Forum

Author Topic: Tetracam multispectral imagery processing  (Read 30015 times)

ricardo9856

  • Newbie
  • *
  • Posts: 6
    • View Profile
Tetracam multispectral imagery processing
« on: May 28, 2013, 06:09:54 PM »
Hello

i?m a uav developer and operator for a few years, and just  now i started to view the post-processing phase, and i loved how easy and accurate are the results of agisoft photoscan with a rgb camera.

but i?m now with a problem, i also need to use a Tetracam ADC multispectral camera, with R-G-Nir bands. actually, its important to me to process the NDVI created with the camera software.

i can have the images in jpg, tiff and multipage tiff, but i still couldn?t aling them, even with gps position on the exif files. Is there any option i?m missing? when the software can align some photos, it does it totally wrong, as if they were taken paralell to the ground, and not 90 degrees from the top.

i?m still on the demo version, but if i could process these images, i?ll buy it for sure. if someone needs some of the raw images, i can post them on my 4shared account.

RalfH

  • Sr. Member
  • ****
  • Posts: 344
    • View Profile
Re: Tetracam multispectral imagery processing
« Reply #1 on: May 28, 2013, 10:03:27 PM »
Have you tried to extract greyscale (single band) images from the multispectral data and process these?

By the way, the tretracam seems to be an interesting product (but still way beyond my budget...)
« Last Edit: May 28, 2013, 10:06:12 PM by RalfH »

ricardo9856

  • Newbie
  • *
  • Posts: 6
    • View Profile
Re: Tetracam multispectral imagery processing
« Reply #2 on: May 28, 2013, 10:24:09 PM »
yes, i tried the 3-band artificially color-processed, R band, G band, NIR band, NDVI Band, multi-page tiff with 3 bands (r-g-nir), and i always get similar results. i?ll run the process again, and take some screenshots when i got time, i believe i can do this tomorrow.

an198317

  • Newbie
  • *
  • Posts: 42
    • View Profile
Re: Tetracam multispectral imagery processing
« Reply #3 on: May 30, 2013, 02:25:25 AM »
We had a company to convert a Canon DSLR to NIR-R-G camera for us. The idea is open the Blue channel  for NIR and keep R and G. But the problem is R and G channels are also effected. We had the same company converted a false NDVI camera: NIR-G-B bands.

I studied the Tetracam cameras, the resolution is too low and some camera technology is too behind based on our findings. I am glad to learn if I am wrong about this camera.

Nan

Alexey Pasumansky

  • Agisoft Technical Support
  • Hero Member
  • *****
  • Posts: 14813
    • View Profile
Re: Tetracam multispectral imagery processing
« Reply #4 on: May 31, 2013, 01:05:11 PM »
Hello ricardo9856,

Please check pre-processing instruction for TetraCam multispectral imagery processing: http://downloads.agisoft.ru/pdf/Tetracam%20multispectral%20imagery%20processing.pdf

After the multipage TIFF files are prepared you can load them to PhotoScan Pro. The workflow is similar to common photos processing, but for multiband data you need to select Master Channel first (from context menu by right-clicking on the chunk label in the Workspace pane). All the following processing steps will be applied to the master channel (camera positions will be estimated based on the master channel data and geometry will be reconstructed accordingly).
But when you use Export Orthophoto command the resulting TIFF file will contain all the original layers (bands).
We recommend to use 0.9.1 pre-release for multipage TIFFs.

However, we've faced problems with aligning the ADC lite imagery caused by long rolling shutter time used for specific datasets.
Best regards,
Alexey Pasumansky,
Agisoft LLC

ricardo9856

  • Newbie
  • *
  • Posts: 6
    • View Profile
Re: Tetracam multispectral imagery processing
« Reply #5 on: May 31, 2013, 10:54:11 PM »
thanks, Alexey. i haven't tried that before, i was trying to run the multipage tiffs, or the single channels.

i made a low-speed flight, about 5m/s, 150m hight, with raw and no delay. i'll try it, as soon as possible i'll come back with news from the results!

ricardo9856

  • Newbie
  • *
  • Posts: 6
    • View Profile
Re: Tetracam multispectral imagery processing
« Reply #6 on: June 01, 2013, 05:04:10 PM »
just tried to align 10 photos, without gps information, and it worked! not so well...but much better than before.

monday i'll add the geotag and try again with the photos of the entire flight. thanks again for the support!

HyperFox

  • Newbie
  • *
  • Posts: 18
    • View Profile
Re: Tetracam multispectral imagery processing
« Reply #7 on: June 05, 2013, 09:02:03 AM »
Hi everyone,

longtime reader, first time poster...

Ricardo, I'm sorry, but I think, you can forget the Tetracam. This camera has an extreme rolling shutter, see attachment of this post. Unfortunately, two years ago, I also trapped on the product description and bought an ADC. This camera is useless in ANY moving camera platform. This means, you'll never ever get undistorted images putting the camera in a UAV (Copter) or even in shaky hands. You can produce the effekt on your own: Just take the camera in your hands and shake it during the exposure...

Ok, there might be one case, where you might get some results that "might look promising":
Mounting the Tetracam in a fixed wing aircraft, which is moving fast forward. Then, you'll also get distorted images, but maybe, with individual camera calibration option in PS, you can "fix" this.

But, the plane WILL roll during exposure, so you'll never get correct results, only results that "might look correct".

In my humble opinion: ADC Tetracam is rubbish. I tried to return it to my vendor, they neglected.

RalfH

  • Sr. Member
  • ****
  • Posts: 344
    • View Profile
Re: Tetracam multispectral imagery processing
« Reply #8 on: June 05, 2013, 09:53:38 AM »
AnsgarG, thanks for sharing your experience. This is a "nice" example for the rolling shutter problem.

Photoscan will not be able to correct for rolling shutter even with individual camera calibration because Photoscan only takes into account radial distortions. Non-radial distortions such as rolling shutter will mess up everything.

Still, there may be two possibilities:
(a) Measure camera rotations at high frequency (using a gyro) and correct distortions after the flight. You'll need a gyro measurement frequency that is at least as high as the sensor line reading frequency of the camera. Do you know the total time the camera needs to take a single picture?
(b) Stabilise the camera with the help of gyros (check out this one: http://www.youtube.com/watch?v=CkUoU0f_o4Q).

ricardo9856

  • Newbie
  • *
  • Posts: 6
    • View Profile
Re: Tetracam multispectral imagery processing
« Reply #9 on: June 05, 2013, 07:10:22 PM »
Thanks fot the reply, AngasG. i never trusted the tetracam, it looks like it stopped in time a decade ago, with the low 3mp and that giant memory card for the adc. i just didn?t found any other multiscpectral camera yet.

i?m now thinking about the maxmax modified cameras, maybe the canon elph300 for its smaller size. Do you have any other hint?

http://maxmax.com/vegetation_stress_mkii.htm

yesterday i achieved a decent result with the combination of a RGB and a NIR-modified camera. i was able to create NRG and NDVI with the RGB and NIR mosaics. i didn?t took too much time aligning the bands, so the result is far from perfect, but now that i learned how to do it, i?ll make it again, beeing more careful.

i?ll upload these images i got and i?ll post them here.
« Last Edit: June 05, 2013, 07:30:21 PM by ricardo9856 »

an198317

  • Newbie
  • *
  • Posts: 42
    • View Profile
Re: Tetracam multispectral imagery processing
« Reply #10 on: June 05, 2013, 08:00:14 PM »
Hi ricardo9856,

I am glad to learn you are giving up on Tetracam that I gave up long time ago. My lab works very closely with MaxMax, which is the company I mentioned in the previous response. We had them convert 4 Canon S95 and S100 cameras to NIR-G-B and 2 Canon XSi cameras to NIR camera.

My lab and Maxmax just came up an idea to convert Canon 6D Full frame cam to NIR-R-G camera. And we are still testing it. I will post more once we have the test results.

Nan

ricardo9856

  • Newbie
  • *
  • Posts: 6
    • View Profile
Re: Tetracam multispectral imagery processing
« Reply #11 on: June 05, 2013, 09:36:11 PM »
Nan, if you get a Canon 6D, or other camera, to get NIR-R-G, as does tetracam, it will be a very nice feature.

heres the generated mosaic of my NIR-R-G, the NIR wasn?t too much aligned with RGB, so the result isn?t perfect, but it?s good enough for the moment.

the area is about 0,5km2, and the resolution about 5cm/pixel with a 12mp camera.

Martin Herkommer

  • Newbie
  • *
  • Posts: 3
    • View Profile
Re: Tetracam multispectral imagery processing
« Reply #12 on: September 16, 2013, 02:57:24 PM »
Dear AnsgarG, dear Agisoft community,

I am a member of the geo-Konzept (European TetraCam reseller) remote sensing team since 15 years and I am using TetraCam sensors for more than 13 years with Aircraft and with UAVs.

It works perfect, if

a) the UAV is NOT a low cost UAV with ultrahigh vibrations
b) you are using a professional gimbal Mount, isolating any vibration
c) the gimbal Mount is active but not too agressive. No fast roll movements! We need smooth movements.
d) we use short exposure times. Therefore we need good daylight conditions

It is not obligatory to use a GPS!

Agissoft Photoscan Pro is used by many customers and a large scientific community.

Here is an example of TetraCam data processed with Agisoft.
Elevation: 1700m NN
Wind Speed 60km/h
8 RTK Ground Control Points

Compared to a SonyNex7 DGM we are only 8cm of in the average and the Standard Deviation is 30cm in Elevation.

Please contact me of you Need any help with your TetraCam sensors! You are welcome!

Cheers
Martin

PS: Demonstrations are possible at any time!   
« Last Edit: September 16, 2013, 07:43:20 PM by Martin Herkommer »

Martin Herkommer

  • Newbie
  • *
  • Posts: 3
    • View Profile
Re: Tetracam multispectral imagery processing
« Reply #13 on: September 16, 2013, 07:46:45 PM »
- The rolling shutter effect is normal for CMOS sensors used on UAVs with high Vibration or fast roll movements.
- We offered a fair trade in of your ADC even many months after your purchase

 


Hi everyone,

longtime reader, first time poster...

Ricardo, I'm sorry, but I think, you can forget the Tetracam. This camera has an extreme rolling shutter, see attachment of this post. Unfortunately, two years ago, I also trapped on the product description and bought an ADC. This camera is useless in ANY moving camera platform. This means, you'll never ever get undistorted images putting the camera in a UAV (Copter) or even in shaky hands. You can produce the effekt on your own: Just take the camera in your hands and shake it during the exposure...

Ok, there might be one case, where you might get some results that "might look promising":
Mounting the Tetracam in a fixed wing aircraft, which is moving fast forward. Then, you'll also get distorted images, but maybe, with individual camera calibration option in PS, you can "fix" this.

But, the plane WILL roll during exposure, so you'll never get correct results, only results that "might look correct".

In my humble opinion: ADC Tetracam is rubbish. I tried to return it to my vendor, they neglected.

Martin Herkommer

  • Newbie
  • *
  • Posts: 3
    • View Profile
Re: Tetracam multispectral imagery processing
« Reply #14 on: September 18, 2013, 01:23:15 PM »
Hi Community,

today was helping a member of this community with his ADC data. I had no Problems to create a perfect Ortho Mosaik with Agisoft PhotoScan.

Attached some Screen shots showing the steps ...

@AngarG: No rolling shutter :-) 

Again: If some others of you need my help with ADC data, please get in touch with me !

Cheers
Martin