From Unstructured File to Structured Scan?

950 views
Skip to first unread message

Jon H

unread,
Apr 28, 2019, 1:29:45 PM4/28/19
to Livox LiDARs
Many meshing programs and photogrammetry applications require structured scan data in the e57 or PTX formats. It seems that Livox's data is unstructured. Is there any way convert and/or 'spoof' the point cloud to ensure it properly imports into these programs? My intent is to merge photogrammetry data with scan data from the livox. Would it be possible for the Livox to create data in a "structured" format like most terrestrial/ground-based lidar units? 

Thanks for your help. Having a lot of fun with the Mid 40, and can't wait to get it outdoors. 

Livox Dev Team

unread,
Apr 30, 2019, 2:23:09 AM4/30/19
to Livox LiDARs
Hi Jon:
There many ways to save the point data as different formats.
1. program with Livox SDK ,you can save the points data as whatever format you want
2. save data bu Livox Viewer into .lvx format, then convert .lvx file to any other, the format of .lvx file is open to all users: https://www.livoxtech.com/3296f540ecf5458a8829e01cf429798e/downloads/20190417/LVX%20Specifications%20EN_20190417.pdf
3. convter .lvx file to .las format by Livox Viewer.Then any other format is possible by other software.

20190430_142157.png


Jon H

unread,
Apr 30, 2019, 10:16:24 AM4/30/19
to Livox LiDARs
Thanks, yes, I've done that process and converted to other formats, but I think the data is still considered "unstructured," in that most terrestrial scanners save in a "gridded" format with the laser position and heading and other key information included that enable the 3D meshing programs to align different scans (I think Autodesk calls it "registering scans"). 

I'm actually trying to import the .las file (and a few other formats) into meshing programs (including RealityCapture, Pix4D, Photoscan) with no luck. I'm assuming if a GNSS receiver and IMU syncs the data with the Mid 40, there's nothing preventing it from saving in a structured format, but I'm not aware of any converters out there that convert an unstructured .las to a structured PTX or e57 format because the .las is missing key information about lidar unit position and yaw/heading. 

bbbe...@gmail.com

unread,
Jun 8, 2020, 3:21:12 AM6/8/20
to Livox LiDARs
Hi Jon, I'm struggling with exactly the same issue... getting mid-40 exports into Reality Capture. Converting .las files into structured e57. Did you manage to solve the issue?

dbrae...@gmail.com

unread,
Jun 15, 2020, 3:28:37 PM6/15/20
to Livox LiDARs
Hi BBE, we are struggling with the same thing - getting mid-40 exports into Reality Capture. We think that we are close, but not quite there yet .... 

Would it be possible to connect and see where you are at and compare notes?

bbbe...@gmail.com

unread,
Jun 16, 2020, 2:06:20 AM6/16/20
to Livox LiDARs
Hi, sure. We are currently trying to get the exported files from loam_livox to cloud compare, merge it, and from there to Reality Capture.. I was hoping that after running loam-livox on the rosbag files will solve the issue with the nonstructured point cloud, but unfortunately, it's still the same. :(

maisa...@gmail.com

unread,
Jul 14, 2020, 10:47:14 PM7/14/20
to Livox LiDARs
Hi Livox Dev Team, 

Does the latest SDK release now can save the log file in lvx format instead of txt file? Can we extract the packet number into x,y, x value for cartesian instead of the packet number. 

maisa...@gmail.com

unread,
Jul 14, 2020, 10:50:43 PM7/14/20
to Livox LiDARs

Capture.PNG

Can the file be saved as csv or other file format that is readable where we can see the x, y,z value (cartesian)

Aldev

unread,
Jul 15, 2020, 5:59:27 PM7/15/20
to Livox LiDARs
Hi maisarahans, is it possible to turn xyz data into structured scan format?

Ch Kiranganesh

unread,
Jul 27, 2022, 3:35:34 AM7/27/22
to Livox LiDARs
did you get any solution for that i'm using Blk 360

Wingman Media

unread,
Feb 11, 2023, 9:34:17 PM2/11/23
to Livox LiDARs
I was trying to play with Mid-40 data trying to figure out how to merge a set of it into an orbital scan for rotating it on a nodal head. There are a few apps for windows on GitHub but there are all for Linux and I do not use it and want it to work with Windows.
My last attempt was purchasing a Bluetooth IMU hopping to get an yawn angle when my Mid-40 rotates but I stuck on not finding anything that can apply the yawn either during scanning or as postprocessing. I guessed I could use timestamp to match data from the IMU and point cloud data. However I could not find any reasonable priced soft to do it. I dropped the idea about a year ago even thought I could install Livox SDK on Windows and looked at the samples code. It is not hard to learn but it will take a lot of time for anyone who has no experience working with Lidars.

For the last 2-3 days I have been trying reality capture for photogrammetry and noticed they can align laser scans but like other mentioned they cannot do it without IMU data presented for every point. That's how I found this discussion. So I hope somebody's here will find a solution.

What I know already is that Livox viewer through a lidar settings lets a user to setup a yawn angle and it is actually placing point cloud data in the viewer correctly when yawn angle is applied.  What is missing though is automatic input from an eternal IMU through a COM port that you can use to get an yawn angle from an external IMU and correct data on the fly. If somebody can figure it out through Livox view app or through any 3rd party or their own written app for windows and get point cloud data from the viewer accounting for yawn it should not be hard to add an angle of tilting later meaning that you can make full 360x180 degree scan. Of course it can be done manually applying yawn and tilt in Livox viewer but you need to stop a Mid-40 after each rotation and enter yawn angle manually.

I do understand that cheap IMUs(mine is $40) are not near accurate and when its data says yawn is 7 degree it can be 7.1 degree or even 6 or 8 degree but when there is a solution to apply it on the fly or as post processing one can invest in more expensive and more reliable IMU. 
Reply all
Reply to author
Forward
0 new messages