|
|
Post by martijn on Apr 19, 2020 15:36:27 GMT -5
Hi,
I'm anxious to start designing my local golf courses like many of you. I've got the OSM part working, but I'm struggling with getting the lidar imported.
I've found a website where I can download many laz files. There's small (600MB) and large ones (2.2GB). I'm not sure what I should add in this op, so let me know and i'll add it.
All I'm seeing in the tool is:
Processing: lidar.las
Overwriting projection with EPSG:28992
Unit in use is 0.9999999999999996
Proj4 : pyproj.Proj('+units=m +init=epsg:28992 ', preserve_units=True)
Could not load lidar.las Please report this issue.
No valid lidar files found, no action taken
Directory was: C:/2
I used laszip to extract the las from the laz, much quicker to process on my pc so it seems. I also tried the laz, though. Same error, same result only took 6-8 times longer.
The las alone doesn't seem to have the epsg, so I forced this one. Not 100% certain this is the right one, but I'd expect a different kind of error message ?
When is a lidar file 'not valid', the requirements are not clear to me, making it a bit difficult to troubleshoot what is going wrong where?
Please advise
|
|
|
|
Post by sandsaver01 on Apr 20, 2020 7:25:27 GMT -5
Hi, I'm anxious to start designing my local golf courses like many of you. I've got the OSM part working, but I'm struggling with getting the lidar imported. I've found a website where I can download many laz files. There's small (600MB) and large ones (2.2GB). I'm not sure what I should add in this op, so let me know and i'll add it. All I'm seeing in the tool is: Processing: lidar.las Overwriting projection with EPSG:28992 Unit in use is 0.9999999999999996 Proj4 : pyproj.Proj('+units=m +init=epsg:28992 ', preserve_units=True) Could not load lidar.las Please report this issue. No valid lidar files found, no action taken Directory was: C:/2 I used laszip to extract the las from the laz, much quicker to process on my pc so it seems. I also tried the laz, though. Same error, same result only took 6-8 times longer. The las alone doesn't seem to have the epsg, so I forced this one. Not 100% certain this is the right one, but I'd expect a different kind of error message ? When is a lidar file 'not valid', the requirements are not clear to me, making it a bit difficult to troubleshoot what is going wrong where? Please advise Martijn I have worked on a bunch of .las and .laz files and they have never been that large. Almost all of them are less than 100 mb each. The one time I had problems was the course I am currently working on, Milwaukee CC, which needed three separate files of about 300 mb each to cover the area. Chad's Tool would process ONE of these, but if you tried to add more than one it would do something like what you show above. I was only able to get it to work by using LasTool (steep learning curve, beware) to cut the areas I needed out of the larger files and they were then a manageable size. I do not understand why your files are so huge. Are they covering a giant footprint of area? |
|
|
|
Post by martijn on Apr 23, 2020 15:21:12 GMT -5
The website I can download these pointcloud laz files from is: downloads.pdok.nl/ahn3-downloadpage/I am not familiar or experienced enough with this to tell if 'this area' is large, compared to areas you are talking about. Take a look at the URL. I don't see an option there to select/cut/clip/split/crop laz files there into small files. It does read 'high resolution' here and there. |
|
|
|
Post by sandsaver01 on Apr 23, 2020 15:39:31 GMT -5
The website I can download these pointcloud laz files from is: downloads.pdok.nl/ahn3-downloadpage/I am not familiar or experienced enough with this to tell if 'this area' is large, compared to areas you are talking about. Take a look at the URL. I don't see an option there to select/cut/clip/split/crop laz files there into small files. I just did go to that website and picked a square at random, C_24HN2.LAZ . The file is 3.7 Gb, which is way to huge to open in chad's tool. The only thing I can find to take out part of a .LAZ file or to break it into smaller tiles is a open-source program called LasTool (google it). I used it to get the Lidar for Milwaukee Country Club because the .LAZ files for that were too big for Chad's tool and they were only 300 Mb! The learning curve to use LasTool is pretty steep, at least it was for me, so you can investigate, but otherwise you may be out of luck. |
|
|
|
Post by martijn on Apr 24, 2020 16:51:30 GMT -5
Thanks for trying that. I found on the same website that indeed the ahn3 files are complete what is called: tiles and they really high res. So, I suppose the system in America allows to select 'open data' and download small pieces of those tiles, which results in small las files. Simultaneously the resolution of those may be lower than the PDOK source (I have not compared those, but I read other threads talking about millions of points). Still being really novice at this, I wonder about chads not not saying anything about minimal pointcloud resolution. Therefor I searched a bit more and there appears to be AHN2 files as well. They are older and smaller resolution and thus smaller in file size (yet still to big, because of the same reason: they are complete tiles). As for the challenge at this point: If I could get my hands on a good tutorial with cropping the las file. I might get a decent file size with AHN2 files and cropping. Yet since those are older I first need to figure out if they're not too old and not contain the golf course  |
|
|
|
Post by martijn on Apr 26, 2020 7:09:08 GMT -5
The only thing I can find to take out part of a .LAZ file or to break it into smaller tiles is a open-source program called LasTool (google it). I found it and it worked! The tool I used was: las2las (also works with laz files. The 2.2GB laz file cropped to 220MB, which loaded fine in Chads tool. (I had to provide the EPSG code manually. Perhaps this got lost in the cropping.) While there may be an even better way to do it, here's what I did: Open the las2las tool (has a GUI). Browse to and open the laz file. Next, if your laz file is really big >300MB, the file contains way too many points, in surface and in density. With las2las you can reduce both. On the right there is a value called sample points. While the default value may work fine, you might want to tweak this number. My thought was: the higher, the better resolution in the result file. This depends heavily on the available RAM in your computer. With 16GB I was easily doing 12000000 instead of the default 5000000. Click on view and wait, depending on your pc this may take while, the image will be rendered in a seperate window. After the image rendered, comes the hardest part. You cannot interact with this render. You'll have to look at the location where the golf course actually is and remember the area to select when you close this window. (I couldn't find a better way to do this, yet). Back in the main screen, on the bottom click on 'Pick' and draw a square approximately on the area that contains the course you want to extract. Click view again to see the result. Repeat until satisfied with the selection (this is mostly a hit and miss approach, I hope there's a better way I'm not yet aware of). Once the selection is ok, click on Output and provide a location and filename. (Selecting more jobs/core did not improve speed for me). Lastly click RUN64 and wait for the file. (It happend a few times the file was ready, but the application appeared to be still processing). |
|
|
|
Post by sandsaver01 on Apr 26, 2020 7:37:47 GMT -5
That is exactly how I do it Martijn! I have also used the "Lasprecision" or "Lasprecision64" to determine if the Lidar files are good enough to use to make a course - it has saved me some time in the past.
|
|
|
|
Post by martijn on Apr 26, 2020 8:41:15 GMT -5
That is exactly how I do it Martijn! I have also used the "Lasprecision" or "Lasprecision64" to determine if the Lidar files are good enough to use to make a course - it has saved me some time in the past. Can you give some examples of issues with not 'good enough' ? I've not read any prerequisites about lidar files, other than being too old and not containing a picture of the (finished or perhaps radically changed) golf course. |
|
|
|
Post by sandsaver01 on Apr 26, 2020 9:52:05 GMT -5
That is exactly how I do it Martijn! I have also used the "Lasprecision" or "Lasprecision64" to determine if the Lidar files are good enough to use to make a course - it has saved me some time in the past. Can you give some examples of issues with not 'good enough' ? I've not read any prerequisites about lidar files, other than being too old and not containing a picture of the (finished or perhaps radically changed) golf course. Martijn, this whole thing is way too complex for my true understanding. I am just feeling my way around at best. What I can show you below is what the Lasprecision output looks like on a good plot and a bad plot. "good data"
lasprecision -i "c:\Users\Mike Ricks\Desktop\Manufacturers\manuf las\USGS_LPC_DE_DelawareValley_HD_2015_18TVK483440_LAS_2017.las"
original scale factors: 0.01 0.01 0.01
loading first 5000000 of 11420141 points
X differences
0 : 4849999 0
1 : 150000 0.01
Y differences
0 : 4937524 0
1 : 62360 0.01
2 : 115 0.02
Z differences
0 : 4994572 0
1 : 5226 0.01
2 : 128 0.02
3 : 18 0.03
4 : 9 0.04
5 : 9 0.05
6 : 3 0.06
7 : 7 0.07
8 : 4 0.08
9 : 3 0.09
10 : 7 0.1
11 : 2 0.11
12 : 1 0.12
13 : 1 0.13
15 : 2 0.15
18 : 1 0.18
19 : 1 0.19
24 : 1 0.24
27 : 1 0.27
37 : 1 0.37
"Bad data"
lasprecision -i "C:\Users\Mike Ricks\Desktop\vicnat\vicnat1\USGS_LPC_IN_WT_B12_Warrick_2013_in2013_28751000_12_LAS_2016.las"
original scale factors: 0.01 0.01 0.01
loading first 5000000 of 2864020 points
X differences
0 : 2365621 0
1 : 496802 0.01
2 : 1592 0.02
3 : 4 0.03
Y differences
0 : 2387279 0
1 : 456167 0.01
2 : 18258 0.02
3 : 2003 0.03
4 : 266 0.04
5 : 37 0.05
6 : 7 0.06
7 : 1 0.07
8 : 1 0.08
Z differences
0 : 2852451 0
1 : 11292 0.01
2 : 109 0.02
3 : 52 0.03
4 : 31 0.04
5 : 12 0.05
6 : 8 0.06
7 : 9 0.07
8 : 6 0.08
9 : 4 0.09
10 : 5 0.1
11 : 6 0.11
12 : 2 0.12
13 : 1 0.13
14 : 3 0.14
15 : 1 0.15
16 : 3 0.16
17 : 1 0.17
18 : 1 0.18
19 : 2 0.19
Notice there is a lot more points in red in the bad example (Victoria National GC, which was so lumpy as to be unusable) than the good example (Milwaukee CC) I am sure there is a lot more that could be gleaned from this data, if I understood the "readme" well in this tool, but this has prevented me from doing some courses like Riviera CC where the data is poor. Hope this helps. |
|
|
|
Post by martijn on Apr 26, 2020 15:14:03 GMT -5
I suppose you used that tool because the lidar data was off and the result on the course looked awkward ?
|
|
|
|
Post by sandsaver01 on Apr 26, 2020 15:25:49 GMT -5
I suppose you used that tool because the lidar data was off and the result on the course looked awkward ? I wish it had happened that way Martijn lol. I put in a lot of work on Victoria National before I realized how lumpy it was. I used the tool in hindsight and saw why that was. It did help me spot that the Lidar data for Riviera CC is not good before I did anything with it though. |
|
esteb22
Caddy
New to this and looking for help.
Posts: 2
|
Post by esteb22 on Jun 6, 2020 13:56:20 GMT -5
Can either of you help with the problem I am having... In the process Lidar step I get the Not valid lidar data, no action taken message.
What do I need to do? I have followed the tutorial steps to the tee.
|
|
|
|
Post by martijn on Sept 2, 2020 8:37:16 GMT -5
Not exactly sure what all the possible things are that can be wrong here. Been a while since I last created a lidar for an rcr. It might be something with the EPSG missing in the file and you need to provide one manually. Don't ask me where you can find your correct EPSG you need, sorry.
|
|
