Convert GPS to HShift and VShift

classic Classic list List threaded Threaded
10 messages Options
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Convert GPS to HShift and VShift

PanoTools NG mailing list
Hi all,

 

I'm doing some aerial crop surveys and am trying to improve and streamline
the stitching process in PTGui. Due to the nature of the crops, (apple
orchards in this case), there aren't always enough unique features for PTGui
to latch onto and create control points.

 

Since these are from a moving platform, the primary difference between each
image is the HShift and VShift parameters, with YPR playing a secondary
role. If I recall correctly, the FoV is best set to something like 2000 and
NOT taken into the optimization. I did a test this morning with a handful of
images, and that produced a nice result.

 

My question is, can I use the embedded GPS coordinates in each image to
produce at least estimated values for the HShift and VShift parameters? That
would get all the images properly positioned in a grid, hopefully making the
control point generator more efficient and accurate. I'm not sure if I can
pull a heading value as well, which would give me approximate Roll values.

 

My initial guess is that I would need to start with a base image, treating
that like an anchor, then calculate HShift and VShift as a relative
departures from the anchor image. Am I correct in that line of thinking?

 

Erik - how's that for something for the group to chew on?  :o)

 

Thanks,


Mark

 

Mark Fink

www.NorthernLight.net <http://www.NorthernLight.net>

1-845-380-0068

Images That Transport You

 

Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Re: Convert GPS to HShift and VShift

PanoTools NG mailing list
I am very interested in this answer as well.. I have been trying to stitch
aerial images from UAV with panotools to no avail...

Regards

Andrés
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Re: Convert GPS to HShift and VShift

PanoTools NG mailing list
In reply to this post by PanoTools NG mailing list
'Mark Fink' [hidden email] [PanoToolsNG] wrote:
> My question is, can I use the embedded GPS coordinates in each image to produce at least estimated values for the HShift and VShift parameters? That would get all the images properly positioned in a grid, hopefully making the control point generator more efficient and accurate.
> My initial guess is that I would need to start with a base image, treating that like an anchor, then calculate HShift and VShift as a relative departures from the anchor image. Am I correct in that line of thinking?

That should work - I would be inclined to make up a small grid of images "by hand", and reverse engineer the
mapping from GPS coord to HShift/VShift from the result.

  BugBear

Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Re: Convert GPS to HShift and VShift

PanoTools NG mailing list
In reply to this post by PanoTools NG mailing list
Am 28.04.2016 um 14:42 schrieb Mark Fink:
> My question is, can I use the embedded GPS coordinates in each image to
> produce at least estimated values for the HShift and VShift parameters? That
> would get all the images properly positioned in a grid, hopefully making the
> control point generator more efficient and accurate

Given the best accuracy that can be achieved with classical GPS is 3
meters (typical 6 meters) this would probably work only if the distance
between two shots is fairly larger.

Apart from that: exiftool command line version can extract any value.
You'd need a script to calculate differences and write a .PTS or .PTO
file f.e.

--
Erik Krause


------------------------------------

------------------------------------

--
<*> Wiki: http://wiki.panotools.org
<*> User Guidelines: http://wiki.panotools.org/User_Guidelines
<*> Nabble (Web) http://panotoolsng.586017.n4.nabble.com/
<*> NG Member Map http://www.panomaps.com/ng
<*> Moderators/List Admins: [hidden email]

------------------------------------

Yahoo Groups Links

<*> To visit your group on the web, go to:
    http://groups.yahoo.com/group/PanoToolsNG/

<*> Your email settings:
    Individual Email | Traditional

<*> To change settings online go to:
    http://groups.yahoo.com/group/PanoToolsNG/join
    (Yahoo! ID required)

<*> To change settings via email:
    [hidden email]
    [hidden email]

<*> To unsubscribe from this group, send an email to:
    [hidden email]

<*> Your use of Yahoo Groups is subject to:
    https://info.yahoo.com/legal/us/yahoo/utos/terms/

Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Re: Convert GPS to HShift and VShift

PanoTools NG mailing list
In reply to this post by PanoTools NG mailing list
To do this manually, make sure you click the Shift boxes for all images in the Lens Settings tab and manually set the Focal length to something like 2000. You’ll need to be in advanced mode for this. Then on the Optimizer tab, uncheck everything on the left side under the Optimize globally column. You should see Yaw, Pitch, Roll and HShift as well as VShift boxes all checked.

 

Let me know if that helps!

 

Mark

 

Mark Fink

 <http://www.NorthernLight.net> www.NorthernLight.net

1-845-380-0068

Images That Transport You

 

From: [hidden email] [mailto:[hidden email]]
Sent: Thursday, April 28, 2016 9:24 AM
To: [hidden email]
Subject: [PanoToolsNG] Re: Convert GPS to HShift and VShift

 






I am very interested in this answer as well.. I have been trying to stitch aerial images from UAV with panotools to no avail...

 

Regards

 

Andrés








Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

RE: Convert GPS to HShift and VShift

PanoTools NG mailing list
In reply to this post by PanoTools NG mailing list
Hi Mark.  Try Googling "drone mapping software."  I've seen other drone
pilots discussing several programs, although I'm not familiar with them
myself.

 

Ed

http://BigEyeInTheSKy.com

 

 

From: [hidden email] [mailto:[hidden email]]
Sent: Thursday, April 28, 2016 7:42 AM
To: [hidden email]
Subject: [PanoToolsNG] Convert GPS to HShift and VShift

 

 

Hi all,

 

I'm doing some aerial crop surveys and am trying to improve and streamline
the stitching process in PTGui. Due to the nature of the crops, (apple
orchards in this case), there aren't always enough unique features for PTGui
to latch onto and create control points.

 

Since these are from a moving platform, the primary difference between each
image is the HShift and VShift parameters, with YPR playing a secondary
role. If I recall correctly, the FoV is best set to something like 2000 and
NOT taken into the optimization. I did a test this morning with a handful of
images, and that produced a nice result.

 

My question is, can I use the embedded GPS coordinates in each image to
produce at least estimated values for the HShift and VShift parameters? That
would get all the images properly positioned in a grid, hopefully making the
control point generator more efficient and accurate. I'm not sure if I can
pull a heading value as well, which would give me approximate Roll values.

 

My initial guess is that I would need to start with a base image, treating
that like an anchor, then calculate HShift and VShift as a relative
departures from the anchor image. Am I correct in that line of thinking?

 

Erik - how's that for something for the group to chew on?  :o)

 

Thanks,


Mark

 

Mark Fink

www.NorthernLight.net <http://www.NorthernLight.net>

1-845-380-0068

Images That Transport You

 



Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Re: Convert GPS to HShift and VShift

PanoTools NG mailing list
In reply to this post by PanoTools NG mailing list
That's a good point about the precision of the GPS numbers. The hexacopter
is traveling at 8m/s, and the camera is firing every five seconds, so there
should be roughly 40 meters between each image.

So, pull the values from the images using exiftool, write them to perhaps a
comma separated value file, then open that in Excel? Once in Excel, I could
probably figure out a way to calculate the differences and convert them into
a format for PTGui. Thanks for pointing me in the right direction. I'll let
you know my progress in the next few days.

Mark

Mark Fink
www.NorthernLight.net
1-845-380-0068
Images That Transport You

-----Original Message-----
From: [hidden email] [mailto:[hidden email]]
Sent: Thursday, April 28, 2016 9:57 AM
To: [hidden email]
Subject: [PanoToolsNG] Re: Convert GPS to HShift and VShift

Am 28.04.2016 um 14:42 schrieb Mark Fink:
> My question is, can I use the embedded GPS coordinates in each image
> to produce at least estimated values for the HShift and VShift
> parameters? That would get all the images properly positioned in a
> grid, hopefully making the control point generator more efficient and
> accurate

Given the best accuracy that can be achieved with classical GPS is 3 meters
(typical 6 meters) this would probably work only if the distance between two
shots is fairly larger.

Apart from that: exiftool command line version can extract any value.
You'd need a script to calculate differences and write a .PTS or .PTO file
f.e.

--
Erik Krause


------------------------------------

------------------------------------

--
------------------------------------

Yahoo Groups Links




Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Re: Convert GPS to HShift and VShift

PanoTools NG mailing list
Am 28.04.2016 um 21:20 schrieb Mark Fink:
> So, pull the values from the images using exiftool, write them to perhaps a
> comma separated value file, then open that in Excel? Once in Excel, I could
> probably figure out a way to calculate the differences and convert them into
> a format for PTGui.

The hard part will be to map coord differences to shift values because
this also depends on height above ground. May be this could be estimated
if you have aligned the first two images.

You write that only few images have no control points. In this case it
might be easier to simply interpolate the positions in between,
eventually even with aid of GPS coords...

--
Erik Krause
http://www.erik-krause.de


------------------------------------

------------------------------------

--
<*> Wiki: http://wiki.panotools.org
<*> User Guidelines: http://wiki.panotools.org/User_Guidelines
<*> Nabble (Web) http://panotoolsng.586017.n4.nabble.com/
<*> NG Member Map http://www.panomaps.com/ng
<*> Moderators/List Admins: [hidden email]

------------------------------------

Yahoo Groups Links

<*> To visit your group on the web, go to:
    http://groups.yahoo.com/group/PanoToolsNG/

<*> Your email settings:
    Individual Email | Traditional

<*> To change settings online go to:
    http://groups.yahoo.com/group/PanoToolsNG/join
    (Yahoo! ID required)

<*> To change settings via email:
    [hidden email]
    [hidden email]

<*> To unsubscribe from this group, send an email to:
    [hidden email]

<*> Your use of Yahoo Groups is subject to:
    https://info.yahoo.com/legal/us/yahoo/utos/terms/

Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

RE: Convert GPS to HShift and VShift

PanoTools NG mailing list
In reply to this post by PanoTools NG mailing list
That's the plan. I have already done a small batch with seven images and
will work backwards from there.

Mark

Mark Fink
www.NorthernLight.net
1-845-380-0068
Images That Transport You

-----Original Message-----
From: [hidden email] [mailto:[hidden email]]
Sent: Thursday, April 28, 2016 9:54 AM
To: [hidden email]
Subject: Re: [PanoToolsNG] Convert GPS to HShift and VShift

'Mark Fink' [hidden email] [PanoToolsNG] wrote:
> My question is, can I use the embedded GPS coordinates in each image to
produce at least estimated values for the HShift and VShift parameters? That
would get all the images properly positioned in a grid, hopefully making the
control point generator more efficient and accurate.
> My initial guess is that I would need to start with a base image, treating
that like an anchor, then calculate HShift and VShift as a relative
departures from the anchor image. Am I correct in that line of thinking?

That should work - I would be inclined to make up a small grid of images "by
hand", and reverse engineer the mapping from GPS coord to HShift/VShift from
the result.

  BugBear



------------------------------------
Posted by: paul womack <[hidden email]>
------------------------------------

--
------------------------------------

Yahoo Groups Links




Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Re: Convert GPS to HShift and VShift

PanoTools NG mailing list
In reply to this post by PanoTools NG mailing list
When I throw all 551 images at PTGui, it isn't finding control points for
dozens of images. I think part of my problem may be not enough side overlap
between rows of images. I have plenty as the hexacopter moves forward, since
that doesn't require extra passes. Extra rows requires more back and forth
passes, which adds to flying time, and requires more battery changes. I will
be doing more aerial surveys this weekend and will add more rows for
increased side overlap. I'll see if that yields better results.

Apologies to those for whom this is becoming increasingly off topic...

Mark

Mark Fink
www.NorthernLight.net
1-845-380-0068
Images That Transport You


-----Original Message-----
From: [hidden email] [mailto:[hidden email]]
Sent: Thursday, April 28, 2016 5:20 PM
To: [hidden email]
Subject: [PanoToolsNG] Re: Convert GPS to HShift and VShift

Am 28.04.2016 um 21:20 schrieb Mark Fink:
> So, pull the values from the images using exiftool, write them to
> perhaps a comma separated value file, then open that in Excel? Once in
> Excel, I could probably figure out a way to calculate the differences
> and convert them into a format for PTGui.

The hard part will be to map coord differences to shift values because this
also depends on height above ground. May be this could be estimated if you
have aligned the first two images.

You write that only few images have no control points. In this case it might
be easier to simply interpolate the positions in between, eventually even
with aid of GPS coords...

--
Erik Krause
http://www.erik-krause.de


------------------------------------

------------------------------------

--
------------------------------------

Yahoo Groups Links




Loading...