Cyclone point cloud classification

Discuss all Cyclone related issues here.
Post Reply
tsepchar
I have made <0 posts
I have made <0 posts
Posts: 1
Joined: Fri Jan 25, 2019 2:06 pm
Full Name: charilaos tsepas
Company Details: datum monitoring ltd
Company Position Title: senior surveyor
Country: uk
Linkedin Profile: Yes

Cyclone point cloud classification

Post by tsepchar » Wed Jul 10, 2019 12:24 pm

Hello to all,

My client asked me to separate my point-cloud into different layers for example ''vegetation'' except from region grow for the road i could not find something else. Any suggestions for cyclone 9.4 would be much appreciated.

Regards to all

User avatar
pburrows145
V.I.P Member
V.I.P Member
Posts: 1781
Joined: Wed Sep 12, 2007 9:16 am
Full Name: Paul Burrows
Company Details: Leica Geosystems Europe
Company Position Title: Scanning Solutions Manager - Europe
Country: UK
Linkedin Profile: Yes
Location: UK
Has thanked: 10 times
Been thanked: 8 times

Re: Cyclone point cloud classification

Post by pburrows145 » Wed Jul 10, 2019 5:12 pm

Give 3DReshaper (soon to become Cyclone 3DR) a go... has some really cool classification tool, plus scripting. You can also segment by colour and distance which sometimes provides really interesting results - and neatly classifies stuff.

There is nothing in Cyclone as of yet which will perform this function. It would be a matter of segmentation by intensity, or manual tidying and layering.

Scott
V.I.P Member
V.I.P Member
Posts: 800
Joined: Tue Mar 29, 2011 7:39 pm
Full Name: Scott Page
Company Details: Scott Page Design- Architectural service
Company Position Title: Owner
Country: USA
Linkedin Profile: No
Location: Berkeley, CA USA
Has thanked: 14 times
Been thanked: 9 times
Contact:

Re: Cyclone point cloud classification

Post by Scott » Wed Jul 10, 2019 5:52 pm

tsepchar wrote:
Wed Jul 10, 2019 12:24 pm
Hello to all,
My client asked me to separate my point-cloud into different layers for example ''vegetation'' except from region grow for the road i could not find something else. Any suggestions for cyclone 9.4 would be much appreciated.
Regards to all
'Segmentation' https://en.wikipedia.org/wiki/Image_segmentation
"In computer vision, image segmentation is the process of partitioning a digital image into multiple segments (sets of pixels, also known as super-pixels). The goal of segmentation is to simplify and/or change the representation of an image into something that is more meaningful and easier to analyze.[1][2] Image segmentation is typically used to locate objects and boundaries (lines, curves, etc.) in images. More precisely, image segmentation is the process of assigning a label to every pixel in an image such that pixels with the same label share certain characteristics."

mike annear
V.I.P Member
V.I.P Member
Posts: 479
Joined: Mon Jun 08, 2009 5:45 am
Full Name: Mike Annear
Company Details: Mike Annear
Company Position Title: 3D Modeler
Country: Australia
Linkedin Profile: Yes
Location: On an island off Brisbane ...
Been thanked: 9 times
Contact:

Re: Cyclone point cloud classification

Post by mike annear » Wed Jul 10, 2019 11:49 pm

CloudCompare or Lastools,
Both have a bit of a learning curve but the results are worth it.
Regards,
Mike.

User avatar
gsisman
V.I.P Member
V.I.P Member
Posts: 318
Joined: Fri Oct 07, 2016 1:51 pm
Full Name: Steve Long
Company Details: Montgomery County DOT _ MD
Company Position Title: Land Survey Supervisor
Country: United States
Skype Name: gsisman1
Linkedin Profile: Yes
Has thanked: 62 times
Been thanked: 14 times

Re: Cyclone point cloud classification

Post by gsisman » Thu Jul 11, 2019 5:44 pm

Scott wrote:
Wed Jul 10, 2019 5:52 pm
tsepchar wrote:
Wed Jul 10, 2019 12:24 pm
Hello to all,
My client asked me to separate my point-cloud into different layers for example ''vegetation'' except from region grow for the road i could not find something else. Any suggestions for cyclone 9.4 would be much appreciated.
Regards to all
'Segmentation' https://en.wikipedia.org/wiki/Image_segmentation
"In computer vision, image segmentation is the process of partitioning a digital image into multiple segments (sets of pixels, also known as super-pixels). The goal of segmentation is to simplify and/or change the representation of an image into something that is more meaningful and easier to analyze.[1][2] Image segmentation is typically used to locate objects and boundaries (lines, curves, etc.) in images. More precisely, image segmentation is the process of assigning a label to every pixel in an image such that pixels with the same label share certain characteristics."
Question, most lidar software claasifies by infrared, reflectivity and color to create segmentation -correct. Can the same be done with Terrestrial scanner using perpendicular to gravity scanners? There are multiple returns in Lidar which also eases some coarse classification (Treetops, rooftops, low vegetation , bare earth,etc)

Scott
V.I.P Member
V.I.P Member
Posts: 800
Joined: Tue Mar 29, 2011 7:39 pm
Full Name: Scott Page
Company Details: Scott Page Design- Architectural service
Company Position Title: Owner
Country: USA
Linkedin Profile: No
Location: Berkeley, CA USA
Has thanked: 14 times
Been thanked: 9 times
Contact:

Re: Cyclone point cloud classification

Post by Scott » Thu Jul 11, 2019 6:13 pm

gsisman wrote:
Thu Jul 11, 2019 5:44 pm
Scott wrote:
Wed Jul 10, 2019 5:52 pm
tsepchar wrote:
Wed Jul 10, 2019 12:24 pm
Hello to all,
My client asked me to separate my point-cloud into different layers for example ''vegetation'' except from region grow for the road i could not find something else. Any suggestions for cyclone 9.4 would be much appreciated.
Regards to all
'Segmentation' https://en.wikipedia.org/wiki/Image_segmentation
"In computer vision, image segmentation is the process of partitioning a digital image into multiple segments (sets of pixels, also known as super-pixels). The goal of segmentation is to simplify and/or change the representation of an image into something that is more meaningful and easier to analyze.[1][2] Image segmentation is typically used to locate objects and boundaries (lines, curves, etc.) in images. More precisely, image segmentation is the process of assigning a label to every pixel in an image such that pixels with the same label share certain characteristics."
Question, most lidar software claasifies by infrared, reflectivity and color to create segmentation -correct. Can the same be done with Terrestrial scanner using perpendicular to gravity scanners? There are multiple returns in Lidar which also eases some coarse classification (Treetops, rooftops, low vegetation , bare earth,etc)
I've been waiting for improved segmentation tools for years--mostly from AI developers. Most of the 'progress' has been towards integration with BIM platforms. While I understand the logic (I'm an architect, after all), it's created a bloated, fragmented software market. Perhaps it's time to re-evaluate the entire process and what it is best suited for. Have a look at medical imaging, too (CT, MR). 2D is still important when it clarifies what you are looking at for basic decision making.

Post Reply

Return to “Leica Cyclone”