Strip Alignment of Livox MID-40 Drone LiDAR with BayesMap software

[Martin Isenburg]

Hello,

On September 11th in 2020 we flew LiDAR with a drone above the destroyed mangroves to get a detailed capture of the topography this former wetland has 25 years after its destruction and of the few areas that were left intact but are equally under threat today.

Our efforts got stalled due to misalignment between the different flight lines that needed to be solved before the post-processing of the data could begin. In this blog post I describe how the data was aligned into a coherent point cloud with the help of the "stripalign" software by André Jalobeanu of "BayesMap" ... ☢💚🐥

https://rapidlasso.com/2021/03/04/strip-aligning-of-drone-lidar-flown-with-livox-mid-40-over-destroyed-mangrove/

Regards,

Martin

[Jochen Bind]

Thanks for writing this article, Martin. This is a pretty impressive result. And I’m curious to find out what you think about the sensor/platform after you’ve done the complete processing of the data (i.e. how does it compare in noise to something like the Velodyne scanners that you’ve processed in the past).

While I’ve toyed with the idea of trying out Andre’s stripalign software for a long time, this post has finally given me an extra incentive. As you know, we’re flying a LiDARUSA Scanlook with Velodyne 32 scanner. I often have to ‘tweak’ the boresight to optimize fit between flight lines. I think this is something that stripalign would do, am I correct?

But since we almost never have man-made features in the environments we are scanning, how would you expect it to perform. I notice that the result of lasoverlap is pretty well cleaned up in the lower left portion of the area, but remains visibly present (even though much improved) for what I imagine are the ‘natural envrionments’ of the mangrove land. What I can make out is an apparent ‘tilting’ of the sensor (i.e. roll) along the flightlines in the areas on the right of the image. This is all too familiar to me, since I see it often. I wonder what this stems from (since the manmade features cleaned up very nicely).

 

Anyhow, thanks for the insightful post.

 

Cheers,

Jo

--
Download LAStools at
http://lastools.org
http://rapidlasso.com
Be social with LAStools at
http://facebook.com/LAStools
http://twitter.com/LAStools
http://linkedin.com/groups/LAStools-4408378
Manage your settings at
http://groups.google.com/group/lastools/subscribe
---
You received this message because you are subscribed to the Google Groups "LAStools - efficient tools for LiDAR processing" group.
To unsubscribe from this group and stop receiving emails from it, send an email to lastools+u...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/lastools/CABSWR-E-1mctWegnPcjm%2Bm3cRr6EX3dYxH-qgF%2Bd1rwDpZb8wA%40mail.gmail.com.

Jochen Bind Hydrodynamics and GIS Technician +64-3-343-8021 National Institute of Water & Atmospheric Research Ltd (NIWA) 10 Kyle Street Riccarton Christchurch New Zealand Connect with NIWA: niwa.co.nz Facebook LinkedIn Twitter Instagram

To ensure compliance with legal requirements and to maintain cyber security standards, NIWA's IT systems are subject to ongoing monitoring, activity logging and auditing. This monitoring and auditing service may be provided by third parties. Such third parties can access information transmitted to, processed by and stored on NIWA's IT systems

[Jochen Bind]

Hi Martin,

I should have had a look at the data before writing my reply, it turns out. Thanks for supplying some real world data from the scanner. It’s highly valued.

 

What seems like a rolled sensor in the densely vegetated areas in your lasoverlap image doesn’t appear like that anymore once you look at the data itself. As a matter of fact, the mean vertical offset in the overlap seems to be a fraction of a cm. Stdev is around 7-10 cm for the areas I had a quick look at, which is to be expected for this sort of high and dense vegetation, I guess. But most importantly, there’s no spatial pattern in the differences, so they behave much like the flat paddock(?) in the lower left part of your image. That’s impressive and now I most definitely have to look at BayesMap Stripalign 😉

 

Just out of curiosity: looking at the fuzziness of the pointcloud over some of the roofs, I see a thickness of about 9-10 cm over what I expect to be solid flat roofs. The roofs I looked at don’t have an obvious corrugation that could account for this (unexpectedly high) variability. Is this the same level of fuzziness that you come up with?

 

Cheers,

Jo

 

From: last...@googlegroups.com <last...@googlegroups.com> On Behalf Of Martin Isenburg
Sent: Tuesday, 9 March 2021 12:53 AM
To: LAStools - efficient command line tools for LIDAR processing <last...@googlegroups.com>
Subject: [LAStools] Strip Alignment of Livox MID-40 Drone LiDAR with BayesMap software

 

Hello,

--

Download LAStools at
http://lastools.org
http://rapidlasso.com
Be social with LAStools at
http://facebook.com/LAStools
http://twitter.com/LAStools
http://linkedin.com/groups/LAStools-4408378
Manage your settings at
http://groups.google.com/group/lastools/subscribe
---
You received this message because you are subscribed to the Google Groups "LAStools - efficient tools for LiDAR processing" group.
To unsubscribe from this group and stop receiving emails from it, send an email to lastools+u...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/lastools/CABSWR-E-1mctWegnPcjm%2Bm3cRr6EX3dYxH-qgF%2Bd1rwDpZb8wA%40mail.gmail.com.

[Martin Isenburg]

Hello Jochen,

in a previous drone flight in March 2019 I used the same system as you (Snoopy HD by LidarUSA) and also successfully had Andre align the flightlines for me. There is definitely fluffiness in the Livox MID40 sensor just like there is in the Velodyne HDL32. How to get a clean ground surface out of this might be a future blog article ... assuming I get one. But after coloring the data with Google Earth Satellite imagery and creating a portal I already have a product with which I can show to government officials and environmental activists where the damage was done and do simple measurements. 

Regards,

Martin

To view this discussion on the web visit https://groups.google.com/d/msgid/lastools/MEXPR01MB093473B51A144E7BC78B476AD9939%40MEXPR01MB0934.ausprd01.prod.outlook.com.

[Mark Levitski]

I have worked with data from the Velodyne 32, the Riegl VUX mini, and the VUX-1 UAV sensors. They all have "fluff" that can be visually assessed by a thin cross section slice on a hard surface. The best so far has been the VUX 1. We used a tool in our LP360 software to measure it:

The precision report is how noisy the data is, deviation from a plane. Point density is listed too. The range is the peak to peak thickness of the data, the noise. The lower we fly, the less noisy. When we drive it and scan at 10 feet above ground, it is very clean. This is at 250 feet altitude. Units here are US survey feet.

 

As far as classifying ground with fluff, the routine will choose the lowest points and the rest will become some sort of unclassified pseudo vegetation. Then a last step of adjusting the Z to any survey control is performed if indeed that is required for the usefulness of the dataset. Since you will be using the lowest strata of the fluff, the adjustment is usually about the same amount as its thickness. 

So of course you will succeed, Martin.

To view this discussion on the web visit https://groups.google.com/d/msgid/lastools/CABSWR-H3NOw_wqRd%2Bx0XawbD88d-ow5rBO6AiDFzqdaNeHsThg%40mail.gmail.com.

[Paul Manley]

This is cool! Nice detailed workflow. 

Is it possible to assign the colors used in lasoverlap? I see in the lasoverlap readme that if you say "-bands 0.1 0.2 0.3" etc, but am trying to make a difference map that uses green, yellow, orange, red, magenta for 0.05, 0.10, 0.16, 0.24, 0.5, respectively.

Regards,

Paul Manley

On Wed, Mar 10, 2021 at 5:33 AM Martin Isenburg <martin....@gmail.com> wrote:

To view this discussion on the web visit https://groups.google.com/d/msgid/lastools/CABSWR-H3NOw_wqRd%2Bx0XawbD88d-ow5rBO6AiDFzqdaNeHsThg%40mail.gmail.com.

[Martin Isenburg]

Hello,

yes you can. Use

-color_bands band1 col1 [band2 col2 [band3 col3 [ ...]]]

which will mean something like

-color_bands 0.05 0xFF0000 0.10  0x00FF00 0.16 0x0000FF 0.24 0xFF00FF 0.5 0xFFFF00

Regards,

Martin

Martin

To view this discussion on the web visit https://groups.google.com/d/msgid/lastools/CAAYwJ14oLer8YAVQ9MANewTkFhF1rtJ3T8PHx7%3DtRWoVFrw-Jg%40mail.gmail.com.