dji phantom 4 fcc330 lens correction profile+parameters for photoscan

[olivine]

Hi
The Adobe Lightroom has new update in June. I have uploaded the lens correction profile files of dji phantom 4 fcc330.

How should I fill the Calibration parameters list in photoscan for all the images taken by phantom 4? What are the values for the following parameters:

fx, fy

cx, cy

skew

k1, k2, k3, k4

p1, p2, p3, p4

I am new to this field, and I feel there are parameters missing from the .lcp files.

Thank you in advance!

[olivine]

 Alexey, could you plz comment on that?

Thanks

[Alexey Pasumansky]

Hello olivine,

You can find the description of the distortion model in PhotoScan Manual:
http://www.agisoft.com/pdf/photoscan-pro_1_2_en.pdf

[olivine]

Hello olivine,

You can find the description of the distortion model in PhotoScan Manual:
http://www.agisoft.com/pdf/photoscan-pro_1_2_en.pdf

Thank you Alexey.

I have read the manual. Two questions for you:
1/the .lcp file is:
<x:xmpmeta xmlns:x="adobe:ns:meta/" x:xmptk="Adobe XMP Core 5.6-c128 79.159124, 2016/03/18-14:01:55        ">
 <rdf:RDF xmlns:rdf="http://www.w3.org/1999/02/22-rdf-syntax-ns#">
  <rdf:Description rdf:about=""
    xmlns:photoshop="http://ns.adobe.com/photoshop/1.0/"
    xmlns:stCamera="http://ns.adobe.com/photoshop/1.0/camera-profile">
   <photoshop:CameraProfiles>
    <rdf:Seq>
     <rdf:li>
      <rdf:Description
       stCamera:Author="Adobe (www.adobe.com)"
       stCamera:Make="DJI"
       stCamera:Model="FC330"
       stCamera:CameraRawProfile="False"
       stCamera:CameraPrettyName="FC330"
       stCamera:LensPrettyName="PHANTOM 4 FC330"
       stCamera:ProfileName="Adobe (PHANTOM 4 FC330)"
       stCamera:SensorFormatFactor="5.7"
       stCamera:ImageWidth="4000"
       stCamera:ImageLength="3000"
       stCamera:FocusDistance="1.322811"
       stCamera:ApertureValue="2.970854">
      <stCamera:PerspectiveModel
       stCamera:Version="2"
       stCamera:FocalLengthX="0.581668"
       stCamera:FocalLengthY="0.581668"
       stCamera:ImageXCenter="0.49812"
       stCamera:ImageYCenter="0.477037"
       stCamera:ResidualMeanError="0.000203"
       stCamera:ResidualStandardDeviation="0.000151"
       stCamera:RadialDistortParam1="-0.004567"
       stCamera:RadialDistortParam2="-0.006062"
       stCamera:RadialDistortParam3="0.003042"/>
      </rdf:Description>
     </rdf:li>
     <rdf:li>
      <rdf:Description
       stCamera:Author="Adobe (www.adobe.com)"
       stCamera:Make="DJI"
       stCamera:Model="FC330"
       stCamera:CameraRawProfile="False"
       stCamera:CameraPrettyName="FC330"
       stCamera:LensPrettyName="PHANTOM 4 FC330"
       stCamera:ProfileName="Adobe (PHANTOM 4 FC330)"
       stCamera:SensorFormatFactor="5.7"
       stCamera:ImageWidth="4000"
       stCamera:ImageLength="3000"
       stCamera:FocusDistance="3"
       stCamera:ApertureValue="2.970854">
      <stCamera:PerspectiveModel>
       <rdf:Description
        stCamera:Version="2"
        stCamera:FocalLengthX="0.581668"
        stCamera:FocalLengthY="0.581668"
        stCamera:ImageXCenter="0.49812"
        stCamera:ImageYCenter="0.477037"
        stCamera:RadialDistortParam1="-0.004567"
        stCamera:RadialDistortParam2="-0.006062"
        stCamera:RadialDistortParam3="0.003042">
       <stCamera:VignetteModel
        stCamera:FocalLengthX="0.581668"
        stCamera:FocalLengthY="0.581668"
        stCamera:VignetteModelParam1="-0.056096"
        stCamera:VignetteModelParam2="0.036316"
        stCamera:VignetteModelParam3="-0.019203"/>
       </rdf:Description>
      </stCamera:PerspectiveModel>
      </rdf:Description>
     </rdf:li>
    </rdf:Seq>
   </photoshop:CameraProfiles>
  </rdf:Description>
 </rdf:RDF>
</x:xmpmeta>

I can only find fx, fy, k1, k2, k3 directly.

What are cx, cy, skew, k4, p1, p2, p3, p4?

2/Assume we have all the parameters, is the DEM generated by Dji Phantom 4 camera reliable for topographic research?

A separate question is:
If there is no GPS info in the images, will Photoscan still be able to align the photoes? Take longer time?
 

[Dmitry Semyonov]

Hello olivine,

I can only find fx, fy, k1, k2, k3 directly.

What are cx, cy, skew, k4, p1, p2, p3, p4?

For photogrammetric processing camera calibration parameters need to be calibrated very precisely. By default PhotoScan performs auto calibration during image alignment/optimization and estimates required calibration coefficients automatically.

Only in case when your particular camera specimen is calibrated in the laboratory you can use such calibration for photogrammetric processing.

Information from Adobe Lightroom profile is definitely not accurate enough as it was created for some other camera specimen (or even averaged over a set of similar units) even if it is of the same model. Such information can still be used as a first approximation that will be further refined during auto calibration, but in most cases it will not make results better than when using PhotoScan defaults.

Assume we have all the parameters, is the DEM generated by Dji Phantom 4 camera reliable for topographic research?

Based on our experience PHANTOM camera is not very good for photogrammetric processing, most likely because of automatic distortion compensation performed by this camera.

For better results we recommend to disable distortion compensation (if this can be configured) or capture photos in RAW format instead of JPEG.

If there is no GPS info in the images, will Photoscan still be able to align the photoes? Take longer time?

Image alignment without GPS data will be much slower, especially for large data sets. Nevertheless, for small data sets (up to 500 images or so) you can perform processing without GPS in reasonable time. In this case Pair preselection option in Align Photos dialog should be set to Generic.

[olivine]

Thank you for your reply! It is very useful!

I appreciate your time!

[sakaic]

I just obtained a phantom 4 as a cheap deployment over my S900+Nikon D5300 setup. The phantom 4 is a large improvement over the previous models especially the phantom 2 vision plus. If there is GPS lock, every photo is geotagged with long, lat, and alt data included in EXIF data. RAW is also available in DNG format which can now be directly processed but can only be taken at a 10 second interval. Jpegs are captured at a 5 second interval. The GPS is definitely not centimeter accurate but good enough for alignment assistance.

If anyone has any experience with the phantom 4, I would like to know if you are obtaining better results with fish-eye model or frame?

[CamMakoJ]

I am intrigued as well, we have had fairly good results with the regular frame but havent tried the fisheye compensation yet