Hi,

I would like to ask the Group about so-called calibration of NDVI maps or connection between real numbers of plant growth and received data.

Do you go to the field in order to measure leaf surface area or make any other research (chlorophyll in a leaf) in order to get a ground truth for data, obtained with UAVs?

Thanks.

Views: 3045

Replies to This Discussion

Depends what are you looking for, I take a photo or satellite image and them go to the field to see if what I think is N deficience (for ex), really is, generally ocular inspection now, in the past with a plant juice analizer and, as Jose do, take my tablet with the georreferenced image with me to find the place.

Looks a wheat image before potato with nitrogen deficiency on potato rows, I walk to that rows, verify it and correct deficiences on that red sites.

The red line upper left is rainy damage, so you need to know the field, only image isn't completly usefull.

I suppose it will be different in each case, depending on coverage, type of crop, crop stage, etc. What is important is to first clarify the final goal of doing such thing as going to the field with tools, otherwise you would´t know which tools to use or how.

Bottomline is that trying to ground truth aerial/satellite images in extensive agriculture makes sense if you can then find the way to make valuable decisions, and that´s a big if.

Andrii: What is the way of your question? are you thinking in something for add to petiole with aerial images?

or work with them?

, perhaps we can oriented better if you clarify your idea.

Cala, you are a mind-reader :)

In my project - Petiole, we have a successful work with agriscientists in measuring leaf surface area and level of chlorophyll using Petiole. In simple words, it is an Android-based app to measure these parameters with a mobile phone + cloud app, which makes an analysis of this data.

In one discussion, a person who is doing aerial imagery for agriculture told that he would like to use our Petiole for ground truth and calibration of the NDVI maps. 

Hence, I decided to ask experts (I mean this Community) about this option and does it have any sense from a practical point of view...

Andrii: I'm only give a quick view to your app and imagine ;) ; yes It can be usefull if, for ex your app can measure cover-uncover areas from an aerial photo, that's is an usefull info for agronomist for ex you have a problem like insect attact and you need to decide re seeding or not, with an aerial photo and a measure app you can have a better desicion, that it's possible to do with complex expensive softwares but UAVs are becoming massive and no everyone have access to that softwares.

Looks if it is your idea (sorry my horrible english or, perhaps any misunderstand)

Here you can see that plants are missing and it's going to affect yield but .how much is the damaged area?

here the procesed photo

but not always NDVI is the more usefull index, perhaps software need to allow specialist to choose the image type that He needs. 

Hi Cala were you able to accurately measure non productive areas

( i am supossing that these would be the areas in red?)

Yes, non productive areas are red and you can measure them, sunflower have high relationship with IAF and yield.

Hi Cala, How did you produce the NDVI ? Which drone/camera did you use for the field image ? Thx.

Which Sattellite images do you use for NDVI or crop yield ? Do you compare it to Drone images ?

Cala, thank you so much for a great idea!

I have a little bit different thought how Petiole app can help with NDVI.

So let's imagine a field. It is the same on every image or NDVI map.

But, actually - it is different because there are many factors with impact on aerial imagery: the time of a day, cloud amount, solar weather, shades. So every time we need to have some constant to compare field on today's image and yesterday's image. In other words, it is important to equalize conditions and adjust to a common basis. But it is very difficult and sometimes impossible.

So I suggest that leaf surface area & chlorophyll might be such parameters and serve as a common basis to show changes. You just measure the leaves every time you fly and after comparing NDVIs also compare leaf area and chlorophyll measurements and, consequently, monitor plant changes.

Does it make any sense?

There are too many factors which influence on aerial imagery: the time of a day, cloud amount, solar weather, shades, etc

what kind of readings would protected agriculture have?  

I am referring to protected agriculture by glass, plastic or mesh.  

These are tomatoes being grown in Mesh HH, the mesh would vary by zone and it could be darker, lighter or shinier. 

how accurate would readings be over mesh?

could there be a calibration for such?

RSS

© 2019   Created by Chris Anderson.   Powered by

Badges  |  Report an Issue  |  Terms of Service