The future of handheld scanners

THOR3D, Artec3D, ASUS Xtrion, Creaform G0-Scan, DotProduct, FARO Freestyle Noomeo, Mantis Vision, ZEB1, Kinect, DAVID, Effortless3D E3D Mk1, etc
Post Reply
MalteHC
V.I.P Member
V.I.P Member
Posts: 140
Joined: Mon May 05, 2014 11:19 am
Full Name: Malte Holm
Company Details: TLS
Company Position Title: Surveyor
Country: Norway
Linkedin Profile: Yes

The future of handheld scanners

Post by MalteHC » Wed Jan 30, 2019 7:46 pm

In the last few years, we have seen a lot of development in the field of
making tablets able to measure, because of alle the AR thats going on.

We have seen some examples of measuring apps for Ipad's like the one below:
https://www.youtube.com/watch?v=XA7FMoNAK9M


youtu.be/XA7FMoNAK9M

This make me think, if in a few years, this technology will make our geoslammer go into retirement.
I could very easy imagine this kind of photogrammetry, taking over for the handheld scanners as our geoslammer.
What do you guys think?

VXGrid
I have made 50-60 posts
I have made 50-60 posts
Posts: 52
Joined: Fri Feb 24, 2017 10:47 am
Full Name: Martin Graner
Company Details: PointCab GmbH
Company Position Title: Research and Development
Country: Germany
Linkedin Profile: No

Re: The future of handheld scanners

Post by VXGrid » Fri Feb 01, 2019 10:15 am

I disagree.
Photogrammtry has always the issue of POI (point of interest) detection to be able to calculate the position of the camera between two images. In addition to that is the missing scale in Photogrammetry, you need to capture this scale somewhere.

This results in two disadvantages: You get holes where the surface is monotone and you need a scale bar or anything else to apply the scale.

Canvas on the other side seems to not only use Photogrammetry, but in addition an infra-red laser.
Infra-red is not working in direct sun light nor in rooms with infra-red emitting lights. Very dark surfaces are absorbing the emitted light, so you can't capture them and the range is normally limited to max ~ 10 meters I guess.


These algorithms will get better, but in my opionion you will never be able to compare the quality of a point cloud, generated with a hand held scanner with a good scanner head, with the point cloud of an IPad infra-red fixture.

User avatar
smacl
V.I.P Member
V.I.P Member
Posts: 155
Joined: Tue Jan 25, 2011 5:12 pm
Full Name: Shane MacLaughlin
Company Details: Atlas Computers Ltd
Company Position Title: Managing Director
Country: Ireland
Linkedin Profile: Yes
Location: Ireland
Contact:

Re: The future of handheld scanners

Post by smacl » Fri Feb 01, 2019 10:30 am

I don't think it comes down to laser scanning versus photogrammetry, I think future solutions will combine both technologies along with IMU to get an optimal solution. Either work well where you have a number of distinct features in view, but both fall over without this. The two technologies use different types of features though, spatially distinct for scanners, visually distinct for photography, so combining the two provides more strength here. Similarly, laser scanning produces good data in all lighting condition but will be sparser than pixels in a photograph and may struggle with some surfaces. Photography is dependent on background lighting and has other error sources, so again the stronger theoretical solution is the combination. Why go for either/or when you can have both? :)

surveypm
I have made 40-50 posts
I have made 40-50 posts
Posts: 41
Joined: Mon May 18, 2015 5:31 pm
Full Name: Tom Elbich
Company Details: Solid Ventures Inc
Company Position Title: President
Country: USA
Linkedin Profile: Yes
Location: Dallas, PA
Contact:

Re: The future of handheld scanners

Post by surveypm » Fri Feb 01, 2019 12:33 pm

I agree with Shane. I use a Faro Freestyle in conjunction with an M70 for scanning in industrial projects. The Freestyle allows you to get into the tight spaces where you want close scanning in structural or piping connections for example. The capture seamless works with the M70 scans in SCENE software. Best of both worlds in my opinion.

MalteHC
V.I.P Member
V.I.P Member
Posts: 140
Joined: Mon May 05, 2014 11:19 am
Full Name: Malte Holm
Company Details: TLS
Company Position Title: Surveyor
Country: Norway
Linkedin Profile: Yes

Re: The future of handheld scanners

Post by MalteHC » Mon Feb 11, 2019 9:31 am

surveypm wrote:
Fri Feb 01, 2019 12:33 pm
I agree with Shane. I use a Faro Freestyle in conjunction with an M70 for scanning in industrial projects. The Freestyle allows you to get into the tight spaces where you want close scanning in structural or piping connections for example. The capture seamless works with the M70 scans in SCENE software. Best of both worlds in my opinion.
Yes, but the freestyle, is kind of a high accuracy, short distance scanner.
It wouldent provide the quick 'n dirty solution for a floorplan, as the geoslammer does.
But once again, does the geoslammer provide to bad a quality for something as BIM modelling.
So once again, it comes down to, that a lot of this kind of solutions is used where, you need cm accuracy and the quality isnt so important,
as the scantime.
I see this as a bit of the same discussion, if you would have told a surveyor 10 years ago, that today, he would be able, to buy a drone
for 2000 dollars and make high accuracy maps.

michaeldutch
I have made <0 posts
I have made <0 posts
Posts: 4
Joined: Mon Sep 05, 2016 12:07 pm
Full Name: Michael Dutch
Company Details: GeoSLAM
Company Position Title: Global Product Manager
Country: UK
Linkedin Profile: Yes

Re: The future of handheld scanners

Post by michaeldutch » Mon Feb 11, 2019 2:57 pm

I would say everyone is making some really good points here.

Martin has clearly pointed out the pit falls of using a completely photogrammetry based solution. The biggest, I would say, is when your in an environment which is 'monotone'. This usually causes a catastrophic fail in processing for photogrammetry and leaves a user with holes in their data or little data all together. A 'monotone' environment is also quite common when thinking about a buildings internals, with white walls and ceilings. So something like LiDAR which functions regardless of lighting is logical.

I see your point as well Malte. Different specifications of jobs usually control the type of instrument on site. When there are time and efficiency pressures a GeoSLAM solution will give you deliverables quicker and more reliably than any other system.

Finally, looking forward, Shane's point on using both techniques together and delivering a hybrid approach is very interesting. If you have not seen any of GeoSLAM's work with Bentley Context Capture please take a look at this video:

https://www.youtube.com/watch?v=LLIZ5m75lME

Obviously it take times and computing power to get to these results. But this is the first steps towards a fully hybrid solution.

If you have any questions or would like to see this solution in more detail please send me a message directly or contact you nearest GeoSLAM distributor. https://geoslam.com/geoslam-dealer-network/

AdrianHedges
I have made <0 posts
I have made <0 posts
Posts: 2
Joined: Wed May 15, 2019 3:46 pm
Full Name: Adrian Hedges
Company Details: T3DMC
Company Position Title: Product Manager
Country: UK
Linkedin Profile: Yes

Re: The future of handheld scanners

Post by AdrianHedges » Thu May 16, 2019 5:53 pm

This is an interesting subject with people looking at a hybrid approach to scanning devices, we have such a device namely the AXE link below, there are a lot of benefits to this fully integrated scanner and photogrammetry solution especially when dealing with scanning objects larger than 1m-1.5m.

These type of scanners are used mainly in engineering and automotive industries due to the very small tolerances required in these areas, but the list of applications are extensive.

http://t3dmc.dev.onpressidium.com/produ ... d-scanner/

The link below is to a video of the AXE system in use on a reverse engineering project on an old E-Type Jaguar

https://www.linkedin.com/feed/update/ur ... 8107781121

VXGrid
I have made 50-60 posts
I have made 50-60 posts
Posts: 52
Joined: Fri Feb 24, 2017 10:47 am
Full Name: Martin Graner
Company Details: PointCab GmbH
Company Position Title: Research and Development
Country: Germany
Linkedin Profile: No

Re: The future of handheld scanners

Post by VXGrid » Fri May 17, 2019 8:52 am

MalteHC wrote:
Mon Feb 11, 2019 9:31 am
Yes, but the freestyle, is kind of a high accuracy, short distance scanner.
It wouldent provide the quick 'n dirty solution for a floorplan, as the geoslammer does.
But once again, does the geoslammer provide to bad a quality for something as BIM modelling.
So once again, it comes down to, that a lot of this kind of solutions is used where, you need cm accuracy and the quality isnt so important,
as the scantime.
I see this as a bit of the same discussion, if you would have told a surveyor 10 years ago, that today, he would be able, to buy a drone
for 2000 dollars and make high accuracy maps.
Somehow this topic was boarded and the discussion topic lost :lol:

@Malte: Your initial post mentioning the IPad and replacing the GeoSLAM device:
I can't think of a way of achieving good results with that kind of gear at this point in time. Yes smart phone / tablet cameras are getting really strong and really great and a lot of devices get a depth sensor as well (or you can add one).

The cameras itself would be enough to use Photogrammetry to build a point cloud (with all mentioned challenges).

The depth sensors won't be good enough for years, since currently they are working with infra red.
There are different aproaches in using infra red: projecting a grid and capturing the aboration of that grid for the measurements, projecting points and measuring the size of the points to get the distance, perhaps time of flight, measuring how long it took to receive an echo of the sent pulse.
It's not very accurate (due to hardware limitations), it's not scalable (you can't cover big areas), the data is very sparse (no dense point cloud).

Speaking of accuracy: Since it is a hand held scanner you have the accuracy of the single point (one shot point cloud, local), and the global accuracy of your cloud (local accuracy + the movement). Even with visual odometry + IMU (finding identical features with the cameras + using the accleration sensors of the smart phone) the positioning is not very well, resulting in a big global error. Since the received single shot "point clouds" are very sparse, you can't use them for SLAM (which might improve the result).

Laser scanners are working with phase comparison and time of flight and on the hardware side have an "expensive" laser head.
As long as the laser head itself costs more than 1/5 of the complete phone price I don't think anybody will develop something for it.

I guess: When the smart phone producers put a laser head into their smart phones the laser scan community will utilise smart drohnes for capturing the interiors, which will be started at the front door and the drohne flys through the complete building by itself producing a complete BIM model, instead of a point cloud :geek:

Post Reply

Return to “Hand Held Scanners”