Discussion: LAS Tools issue on a new machine

Hey all!

I have a new (< 1 year old) Windows 10 machine, and I'm having a particular issue with LAS Tools. I've tried re-downloading and unzipping, restarting, and such, but I still have the same issue that I didn't have on a previous Windows 10 machine (~7 years old). I'm wondering if I've set something up wrong? ie: forgot to do a setting, or it's Windows Defender getting in the way? (when I first open up the .exe file, it Windows Defender says it's an "unrecognized app" and I tell it to "run anyway".

The issue: I open up the LASTools simple GUI, I browse to find my las/laz files, I double-click on a file name, and the file now shows up in the window in the upper left corner, like it should. But I would also usually see a footprint of the lidar tile(s) in the large window in the middle, (which I think was shown in blue?). I just have a blank gray area here instead. And when I click the 'pick' button in the clip input area below the main window, I get a crosshairs mouse cursor, but it won't draw a bounding box (which I think should be red?). It's as if nothing in that main window area is displaying properly.

Thoughts?

Thanks!

Do you have (I think you should have) an other window open behid the app, black background with some text. Is there any error messages?

Nope, no error messages there.

I was able to merge some tiles successfully today, despite their footprints not showing up on the main window. What I don't think I can do right now is clip tiles, since I can't really see what I'm doing.

Excuse me butting in but LAS Tools seems a pretty expensive bit of software.
What is it used for?

A portion of LAStools is Open Source and is mega useful for merging, clipping, converting projections of Lidar LAS and LAZ files, and far more - see License Agreement

LAS Tools can do pretty much anything you need to a LAS or LAZ file. You can convert from state plane to WGS meters. You can reclassifiy points, strip out all but ground points, visualize the data set, crop, or merge datasets. Very powerful tools for manipulating point clouds.
The software may be used for "all non-profit personal, non-military educational, or non-profit humanitarian purposes."

@Pink Socks I also have a less than one year old Windows 10 machine and have been experiencing the same problem as you. But it only started happening for me in the past few weeks.

I didn't think much of it initially because I don't usually need to see the blue box for what I use LAStools for. But it is interesting to see someone else experiencing the same issue.

I don't have any idea about how to fix it, but it makes me wonder if it's caused by a software update.

@Gswede - did you maybe reinstall recently? Looks like the Download was updated Aug. 13.

Just installed and used LAStools on a new Windows 10 machine myself yesterday (finally got tired enough of crashes and having to restore the OS on the previous computer and succumbed to the lure of solid state drives) and everything I did with it, including clipping a point cloud with a polygon, worked with no issues. Possibly the problem is solely with the simple GUI and clipping from the command line would work on your new computer as on mine?

My old laptop is at home, with an older version of LAS Tools, so when I get back to it, I'll move that version over to the new laptop to see if it works.

It's been almost two years since I've done a lot of LAS/KP stuff, and I was pretty clunky at it then, so I was assuming that I had done something wrong today.

Related to my earlier question (in another thread) about about having KP skip some steps... I was hoping to use LAS Tools to crop out all of the giganto cliffs to make the KP processing quicker.

I did a full KP run of 6 merged tiles today, and it was surprisingly speedy because I fed it tiles without any vegetation or cliffs. Then I got cocky and merged 15 of them, many with the giganto cliffs, and it's still going....

I suggest processing tiles in batch mode 4 (or even 8) threads parallel, instead of merging files. Many times faster that way, because you can make. And setting cliffthin to 0.5 should make that step about 4 x faster. So overall it should be about 3...8 times faster or so depending on your computer.

- make "in" subfolder and move all laz tikes there
- open ini file and look for batch=0 and change it to batch=0 batch=1
- just below that there is processes, set that to 4
- look for cliffthin, set that to 0.5 or so
- look for savetempfiles, set that to 1 if you like to get dxf files.
- look for basemapinterval=0 and set contour interval you want for mapping, like 0.625

When you start it it should open 4 black windows and each processing tiles, and results should one by one appear to "out" subfolder. Seamless result, because in batch mode it reads buffer from neighbouring tiles and crops them to original extent in the end. See, you can't process thousands of km2 by merging them all together, but in tiles it goes just fine.

Change it to batch=1, I think you meant

Thanks for the tips!

When I got up this morning, I had everything I needed for this project, it just took the computer a little longer. I'm used to making tiny little sprint maps, so this was really the first time that I'm mapping something *substantially* larger (~20x larger than my previously largest map), so my process is definitely not optimized yet.

Now that Pink Socks is sorted out may I jump back in here?
Thank you for the advice on what LAS Tools does.
Maybe I just didn't understand something simple here. So are you all talking about this because you are using OOM and/or older versions of OCAD?
The functions of LAS Tools described above are all done pretty well automatically by OCAD 12 and OCAD 20xx for .laz and.laz tiles There are some functions in the later OCAD such as contour smoothing and importing geo-referenced images directly into background that were not available in OCAD 12 but importing, processing and merging multiple .laz or .las tiles, choosing the contour interval you desire, getting point cloud and getting a bunch of Kartapaullatin-like features, etc. are there in both.
Or am I missing something in OCAD 2020 that I could be getting from LAS Tools?

With lastools you can tweak laz files and make various analysis. For example take gound points from older leaf-off laz and non-gound points from new leaf-on scanning, merge them together and make analysis images for vegetation updates - new leaf-on can have poor ground for being leaf on scanning. Or strip off all points over 6 m from gound and make slope image of the rest to make edges of lower vegetation under canopy more visible. Typical use case is re-classifying ground points, often bushes/boulders are mapped as gound or sharp knolls as vegetation and by iterating ground classifyng parameters you can sometimes improve yout input laz files quite a lot before impoting them to 0cad or whatever. And things like that. When you get an idea how to do or detect something from point could lastools is most likely the most flexible tool to boil it together.

One use case is just looking laz in 3D, lasview. I believe ocad uses it too and comes with lasview of lastools baked in.

@gordhun
As another example - I had a file with a number of large square sections in a checkered pattern missing the ground points, but they also had files containing only the ground points. I combined the files in LAStools and eliminated the duplicate points (in the areas that were not corrupted and thus had them in twice), and then had perfectly working files.
In WI, every county have their own projections, and there is no proper way in OCAD to convert individually by county.
And that is before getting even deeper into the added understanding of how to re-classify or all that, as Jagge describes.

Thanks Jagge and Andreais. That is some work way above my pay grade, not needed for the level of work I am doing.
Except I'm not sure I understand the point about projections. Do you mean UTM, State Planes, Lambert etc? I find I just have to tell OCAD the projection and the area and it handles the rest and I end up with a geo-referenced map.
Sometimes USGS is sloppy in identifying the correct projection in the metadata but I've figured out a way to figure out when they are wrong. For instance all of MN is UTM and that is great. On the other hand a lot of FL counties' laz tiles are identified as UTM by USGS but they are really State Plane. The way to tell the difference is that in the metadata the state plane ones identify their planar distance as feet. UTM is always meters. Does that make sense?

BTW I recently played with PDAL and was able to reproject to UTM and get some contours out for ground points pretty quickly, I have not tried anything heavy duty but it looked like maybe a good alternative to LAStools (especially if you are not on a Windows machine). Has anybody tried that for Lidar processing?

@gordhun I am at the border of a great easy state (MN) and the most frustrating one (WI), as it has, not like most other states, one (or a few) coordinate reference systems for their entire state, but has one reference system per county. And they are not consistent in choosing feet or survey_feet, and most frustrating if you have a place at the border of 2 counties... I luckily had lots of help every time I got stuck.

I know the frustration andreais. In my US part of the year most of the mapping is Florida. They have a few counties with UTM, more with State Plane and some that share no LiDAR through USGS so I have to go to the local Water Management District.
Canada, province by province is hit and miss, a lot of miss actually except Quebec which has just made the whole provincial LiDAR open and available but the complication is they use a variation of UTM called MTM.
I loved working with the MN LiDAR and if you ever find anyone in Duluth, Bemidjii, Lutsen, Woodbury HS/ Ojibway Park or Paul Bunyon SF who want an orienteering map I have one for them., that plus St John University, of course.

For example take gound points from older leaf-off laz and non-gound points from new leaf-on scanning, merge them together and make analysis images for vegetation updates - new leaf-on can have poor ground for being leaf on scanning. Or strip off all points over 6 m from gound and make slope image of the rest to make edges of lower vegetation under canopy more visible.

You could do both of those in OCAD. You could make a DSM from the leaf on lidar and a DTM from the leaf off. Use the DTM for contours and the difference between the DTM and the DSM for the vegetation. I guess you might get some anomalies in places where there is no vegetation if the DTMs of the two lidar sets don't match well.

And using the veg height function you could create a image showing only the first 6m of vegetation,

Can you do all the things people are talking about on this thread with the free version of lastools? I tried using lastools some time ago (through QGIS I think) and kept getting messages saying I couldn't do things.

I use free version of Lastools in command line mode to reclassify buildings. This adds lots of boulders to the ground points which then turn up in the slope and terrain images. The only issue I see with the free version is that it introduces small distortions which might be meaningful for serious survey work, but as far as I can tell, they make little difference for orienteering uses. If I misunderstand the distortions, I would like to be set correct. Everything else appears standard. The issue might have been limitations in the QGIS system way back. It seems much more well integrated today.

Thanks Neil. I am just watching some LASTools videos. In one the developer says that if you are worried about the distortions in the free tools you can create very small tiles (200x200m) and then they will work with no distortion. You could try that for the sake of comparison and, hopefully, peace of mind.

I believe you can that with 0cad but that's not quite the same I was trying to achieve.