I want to have a discussion here related to various methods of collecting NIR imagery in support of generating Vegetation Index data such as NDVI, specifically with regards to using OTS digital cameras or modified digital cameras.

In general there are two approaches:

  1. Single Camera in which one of the 3 bands captures NIR instead of visible light
  2. Dual Camera One camera captures R,G,B and another captures only NIR

NDVI requires a visible band (usually Red or in the case of single camera NDVI, Blue) and an NIR band.

I've been working on a single camera NDVI mapping solution using canon P&S. Testing, testing, testing, and posting results at  www.flightriot.com

Please offer any guidance, suggestions, experience, or questions related to using NIR imagery in UAV aerial mapping. 

I look forward to the discussion.

Views: 7465

Attachments:

Reply to This

Replies to This Discussion

Hey Brenden,
Looks like we've got pretty similar setups. I started flying my skywalker with two cameras so I could create a four-band image. Shadows are pretty much the enemy in my mind, so I fly both cameras at the same time to keep the images as consistent as possible. I made a couple pods out of foam core board and mounted them to the fuselage on some metal rods that I installed just below the wing mount.

Great site by the way!
-Colni

Colin,

I'd love to see your 2-cam set up. It sounds very effective. I have not yet flown with 2 cameras but I intend to so any insight is greatly appreciated.

Do you have any video or images of your set up you can share?

Shadows have been a challenge in my filter tests too as they can sometimes show up as very high or very low NDVI values. On your NIR camera, what filter are you using and is it mounted internally or externally? Also, what cameras are you using? I ask because if you are interested, I'd be happy to send you some filters to test the single camera NDVI method.

I have tested a bunch of filters on 3 cameras and I will continue to test...think the more testing and input the better. I'd only ask that you share your results so others can benefit. 

One discovery along the way is that CMOS sensors seem to be more of a challenge to use than CCD sensors. I was planning to use a canon sx230 as my main NDVI camera, only to find out that cameras such as the A490 and A810 seem to be much easier to get good results with. 

Do you think a modified version of your approach to mounting the cameras may work for collecting oblique imagery? At some point I want to try running two cameras simultaneously to capture orthogonal and oblique images for generating better 3D models.

Also, thank you for the positive feedback. I really scaled back the site to focus on UAV mapping as it seems to be the only one devoted to the topic. The goal right now is to develop a "complete" set of UAV mapping resources.

Brenden

Hi Brenden, 

I am doing the samething in Beijing, China as well. I am a newbie in the area and I hope to keep in touch with you guys. Now, I am planning to carry out a commercial strategy. And sell those images and services to government and farmers. 

Best wishes,

Kun

This page, this page, and this page may be useful as references for this discussion.

Thank you John,

It seems a lot stuff to study..One more concern, to monitor vegetation's condition and stress, (get accurate  NDVI pics) do we actually need a professional camera? Or the modified camera can do this job perfectly?

Many thanks,

Kun

Unfortunately, it is a lot of stuff to study. Searching is even more time-consuming on diydrones.com (DIYD). This page may help you with searching, including the tip at the bottom of the page.

Because of limited budgets, and an interest in do-it-yourself (DIY), many members/visitors to DIYD keep looking for less expensive but good enough products, or products that can be hacked or modified. A recent example is this page. And an interesting variety of opinions on this page.

I don't have expertise in NIR or NDVI topics, including specific camera products or hacks of products. However, while cataloging the DIYD blog posts (starting January 2013), I did notice a few that may be useful to you, even though they may not answer your immediate questions.

I have often found that the most useful info for me was in the comments, rather than in the initial post/discussion topic.

http://diydrones.com/profiles/blogs/infragram-a-kickstarter-project...

http://diydrones.com/profiles/blogs/ir-and-eoi-brushless-gimbal

http://diydrones.com/profiles/blogs/raspberry-pi-noir-and-public-la...

http://www.diydrones.com/profiles/blogs/results-of-two-years-of-work

http://diydrones.com/profiles/blogs/weed-mapping-in-maize-fields-us...

http://diydrones.com/profiles/blogs/100-off-agisoft-photoscan-pro-u...

http://www.diydrones.com/profiles/blogs/geo-referenced-tetra-cam-ad...

http://diydrones.com/profiles/blogs/visible-rgb-and-full-spectrum-r...

http://diydrones.com/profiles/blogs/remote-sensing-with-uav-for-bio...

Best wishes on your project!

Hey! Thank you for your website, I love your videos!!!

Hi John,

Many thanks for your websites and I much appreciate your help so much! I am gidding it in much more and found this website. I hope it could help.

Kun 

Hi Kun,

You are most welcome. More stuff in this blog post and this blog post. Read the comments on this blog post from 2012. This product is an alternative to the one you found. Here is one more link to look at. You may find more info by searching DIYD for 2011-2012 blog posts and discussions. Hope you are resting your eyes between all that reading. :-)

Hi John,
I am glad to receive your reply and thank you so much~ It really helps mine study!! I will post my least study and share it with you~
P.S. Your are right, recent weeks, I feel my eyes are not good! Lol...
Cheers,
Kun

This 2014 discussion may be one of the most useful on DIYD for your interests.

Hi John,

Brilliant!~ this is exactly what I mean and want!! 

Kun

Reply to Discussion

RSS

© 2019   Created by Chris Anderson.   Powered by

Badges  |  Report an Issue  |  Terms of Service