LASmoons: Stéphane Henriod

Stéphane Henriod (recipient of three LASmoons)
National Statistical Committee of the Kyrgyz Republic
Bishkek, Kyrgyzstan

This pilot study is part of the International Climate Initiative project called “Ecosystem based Adaptation to Climate change in the high mountainous regions of Central Asia” that is funded by the Federal Ministry for the Environment, Nature Conservation, Building and Nuclear Safety (BMU) of Germany and implemented by the Deutsche Gesellschaft für Internationale Zusammenarbeit (GIZ) GmbH in Kyrgyzstan, Tajikistan and Kazakhstan.

lasmoons_Stephane_Henriod_1

Background:
The ecosystems in high mountainous regions of Central Asia are characterized by a unique diversity of flora and fauna. In addition, they are the foundation of the livelihoods of the local population. Specific benefits include clean water, pasture, forest products, protection against floods and landslides, maintenance of soil fertility, and ecotourism. However, the consequences of climate change such as melting glaciers, changing river runoff regimes, and weather anomalies including sharp temperature fluctuations and non-typical precipitation result in negative impacts on these ecosystems. Coupled with unwise land use, these events damage fragile mountain ecosystems and reduce their regeneration ability undermining the local population’s livelihoods. Therefore, people living in rural areas and directly depending on natural resources must adapt to adverse impacts of climate change. This can be done through a set of measures, known in the world practice as ecosystem-based adaptation (EbA) approach. It promotes the sustainable use of natural resources to sustain and enhance the livelihood of the population depending on those resources.

lasmoons_Stephane_Henriod_2 Goal:
In two selected pilot regions of Kyrgyzstan and Tajikistan, planned measures will concentrate on climate-informed management of ecosystems in order to maintain their services for the rural population. EbA always starts with identifying the vulnerability of the local population. Besides analyzing the socio-economic situation of the local population, this includes (1) assessing the ecological conditions of the ecosystems in the watershed and the related ecosystem services people benefit from, (2) identifying potential disaster risks, and (3) analyzing glacier dynamics with its response to water runoff. As a baseline to achieve this and to get spatially explicit, remote sensing based techniques and mapping activities need to be utilized.

A first UAV (unmanned aerial vehicle) mission has taken place in the Darjomj watershed of the Bartang valley in July 2016. RGB-NIR images as well as a high-resolution Digital Surface Model have been produced that now need to be segmented and analysed in order to produce comprehensive information. The main processing that will take advantage of LAStools is the generation of a DTM from the DSM that will then be used for identifying risk areas (flood zones, landslides and avalanches, etc.). The results of this approach will ultimately be compared with lower-cost satellite images (RapidEye, Planet, Sentinel).

Data:
+ High-resolution RGB and NIR image (10 cm) from a SenseFly Ebee
+ High-resolution DSM (10 cm) from a SenseFly Ebee

LAStools processing:
1)
classify DSM points obtained via dense-matching photogrammetry into a SenseFly Ebee imagery into ground and non-ground points via processing pipelines as described here and here [lastile, lassort, lasnoise, lasground]
2) create a DTM [las2dem, lasgrid, blast2dem]
3) produce 3D visualisations to facilitate the communication around adaptation to climate change [lasview]
lasmoons_Stephane_Henriod_0

New ‘laspublish’ creates Web Portals for 3D Viewing and Downloading of LiDAR

PRESS RELEASE (for immediate release)
February 18, 2016
rapidlasso GmbH, Gilching, Germany

Just in time for ILMF 2016, the makers of the open LASzip LiDAR compressor annouce the latest addition to their LiDAR processing software LAStools. The new ‘laspublish‘ from rapidlasso GmbH creates stand-alone Web Portals for interactive 3D viewing of LiDAR points and for selective downloading of LAZ or LAS files. The new tool is based on the cutting-edge streaming point cloud viewing technology of Potree that optimizes large LiDAR point clouds for streaming via the Web such that anyone can visualize, explore, and (optionally) download them with any modern browser.

3D viewer and download portal created with 'laspublish'

3D viewer and download portal created with ‘laspublish’

The interactive 3D viewer streams on demand only the relevant parts of the point clouds. It not only visualizes the LiDAR in many useful and intuitive ways, but is also equipped with measurement tools to calculate distances or areas and profile or clipping tools for close-up inspections. With its integrated LASzip compression, options to color LiDAR by classification, return type, intensity, and point source ID, stunning visuals via Eye Dome Lighting (EDL), and the optional 2D download map, the new ‘laspublish‘ empowers professional and novice users alike to create stand-alone LiDAR Webportals with just a few button clicks.

a few clicks and 'laspublish' creates a professional LiDAR portal

a few clicks and ‘laspublish’ creates a professional LiDAR portal

The new ‘laspublish‘ is now an integral part of the LAStools software and is bundled together with all necessary components of the Potree software. It provides an instant and cost-effective solution for generating a set of Web pages that realize a self-contained LiDAR portal offering interactive online visualization and exploration as well as easy and intuitive distribution of large LiDAR data sets via the optional download map.

About rapidlasso GmbH:
Technology powerhouse rapidlasso GmbH specializes in efficient LiDAR processing tools that are widely known for their high productivity. They combine robust algorithms with efficient I/O and clever memory management to achieve high throughput for data sets containing billions of points. The company’s flagship product – the LAStools software suite – has deep market penetration and is heavily used in industry, government agencies, research labs, and educational institutions. Visit http://rapidlasso.com for more information. As the only Diamond sponsor, rapidlasso GmbH has been the main financial supporter of the open source Potree package by Markus Schütz over the past two years.

About Potree:
Potree is a WebGL based viewer for large point clouds. The project evolved as a Web based viewer from the Scanopy desktop point cloud renderer by TU Wien, Institute of Computer Graphics and Algorithms. It will continue to be free and open source with a FreeBSD license to enable anyone to view, analyze and publicly share their large datasets. Visit http://potree.org for more information.

Potree puts Big and Beautiful LiDAR in Your Browser

PRESS RELEASE (for immediate release)
September 14, 2015
rapidlasso GmbH, Gilching, Germany

Just in time for INTERGEO 2015, the Potree software was released in its latest 1.3 version. Potree is a WebGL based point cloud viewer for very large datasets. The Potree software allows to publish large LiDAR point clouds on the Web such that anyone can explore the data with nothing more but a modern browser. The interactive 3D viewer not only visualizes the LiDAR in many useful and intuitive ways but also comes with tools to perform various measurements. As its only Gold Sponsor, rapidlasso GmbH is the main supporter of this powerful open source package by Markus Schütz.

17.7 billion points around San Simeon,, CA courtesy of Open Topography

17.7 billion points from San Simeon, CA courtesy of Open Topography

The long-term sponsorship of rapidlasso GmbH has directly supported a number of useful features such as the integration of our award-winning LASzip compressor using the pure javascript version contributed by Hobu Inc, optimization for massive airborne LiDAR data, profile selection, tools for distance and area measurements, options to color by classification, return type, and point source ID, and a clipping tool. The particular features sponsored in the most recent 1.3 release of Potree are the incredible Eye Dome Lighting (EDL) and faster data conversion for large data sets. A number of interesting showcases (including the CA13 example shown here) are available on the Potree page.

In the near future the Potree software will be distributed together with the LAStools package to offer a one-click solution for generating Webportals that host and distribute large LiDAR data sets and offer interactive online visualization and exploration. Potree is open source software that is free for anyone to acquire and to deploy. Please remember that using open source software is not the same as supporting open source software. Given the positive experience that rapidlasso GmbH has had with Potree we can only encourage other geospatial companies to support with time or money those open source projects that help your business.

About rapidlasso GmbH:
Technology powerhouse rapidlasso GmbH specializes in efficient LiDAR processing tools that are widely known for their high productivity. They combine robust algorithms with efficient I/O and clever memory management to achieve high throughput for data sets containing billions of points. The company’s flagship product – the LAStools software suite – has deep market penetration and is heavily used in industry, government agencies, research labs, and educational institutions. Visit http://rapidlasso.com for more information.

About Potree:
Potree is a WebGL based viewer for large point clouds. The project evolved as a Web based viewer from the Scanopy desktop point cloud renderer by TU Wien, Institute of Computer Graphics and Algorithms. It will continue to be free and open source with a FreeBSD license to enable anyone to view, analyze and publicly share their large datasets. Visit http://potree.org for more information.

Removal of Cloud Returns With a Coarse DTM

Flying LiDAR in regions with frequent cloud cover presents a significant challenge. If flight plan constraints do not allow to stay below all of the clouds then some of them will be scanned from above. For denser clouds this often means that all of the laser’s energy gets reflected or absorbed by the cloud and no returns on the terrain are generated. Clouds of points that are all cloud points can spell trouble in subsequent processing steps like automated classification with lasground. This is especially true for large dense clouds that start to look like features.

lasheight_masbate_lidar_clouds

scanned clouds in four neighboring LiDAR flight lines

Airborne LiDAR surveys are often carried out to create an improved Digital Terrain Model (DTM) with higher resolution than previous elevation products. Hence, usually there is already some lower resolution model that – as we will see in the following – can be used to robustly remove or mark all cloud points, at least those sufficiently high above this older ground approximation. We use data from the DREAM LiDAR Project in the Philippines who often acquire LiDAR in areas with a lot of cloud cover.

Our input are the 4 very short LiDAR strips in LAZ format shown above with lots of cloud returns and a coarse Aster DTM in ASC format at 50 m resolution. We will classify all LiDAR points far above the Aster DTM because they correspond to cloud returns. We need to be very conservative (because the Aster DTM is coarse and inaccurate) and only remove points that are really far above the Aster DTM whose ASC header is shown below.

ncols         2109
nrows         2162
xllcorner     512546.427859
yllcorner     1290961.682335
cellsize      50
NODATA_value  -9999
-9999 -9999 -9999 -9999 -9999 [...]

First we convert the Aster DTM from the inefficient ASC format to the efficient LAZ format using lassort. Why lassort? Because that puts the rasters that are on-the-fly converted to points on a grid into a spatially coherent order. This will allow efficient area-of-interest (AOI) queries once we have LAXxed the file with lasindex.

>> lassort -i masbate.asc -o masbate.laz
>> dir masbate*
    10,993,886 masbate.asc
     2,120,442 masbate.laz

Next we create a spatial indexing for ‘masbate.laz’ with lasindex. The default granularity of lasindex is 100 by 100 meter because it is set for airborne LiDAR. Given that there is only one point every 50 by 50 meters in the point cloud grid of the Aster DTM we increase the granularity to 1000 by 1000 meters.

>> lasindex -i masbate.laz -tile_size 1000 -append
>> dir masbate*
    10,993,886 masbate.asc
     2,132,006 masbate.laz

The ‘masbate.laz’ file is a little bit bigger than before because the tiny ‘masbate.lax’ file (that can also be stored separately) was appended to the end of ‘masbate.laz’ via the ‘-append’ option. Spatial indexing is realized with an underlying quadtree that can be visualized with lasview by pressing ‘Q’ (or with the pop-up menu) and the points corresponding to each quadtree cell can be turned blue by hovering above a cell and pressing ‘q’.

The points of the 50m Aster DTM as a LAZ file with the LAX spatial indexing quadtree  generated by lasindex.

Points of the 50m Aster DTM as a LAZ file and its spatial indexing quadtree generated by lasindex.

Before we use the sorted points from the Aster DTM to classify, flag, or delete the LiDAR returns far above the ground with lasheight we do a visual sanity check with lasview using the GUI settings shown below.

>> lasview -i raw_strips\*.laz ^
           -i masbate.laz ^
           -gui
GUI settings in lasview for picking a small area

GUI settings in lasview for picking a small area

By triangulating only the Aster DTM points we can visually confirm that the LiDAR points on the terrain are also close to the Aster DTM whereas the cloud returns are far up with a clear seperation between. The pink “drop lines” of length 50 meters that are dangling off each point give us a sense of scale (enabled via the pop-up menu that appears with a right click). Note how many last returns get stuck in the clouds. Those would likely cause troubles in subsequent ground classification with lasground.

Finally we run lasheight on 4 cores to classify all points that are 150 meter or more above the Aster DTM as noise (class 7).

>> lasheight -i raw_strips\*.laz ^
             -ground_points masbate.laz ^
             -classify_above 150 7 ^
             -odix _cloud -olaz ^
             -cores 4

Below the result with the cloud returns in pink (i.e. the points classified as 7). This process scales well even for larger ground points files, because lasheight uses an area-of-index (AOI) query to load only those ground points from the spatially indexed ‘masbate.laz’ file, that fall into the (slightly enlarged) bounding box of the raw LiDAR strip that is being processed.

lasheight_masbate_lidar_aster_classified_clouds

Tutorial: editing LAS or LAZ files “by hand” with lasview

This tutorial describes how to manually edit LiDAR using the new inspection and editing functionality available in ‘lasview.exe’ with the latest release of LAStools (version 140301). We will work with the familiar ‘fusa.laz’ sample LiDAR data set from the LAStools distribution that was recently reported to have shown strange symptoms assumed to be side-effects of the LAZ cloning experiments in the ESRI labs … (-;

Inspecting LiDAR files with cross sections

Copy ‘fusa.laz’ from the folder ‘lastools\data’ to the folder ‘lastools\bin’. Run ‘lasview.exe’ so that it loads ‘fusa.laz’. Either do this via the GUI by double-clicking ‘lasview.exe’, loading ‘fusa.laz’ via the ‘browse …’ menu, and then clicking the ‘VIEW’ button or by entering the command below:

C:\lastools\bin>lasview -i fusa.laz

Press the <x> key to toggle to “select cross” where you can pick a rectangular “cross” section. The default cross section is a profile extending across the bottom of the bounding box.

tutorial4_lasview_01_select_cross

By pressing the <x> key again you toggle back to actually view the cross section. Holding down <ALT> you can rotate the view to look at the cross section from the side. Holding down <CTRL> you can zoom in and out. Holding down <SHIFT> you can translate up and down or left and right. Increase or decrease the size of the points pressing <=> or <->. Hover with the mouse over a point and press <i> to inspect its coordinates and attributes.

tutorial4_lasview_02_cross_inspect_point

Traverse the LiDAR file visually by moving the cross section with the arrow keys <UP> <DOWN> <LEFT> and <RIGHT>. You can move either in the “select cross” view and see the picked rectangle move or in the “cross” view and “walk” through the LiDAR. Hold down the <SHIFT> key simultaneously to take bigger steps or the <ALT> key to take smaller steps. Inspect other points by hovering over them with the cursor and pressing <i>. The point information disappears when pressing <i> with the cursor over the background.

tutorial4_lasview_03_traverse_with_arrow_keysToggle back to the “select cross” view with <x> and pick approximately the same rectangle as shown below:

tutorial4_lasview_04_pick_missclassified_roof

Changing Classifications and Deleting Points

Continuing the steps above, toggle back to the “cross” view by pressing <x>. Note that part of the roof of the house has been miss-classified as vegetation while others are left unclassified. Press <e> to turn on the “EDIT” mode and right-click to select “reclassify points as building (6)” via the pop-up menu.

tutorial4_lasview_05_reclassify_menu

Now use the cross-hair cursor to draw a polygonal fence around all points that should be reclassified. Press <ESC> to remove the last vertex of the polygon if you miss-placed it by a mistake.

tutorial4_lasview_06_reclassify_fenceOnce you are happy with your polygon press <r> to register the edit. A note appears informing you how many points had their classification changed. In the top right corner an “undo” counter appears informing you how many changes you can undo by pressing <CTRL-u>. Try it. Immediataly the changes disappear and a “redo” counter appears instead. Press <CTRL-o> to redo the change you have just undone.

tutorial4_lasview_07_edit_result

Press <CTRL-s> to save this edit as a tiny LAY file using the recently introduced LASlayers concept. In case there was already an existing LAY file (that was not applied with ‘-ilay’ when starting lasview) you will be warned and have to press <CTRL-f> to force overwriting it as shown below.

tutorial4_lasview_07a_first_save

Press the key sequence <SHIFT-b>, <t>, and <a> to get the same visuals above.

tutorial4_lasview_07a_second_save

Press <SHIFT-t> to remove the triangulation again. After saving an edit it can no longer be undone via <CTRL-u>. Instead you will have to strip off this particular layer with the layer management available through “laslayers.exe” as described here. Now press <x> to toggle to the “cross select” view.

tutorial4_lasview_08_move_select_cross

Use the <DOWN> arrow to move the selected cross section to the area shown above that has a few unclassified points in the middle of the roof. Press <x> to go back to the “cross” view and try to understand why these points are not part of the roof. Looks like they are from the top of a chimney, and antenna, or a satellite dish as they do not fit the otherwise planar roof.

Assume we need to remove them for some reason. Pan, translate, and zoom the view such that these points can be easily surrounded by a polygon. Now press <d> to enter the “DELETE” mode, fence in these points, and press <r> to register the deletion.

tutorial4_lasview_09_delete_points

It can be tricky to place a clearly seperating polygon and you may be worried about deleting a few orange building points as well. Press <u> to only display the unclassified points before pressing <r> to register the deletion.

tutorial4_lasview_10_delete_points_unclassified_only

Press <a> to see all points again, then delete the other two points by finding a good view point, pressing <d>, and drawing a polygon.

tutorial4_lasview_11_delete_other_points

After registering this deletion of two points your “undo” counter should be at two. Press <CTRL-u> twice to undo this and the last deletion, then press <CTRL-o> twice to redo them both.

tutorial4_lasview_12_delete_other_points_undo_count_2

Now press <CTRL-s> to save this deletion as another layer. It will be appended to the LAY file that already contains one layer with the roof re-classification edit we did first. Press <t> to triangulate the points in the “cross” view. See how nicly flat the triangulated roofs are now that we deleted these 6 chimney points.

tutorial4_lasview_13_second_save

Look at the size of the tiny LAY file called ‘fusa.lay’ that is in the same folder as the ‘fusa.laz’ file. It contains all the edits we have done so far and mine is only 681 bytes in size. The original LAZ file has not changed. Maybe this is all you want for now. You could send only this tiny LAY file to a colleague elsewhere and he or she could apply those changes locally when needed using the ‘-ilay’ switches. For more on this see the LASlayers page.

C:\lastools\bin>lasview -i fusa.laz
C:\lastools\bin>lasview -i fusa.laz -ilay 1
C:\lastools\bin>lasview -i fusa.laz -ilay 2

However, you may want to eventually apply the changes and produce a new LAZ file. This will be a lot slower as it requires rewriting the entire file. It will also make changes permanent. Press <CTRL-a> and a new file is produced called ‘fusa_1.laz’ that has 6 points less than ‘fusa.laz’ and 69 points with a different classification as “building”. One more thing, press <CTRL-x> if you want to toggle between the “cross” section view and the default view.

tutorial4_lasview_14_applied_laslayers You need to have a license to LAStools to save edits for file that contain 1 million points or more.

online Web viewer for LiDAR

Nifty! Am online Web viewer for LiDAR data in LAS or ASCII format that can load files from your local drive. Very handy in case you need to inspect or show off a LAS file but do not have any LiDAR viewer installed on your computer – especially in case it should run cross-platform in any browser (does it?). But no LAZ support yet … (-;

Below you see a screenshot of me testing the viewer after loading rapidlasso‘s decompressed “fusa.las” data set that is part of the LAStools distribution. Try it for yourself at this site: http://lidarview.com/ … who is behind this?

lidarview displaying the "decompressed "fusa.las" model that is part of the LAStools distribution

lidarview displaying the “decompressed “fusa.las” model that is part of the LAStools distribution