Infragram Webcam Calibration for NDVI

[bernardo araujo rodrigues]

Hello.

My name is Bernardo Rodrigues, I am working at a Brazilian Agricultural Research Institute called Embrapa.

We are trying to implement some solutions for NDVI calculation using PublicLab's Infragram Webcam.

I am having some issues with it.

I have disabled UVC's auto white balance and exposition.

The webcam image looks too reddish and dark (different from the orangish expected tone), and all NDVI pixels are mostly inside the range 0 to 1.

Feels like I'm missing some calibration with the webcam.

Anyone has any hints?

​

Print Screen above shows webcam image, ndvi image, v4l controls and histograms of blue channel, red channel and ndvi (gray).

​Outdoor test with unsatisfactory results.

What am I missing?

[Liz Barry]

Hi Bernado, 

I'm bumping this thread back up in case someone can answer your questions. Also, i'm coping the Public Lab Portugues list so that other Brasil-based members can meet you! Please consider joining this list as well, you can find it at the top of publiclab.org/lists

Yours, 

Liz

--

Liz Barry
director of community development
@publiclab

publiclab.org

+1-336-269-1539

Love our work? Become a Public Lab Sustaining Member today!

--
Post to this group at plots-i...@googlegroups.com
 
Public Lab mailing lists (http://publiclab.org/lists) are great for discussion, but to get attribution, open source your work, and make it easy for others to find and cite your contributions, please publish your work at http://publiclab.org
---
You received this message because you are subscribed to the Google Groups "plots-infrared" group.
To unsubscribe from this group and stop receiving emails from it, send an email to plots-infrare...@googlegroups.com.

[Chris Fastie]

Bernardo,

I have not used a camera like yours or the control program you have. If I had your setup:

1. I would not bother with indoor photos. The setup probably can't tell you much about plant health in a room with dim artificial lighting.
   
2. I would not expect good indoor results with the gain and the exposure set to zero.
3. To get useful NDVI results, I would experiment with "White Balance Temperature." If you actually have control over that, you could get good results.
4. I would use your control of the camera to get a well exposed photo of plants outside. The outdoor photo you show is too overexposed to produce meaningful NDVI.
5. I would first try auto exposure, then maybe use manual exposure to improve the photo.
6. If you really have control of all those things on the control panel, you should be able to get much better results.
7. If that control panel does not actually control the camera, I can't help with that.

Chris

[Ives Rocha]

Oi pessoal,

Alguém aqui no Brasil usa este kit e pode auxiliar o Bernardo?

Abs,

Ives

--
Publicado neste grupo através do publiclab...@googlegroups.com
 
Listas de discussão do Public Lab (http://publiclab.org/lists) são ótimas para a troca de ideias mas, para obter autoria, tornar seu trabalho aberto e suas contribuições mais fáceis de encontrar e citar, por favor, publique o seu trabalho em http://publiclab.org
--
Post to this group at publiclab...@googlegroups.com

 
Public Lab mailing lists (http://publiclab.org/lists) are great for discussion, but to get attribution, open source your work, and make it easy for others to find and cite your contributions, please publish your work at http://publiclab.org
---

Você recebeu essa mensagem porque está inscrito no grupo "Public Lab - Português" dos Grupos do Google.
Para cancelar inscrição nesse grupo e parar de receber e-mails dele, envie um e-mail para publiclab-portu...@googlegroups.com.
Para mais opções, acesse https://groups.google.com/d/optout.

[bernardo araujo rodrigues]

Thanks a lot for the help guys, specially Ivan and Chris!

I think I'll just stick to the auto settings.