Need some help about the principle of hand held laser scanner

Apple iPhone & iPad, Artec3D, ASUS Xtrion, Creaform, DotProduct DPI-8S &DPI-10, FARO Freestyle, G0-Scan, Mantis Vision, Microsoft Kinect, Noomeo, THOR3D, ZEB1, viDoc RTK Rover etc.
Post Reply
Solaris
I have made <0 posts
I have made <0 posts
Posts: 4
Joined: Sun Jul 07, 2024 10:00 am
Full Name: Yu Huang
Company Details: ZY intelligence
Company Position Title: Algorithm Engineer
Country: China

Need some help about the principle of hand held laser scanner

Post by Solaris »

Hi everyone!
I'm trying to make a handheld laser scanner similar to HandyScan or ATOS. I understand that the 3D coordinates or depth are acquired through laser triangulation, but the role that stereo vision plays is still confusing to me. One camera is sufficient for triangulation, so what is the purpose of applying stereo vision? Is it used to build a coordinate system based on marks?
User avatar
smacl
Global Moderator
Global Moderator
Posts: 1467
Joined: Tue Jan 25, 2011 5:12 pm
13
Full Name: Shane MacLaughlin
Company Details: Atlas Computers Ltd
Company Position Title: Managing Director
Country: Ireland
Linkedin Profile: Yes
Location: Ireland
Has thanked: 645 times
Been thanked: 694 times
Contact:

Re: Need some help about the principle of hand held laser scanner

Post by smacl »

Solaris wrote: Tue Jul 09, 2024 11:16 am Hi everyone!
I'm trying to make a handheld laser scanner similar to HandyScan or ATOS. I understand that the 3D coordinates or depth are acquired through laser triangulation, but the role that stereo vision plays is still confusing to me. One camera is sufficient for triangulation, so what is the purpose of applying stereo vision? Is it used to build a coordinate system based on marks?
One camera (or one eye) is not enough to compute depth. You can triangulate a position from a single point of reference using resection only where you have three or more known points in view. Stereo vision uses two cameras (or eyes) to observe the same object point from different positions a known distance apart and triangulates that way to compute a distance to the object.

A hand held laser scanner is not like a camera, it actively computes depth. This is done by sending out a beam of light, timing how long it takes to receive a reflection of that beam of light, and dividing by the speed of light to get a distance.
Shane MacLaughlin
Atlas Computers Ltd
www.atlascomputers.ie

SCC Point Cloud module
Solaris
I have made <0 posts
I have made <0 posts
Posts: 4
Joined: Sun Jul 07, 2024 10:00 am
Full Name: Yu Huang
Company Details: ZY intelligence
Company Position Title: Algorithm Engineer
Country: China

Re: Need some help about the principle of hand held laser scanner

Post by Solaris »

smacl wrote: Tue Jul 09, 2024 11:24 am
Solaris wrote: Tue Jul 09, 2024 11:16 am Hi everyone!
I'm trying to make a handheld laser scanner similar to HandyScan or ATOS. I understand that the 3D coordinates or depth are acquired through laser triangulation, but the role that stereo vision plays is still confusing to me. One camera is sufficient for triangulation, so what is the purpose of applying stereo vision? Is it used to build a coordinate system based on marks?
One camera (or one eye) is not enough to compute depth. You can triangulate a position from a single point of reference using resection only where you have three or more known points in view. Stereo vision uses two cameras (or eyes) to observe the same object point from different positions a known distance apart and triangulates that way to compute a distance to the object.

A hand held laser scanner is not like a camera, it actively computes depth. This is done by sending out a beam of light, timing how long it takes to receive a reflection of that beam of light, and dividing by the speed of light to get a distance.
So, is it actually based on Time-of-Flight (TOF)? Do the cameras on a handheld laser scanner function as two receivers?
User avatar
smacl
Global Moderator
Global Moderator
Posts: 1467
Joined: Tue Jan 25, 2011 5:12 pm
13
Full Name: Shane MacLaughlin
Company Details: Atlas Computers Ltd
Company Position Title: Managing Director
Country: Ireland
Linkedin Profile: Yes
Location: Ireland
Has thanked: 645 times
Been thanked: 694 times
Contact:

Re: Need some help about the principle of hand held laser scanner

Post by smacl »

Solaris wrote: Tue Jul 09, 2024 2:27 pm
smacl wrote: Tue Jul 09, 2024 11:24 am
Solaris wrote: Tue Jul 09, 2024 11:16 am Hi everyone!
I'm trying to make a handheld laser scanner similar to HandyScan or ATOS. I understand that the 3D coordinates or depth are acquired through laser triangulation, but the role that stereo vision plays is still confusing to me. One camera is sufficient for triangulation, so what is the purpose of applying stereo vision? Is it used to build a coordinate system based on marks?
One camera (or one eye) is not enough to compute depth. You can triangulate a position from a single point of reference using resection only where you have three or more known points in view. Stereo vision uses two cameras (or eyes) to observe the same object point from different positions a known distance apart and triangulates that way to compute a distance to the object.

A hand held laser scanner is not like a camera, it actively computes depth. This is done by sending out a beam of light, timing how long it takes to receive a reflection of that beam of light, and dividing by the speed of light to get a distance.
So, is it actually based on Time-of-Flight (TOF)? Do the cameras on a handheld laser scanner function as two receivers?
Cameras on most handheld laser scanners are not being used compute coordinates. They're used to capture additional images that can be used to color the point cloud generated by the laser scanner. There are some handheld devices out there that use photogrammetry to generate a point cloud or model, but they all have multiple cameras.

There is work going on out there to calculate depth based on a single image, see https://towardsdatascience.com/how-to-e ... 421d86b22d for example, but I don't think it has been applied in this market as yet. I doubt it has anywhere near the required accuracy as yet.
Shane MacLaughlin
Atlas Computers Ltd
www.atlascomputers.ie

SCC Point Cloud module
Solaris
I have made <0 posts
I have made <0 posts
Posts: 4
Joined: Sun Jul 07, 2024 10:00 am
Full Name: Yu Huang
Company Details: ZY intelligence
Company Position Title: Algorithm Engineer
Country: China

Re: Need some help about the principle of hand held laser scanner

Post by Solaris »

smacl wrote: Tue Jul 09, 2024 3:33 pm
Solaris wrote: Tue Jul 09, 2024 2:27 pm
smacl wrote: Tue Jul 09, 2024 11:24 am

One camera (or one eye) is not enough to compute depth. You can triangulate a position from a single point of reference using resection only where you have three or more known points in view. Stereo vision uses two cameras (or eyes) to observe the same object point from different positions a known distance apart and triangulates that way to compute a distance to the object.

A hand held laser scanner is not like a camera, it actively computes depth. This is done by sending out a beam of light, timing how long it takes to receive a reflection of that beam of light, and dividing by the speed of light to get a distance.
So, is it actually based on Time-of-Flight (TOF)? Do the cameras on a handheld laser scanner function as two receivers?
Cameras on most handheld laser scanners are not being used compute coordinates. They're used to capture additional images that can be used to color the point cloud generated by the laser scanner. There are some handheld devices out there that use photogrammetry to generate a point cloud or model, but they all have multiple cameras.

There is work going on out there to calculate depth based on a single image, see https://towardsdatascience.com/how-to-e ... 421d86b22d for example, but I don't think it has been applied in this market as yet. I doubt it has anywhere near the required accuracy as yet.
Sorry, but I think we are talking about different things. The handheld laser scanner I mean is something like the HandyScan Black in this video: https://www.youtube.com/watch?v=pXmQR4dgpfg. The laser triangulation I refer to is a method to calculate depth from different angles of reflection, which is demonstrated here: https://hermary.com/learning/principles ... s%20plane..
User avatar
smacl
Global Moderator
Global Moderator
Posts: 1467
Joined: Tue Jan 25, 2011 5:12 pm
13
Full Name: Shane MacLaughlin
Company Details: Atlas Computers Ltd
Company Position Title: Managing Director
Country: Ireland
Linkedin Profile: Yes
Location: Ireland
Has thanked: 645 times
Been thanked: 694 times
Contact:

Re: Need some help about the principle of hand held laser scanner

Post by smacl »

My bad, I was thinking more of survey units like the BLK2GO, Geoslam etc...
Shane MacLaughlin
Atlas Computers Ltd
www.atlascomputers.ie

SCC Point Cloud module
User avatar
Daniel Wujanz
V.I.P Member
V.I.P Member
Posts: 233
Joined: Fri Aug 24, 2018 11:26 am
5
Full Name: Daniel Wujanz
Company Details: technet GmbH
Company Position Title: 3D Laser Scanning Specialist
Country: Germany
Linkedin Profile: Yes
Location: Berlin
Has thanked: 212 times
Been thanked: 199 times
Contact:

Re: Need some help about the principle of hand held laser scanner

Post by Daniel Wujanz »

Dear Yu,

the two mentioned sensors use different principles. As for the GOM Atos a "classical" structured light procedure is used. These systems (typically) apply the general principle of stereoscopic coordinate determination. If I remember correctly, you could also use the Atos with just one camera for occluded areas.

Two industrial cameras are mounted to a stable base which fixes their relative orientation. After calibration the absolute orientation is known which allows metric measurements. The third vital component is a projector whose task it is to provide contrast to areas that contain no or poor contrast (where photogrammetry would usually fail). There are two basic principles: phase shift and the coded light approach. A book that I highly recommend is the following:

https://www.degruyter.com/document/doi/ ... #contents

I have an older addition of the book which features everything you need.

The second procedure looks like "laser" triangulation...without using a laser. Instead a projected pattern appears to be used instead of a single projected laser spot. Many systems project in the near infrared. The first system (that I was aware off) which used a comparable strategy was the first generation of the Kinect https://en.wikipedia.org/wiki/Kinect which was of course designed for non-metrological applications.

I hope this helps for a start.

All the best

Daniel
Solaris
I have made <0 posts
I have made <0 posts
Posts: 4
Joined: Sun Jul 07, 2024 10:00 am
Full Name: Yu Huang
Company Details: ZY intelligence
Company Position Title: Algorithm Engineer
Country: China

Re: Need some help about the principle of hand held laser scanner

Post by Solaris »

I now understand that the handheld laser scanner is essentially functioning as a stereo camera. The laser in this system serves as structured light, which provides matched points in two images.
Thank you for the assistance from all of you.
Post Reply

Return to “Hand Held Scanners”