14/12/13

Working with LiDAR *.las files eCognition Com m unity

Working with LiDAR *.las files

View Edit History Actions

TABLE OF CONTENTS
1. Working with LiDAR *.las files 1. Loading LiDAR point cloud files in *.las format 1. Creating a new project using LiDAR data 2. Importing LiDAR data using the customized import 2. Using the LiDAR converter algorithm 1. Image Layer (Las file source) 2. Convert Parameter 3. Returns 4. Calculation 5. Filtering 6. Output Layer 3. Interpolating LiDAR data based on image objects

Working with LiDAR *.las files

When dealing with LiDAR processing and analysis, there are several components which provide you with means to directly load and analyze LiDAR point clouds as well as export raster result images such as DSM / DTM. Working with LiDAR point clouds in eCognition is implemented in a two step approach. 1. LiDAR point clouds are loaded and displayed using a quick resampling of the intensity data. 2. Once loaded additional layers can be generated using a special LiDAR converter algorithm. 3. Gaps within the LiDAR data can then be interpolated based on image objects

Figure: LiDAR intensity as displayed after loading

Figure: LiDAR first return as displayed after applying the LiDAR converter algorithm

Figure: LiDAR last return as displayed after applying the LiDAR converter algorithm

com m unity.ecognition.com /hom e/working-with-lidar-.las-files

1/4

14/12/13

Working with LiDAR *.las files eCognition Com m unity

Loading LiDAR point cloud files in *.las format

To allow for quick display of the point cloud, rasterization is implemented in a simple averaging mode based on intensity values. Complex interpolation of data can be done based on the LiDAR file converter algorithm.

In the loading process a resolution needs to be set which determines the grid spacing of the raster generated out of the las file. The resolution is set to 1 per default, which is the optimal value for LiDAR data with a point density of one point per square meter. For coarser resolution data, set the value to 2 or above, for higher resolution data set it to 0.5or below.

Creating a new project using LiDAR data

To create a new project using LiDAR data: 1. open the create project window 2. select the *.las file to open 3. set the resolution in which to load the las file and hit OK 4. continue loading as with any other dataset

Importing LiDAR data using the customized import

When importing LiDAR data using the customized import, ...

Using the LiDAR converter algorithm

After loading LiDAR data, you need to extract additional information from the dataset as required by your solution This can be the elevation, number of returns and more. To do so, use the LiDAR converter algorithm.

Image Layer (Las file source)

The image layer source defines the layer respectively the dataset from which to extract additional layers.

Convert Parameter
The convert parameter defines which information to use for generating new image layers. The following parameters can be set:

Mode
Intensity Elevation Class

Operation
The point intensity values are used to generate the raster layer The point elevation values are used to generate the raster layer The class information stored in the points is used to generate the raster layer
2/4

com m unity.ecognition.com /hom e/working-with-lidar-.las-files

14/12/13

Working with LiDAR *.las files eCognition Com m unity

Number of returns

A raster layer representing the number of returns is generated

Returns
Since the data is resampled into a 2D Grid with a fixed spacing, several LiDAR points can be within one pixel. The Return type allows to define which points to use in generating the raster. The following settings are available.

Mode
First Last All

Operation
Only first return points are used Only last return points are used All points are used

Calculation
Since the data is resampled into a 2D Grid with a fixed spacing, several LiDAR points can be within one pixel. The calculation mode allows determining how to compute the resulting pixel value out of several LiDAR points. The following modes are available:

Mode
Average Minimum Maximum Median (lower) Median (upper) Most frequent value

Calculation
The average out of all point values is calculated The minimum value out of all available point values is used The maximum value out of all available point values is used The lower median out of all available point values is used The upper median out of all available point values is used The most frequent value out of all available point values is used

Filtering
The filtering option allows limiting the points used for generating the raster representation based on available point properties. The following options are available:

Mode
No filtering Selected classes Non selected classes

Operation
No filtering is applied. This is the default setting All operations are applied to points of the selected classes only All operations are applied to points of the classes which are not selected only

If filtering is applied, a list of classes as standardized for the *.las file format is provided. Standard classes allow using predefined standardized class types, for custom classes the
com m unity.ecognition.com /hom e/working-with-lidar-.las-files 3/4

14/12/13

Working with LiDAR *.las files eCognition Com m unity

other classes selection is available.

Figure: Class selection of the LiDAR converter algorithm

Output Layer
The Output layer defines the layer name

Interpolating LiDAR data based on image objects

Interpolating LiDAR data based on image objects has distinct advantages: the interpolation can be freely adapted to the specific application in use ramping effects as seen in TIN based approaches can be avoided

Sample Rule Set: LiDAR sample project

com m unity.ecognition.com /hom e/working-with-lidar-.las-files

4/4