the option to 'Delete Samples Over _ Standard Deviations from the Mean' in the Lidar Load Options menu.
Most lidar data will contain a certain percentage of ground points, along with a number that are unclassified. The Auto-Classification Ground Points tool can be used to identify previously unclassified ground points for use in eliminating unclassified points as possible building or tree features, or for employing in the creation of a digital terrain model. Once ground points have been classified a different algorithm may be employed to identify previously unclassified non-ground features, such as buildings, trees, and powerlines by using the Auto-Classify Non-Ground Points. In the classification of non-ground points, relatively flat surfaces that are above the height determined to be ground height will be classified as buildings and those that are vertically offset from neighboring points by user-defined parameters will be classified as trees or powerlines.
The typical workflow for auto-classification of unclassified lidar points is to classify noise points, ground points, and then non-ground points. Identifying previously unclassified noise points will improve the ground auto-classification results, and classifying previously unclassified ground points will improve the auto-classification of non-ground points. Some of the classification dialogs and will allow for user defined bin size settings, the bin size you want to set is based on the resolution of your Lidar data.
Automatic Noise Classification
To automatically detect and remove likely noise points from loaded or selected Lidar data, select the Auto-Classify Noise Points button on the Lidar Toolbar.
The Auto-Classify Noise points tool will calculate a sub-sampled grid and in addition to calculating min/max/avg average and standard deviation grids. Anything more than a user-identified number of standard deviations away from a local average hight will classify points as noise.
Select Unclassified Point Cloud(s) to Find Likely Noise Points In - If more than one Lidar data set
is loaded into workspace, specific Lidar layers may be selected, or unselected,
for classification. Check the option to 'Only Classify Lidar Points Selected
in Digitizer Tool' to run noise classification tool only on points selected
by the Digitizer.
Maximum Allowed Variance from Local Average: Specify the maximum allowed variance from the local average of height values for the loaded or selected points in standard deviations. The local area is defined as a 5x5 chunk 128 times the sample spacing in each direction. So if the cloud resolution is about 0.1 meters per pixel, the local group would be a 5x5 set of sample cells 12.8 meters on each side.
Mark Noise Points as Withheld/Deleted Rather than Reclassifying - Mark the selected or loaded points to be withheld, rather than classifying.
Mark Noise Outside Elevation Range - Manually define an elevation range, outside of which points will be defined as noise.
Mark Noise Outside Height Above Ground Range - Manually define a height above ground range, outside of which points will be defined as noise.
Identify Likely Noise Points from Points that are Already Classified- This will also consider points that have existing classifications as possible noise points.
Reset Existing Noise Point to Unclassified at Start - Resets existing noise points to unclassified.
Specify Bounds... - Specify the bounds of the classification operation by drawing a box or using coordinates.
Filter Points by Elev/Class/etc... - Use the Filter Lidar by Class options to define the operation.
Restore Defaults - Restore defaults for running classification, using all loaded or selected points.
Automatic Ground Classification
The Auto-Classify Ground Points
tool on the Lidar Toolbar brings up the Automatic
Classification of Ground Points settings window (below). These
values may be changed according to optimize the model output based on
the local terrain, the range of elevation values in the data set, user-defined
preferences for filtering points prior to auto-classification or known
features in the landscape. 
Select Unclassified Point Cloud(s) to
find Likely Ground Points In - If more than one Lidar data set
is loaded into workspace, specific Lidar layers may be selected, or unselected,
for classification. Check the option to 'Only Classify Lidar Points Selected
in Digitizer Tool' to run classification tool only on points selected
by the Digitizer.
Base Bin Size to Check for Curvature
Deviations - This may be set to Point Spacings or Meters, smaller
distances will result in higher accuracy but will require higher resolution
lidar data. Those working with low resolution data, can use a higher value.
Minimum Height Departure from Local Mean
for Non-Ground Point - Use this setting to change the minimum height
above the (averaged) minimum elevation that a point should fall above
in order to not be classified as a ground point.
Removal of Likely Non-Ground Points
- These parameter settings allow the user to control the values used for
the removal of points that are not likely to be ground points. For areas
with high built structures, use larger values.
Reset Existing Ground Points to Unclassified
at Start - Resets the Unclassified Ground Point data, resetting
any points classified as ground. Removes all manual and automatic classification
of ground points in selected point data.
Specify Bounds... Manually specify
the bounds for applying auto-classification of ground points. May use
North, West, South, East coordinates to define coordinates, or draw an
area feature to define classification boundaries.
Filter Points...Contains additional
settings for filtering by points for classification by elevation and color
values, Source ID and existing classification. The user may also exclude
all points outside of a specified scan angle.
Restore Defaults - Restores the default settings for the Auto-Classify Ground Points tool.
Automatic Non-Ground ClassificationThe Auto-Classify Non-Ground Points tool allows the user to automatically classify Building , Vegetation, and Powerline Points. Clicking this option brings up the Automatic Classification of Non-Ground Lidar Points settings window (below).
There are a few different ways to select lidar points for classification. Use the options in the Select Unclassified Point Cloud(s) to find Likely Ground Points In if more than one Lidar data set is loaded into workspace, specific Lidar layers may be selected (check in box), or unselected (box empty), for automated classification. Check the option to 'Only Classify Lidar Points Selected in Digitizer Tool' to run classification tool only on points selected by the Digitizer tool. Use 'Specify Bounds...' to set the bounds for the classification by drawing a box on the workspace, using coordinate extents, or a selected area feature.
The setup options found under Building/High Vegetation Classification and Powerline Classification Setup can be used to adjust the classification parameters to ground conditions and the density and quality of the lidar points.
This setting specifies the size of each bin (meters per edge) when evaluating points to see if they are clustered as needed for the powerline or building/high vegetation classification algorithms. Smaller distances will result in higher accuracy but will also require higher resolution lidar data. Those working with low resolution data can use a higher value. For Building/ High Vegetation Classification the bin size to check for planar points may be set using either Point Spacings or Meters.
Base Bin Size to Check for Planar Points - This setting specifies the size of each bin using either Point Spacings or Meters when evaluating points to see if they are clustered as needed for the building/ high vegetation classification algorithm. For example, using Point Spacings a value of "3.0" would make each square bin 3 times the calculated native spacing of the point data. If you want to specify a spacing in meters rather than as a multiple of the native spacing for the point cloud, select 'Meters' from the drop down menu.
Minimum Height Above Ground - Use this setting to specify the minimum height above ground that a point has to be in order to consider it as a possible building or high vegetation point.
Maximum Co-Planar Distance - Use
this field to set the maximum co-planar distance to use for classifying
non-ground points in meters. Specifies the maximum RMSE (root mean square error) in meters from a best-fit local plane that the points in a small region all have to be within, in order to consider the region a potential planar (building) region.
Minimum Vegetation Distance - Use
this field to set the maximum distance (in meters) that the points in a small region all have to be within, in order to consider the region a potential vegetative region. Used for auto-classifying non-ground,
vegetation points.
Max Co-Planar Angle Difference - Use this field to set the maximum angle difference (in degrees) to be used when auto-classifying non ground points. Specifically, this will determine the maximum angle (in degrees) between adjacent best-fit planes such that they can still be considered part of the same plane when identifying flat building surfaces.
Minimum Height Above Ground - Use this field to specify the minimum height above ground that a point has to be in order to consider it as a possible powerline point.
Maximum Distance from Best Fit Line - Use this field to set the maximum distance from best fit line, specifying the maximum distance (in meters) from the best-fit 3D line of points with similar elevations in a bin that any points can be and still be considered powerlines.
Bin Size to Check for Planar or Linear Points - This setting the size of each bin (meters per edge) when evaluating points to see if they are clustered as needed for the powerline classification algorithm.
Maximum Height Change per Meter - Use this field to specify the maximum difference in elevation allowed per meter to consider points as possibly part of the same powerline segment. The default value is 0.5m, which allows for a change in elevation of 0.5m over a 1m distance between points. You might specify a slightly larger value if your data is noisy.
Resets the Unclassified Non-Ground Point data, resetting
any points classified as non-ground. Removes all manual and automatic
classification of ground points in selected point data, setting all points to unclassified.
Lidar Feature ExtractionGlobal Mapper allows you to automate the process of locating features on
the ground by using Lidar data in the automated feature extraction tools
available with the Lidar Module. The default parameter values can be changed
by the user for improved results.
The Extract Vector Features tool
can be used on classified ground points, either an entire point cloud
or a user defined subsection. This lidar feature extraction tool allows
the user to derive features such as building footprints, building roof
structure, power lines and other structures from classified Lidar ground
points.
Optional settings for extracting building roof and tree top polygons are
available in the Lidar Feature Extraction
Settings menu (below).
Select Point Cloud(s) to Extract Features
From - If more than one Lidar data set is loaded into workspace,
specific Lidar layers may be selected, or unselected, for automated feature
extraction. Check the option to 'Only Extract From Lidar Points Selected
in Digitizer Tool' to run classification tool only on points selected
by the Digitizer.
Resolution to Extract at: This may be set to Point Spacings or Meters, smaller distances will result in higher accuracy but will require higher resolution lidar data. Those working with low resolution data can use a higher value.
The average Point Spacings comes from the calculated point cloud statistics. For consistent and repeatable results when working with multiple lidar files, it is recommended to specify an exact resolution for the extraction in meters, rather than using the calculated Point Spacings.
Extract 3D Building Footprints - Select
this for extracting building footprints and other, similar features from
classified ground points.
Create Separate 3D Areas for Different
Roof Pieces if Possible - Select this option to identify roof features,
this requires high resolution Lidar data and may not be possible with
all data sets.
Max Co-Planar Angle Difference:
Use this field to set the maximum co-planar angle difference in degrees
for use for extracting separate roof pieces. For lower resolution data,
use a higher value and for higher resolution data, use a lower value.
Max Distance From Adjacent Plane: Use this field to set the maximum distance in meters from adjacent planes to use for extracting roof piece features. For lower resolution data, use a higher value and for higher resolution data, use a lower value.
Create Side Wall Area Extending to Ground
- Select this to create 'side walls', which extrude to ground from each
roof area edge.
ADVANCED: Simplify Multiplier for Smoothing
Buildings: Select this option to simplify the building outlines.
The specified value is a multiplier of the bin spacing for the feature
extraction operation. For example, if the extraction is set at 1 point
spacing and a simplification of 2 is used, the simplification factor is
then 2 point spacings.
Extract Tree Points - Select this for extracting trees from
classified ground points, this option creates point features that are
tree tops.
Minimum Tree Height - The minimum height above ground for a tree.
Minimum Tree Spread -The minimum width (edge-to-edge) of a single tree.
Maximum Tree Spread - Specify the maximum tree (canopy width expected per tree) spread in meters.
Select Create Approximate Tree Coverage
Polygons to extract tree top polygons.
Extract Powerline Features - Select this option to extract powerline features from loaded or selected lidar points.
Maximum Distance from Best Fit Line - Define the maximum allowable distance from the mathematical determined Best Fit Line of Classified Powerline points, in meters.
Maximum Angle Delta Allowed - Maximum angle between points to define a colinear powerline feature.
Minimum Powerline Length to Keep - Define shortest allowable powerline segment.
