completely out of my depth here please help

[Jeremy Cullen]

Hi so i have recently been talking with a professor in the uk who has been developing a paper about using a drone to measure biomass in pasture using a drone and photogrammetry software. now i am just a dairy farmer but im not too shabby with a computer but so far im stumped i know i need to do the following.

First one needs to process the photographs via something called 'structure-from-motion photogrammetry' (with software like Agisoft Metashape or Pix4D, other free alternatives that are a bit less user friendly), in order to generate 3D point clouds.

Then one needs to analyse the point cloud, to quantify plant canopy height across the area of interest

please someone help me out or direct me to a tutorial on how to use this software

[Robert Smith]

Hello Jeromy,

Actually you may be better off using a open-source software called 'Paraview' by Kitware and used by millions in the scientific community. It is fairly simple to use with amazing use. Many national labs and dept.s of government also use it so you should be able to handle it well.

If you don't mind my asking, what in particular are you trying to research, and do you have knowledge of the spectrometry useful to regular RBG images for analysis of plant matter? I may be able to advise on this.

Good Luck,

Bob Smith

---

From: icy-so...@googlegroups.com <icy-so...@googlegroups.com> on behalf of Jeremy Cullen <jezcu...@gmail.com>
Sent: Wednesday, September 2, 2020 7:55 PM
To: Icy imaging <icy-so...@googlegroups.com>
Subject: [icy] completely out of my depth here please help

 

--
You received this message because you are subscribed to the Google Groups "Icy imaging" group.
To unsubscribe from this group and stop receiving emails from it, send an email to icy-software...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/icy-software/a6f09c2c-3158-486c-9f55-16a3af4dcd74n%40googlegroups.com.

[Jeremy Cullen]

Hi thanks for the advice. I have no experience whatsoever this is all new to me. Basically I spend about 4 hours every week walking over my whole farm measiung how much grass I have. So I'm trying to develop a way to simply have my drone fly over the farm and fo it for me. I know i will need to continue doing this to correlate the readings in the paddock to what the computer analysis comes up with but I do have access to 3 other farms so potentially 300 plus reading every week. I would like to long term turn this into something that I could develop and potentially release to the market as the closest there is at the moment is something called LIC space which works under the same principles but is very unreliable as it relies on images taken by a satellite and if there is cloud cover when it flies over you don't get any readings. 

To view this discussion on the web visit https://groups.google.com/d/msgid/icy-software/CH2PR19MB371996822BA593FEE38F9E1ABD2C0%40CH2PR19MB3719.namprd19.prod.outlook.com.

[Robert Smith]

Hello again Jeromy,

Only grass?  

Meantime, the FAA requires a specific license if the drone will be more than 400 ft. in altitude and the altitude you will be using is also important to how much data you can collect.

Are you using a typical RGB camera or one fitted with Infra Red capability? Either one works well, but IR just allows more bandwidth to the data.

ICY as well as ImageJ, which is a part of ICY so you already have it, are geared to the imaging and data aquisition, then the Paraview software would be used in the analysis of the data and visualization of the data.

And what charactoristics  are you interested in obtaining. There are hundreds of things to look for, but you don't want it to be overly large. 

Get the Paraview software, and with the ICY software you will have everything you would need, as much as NASA and satellite images. As for the camera, again I need to know the image size and altitude you will be using as well as parameters such as time of day, direction of the flight path and sky condition.

It would be great if you could supply an example image with the equipment you will be using. 

It is a very worthwhile project and is greatly needed all over the world, yet low level imaging with useful information is lacking. I will assist you as to how to look for what you want.

Type with you soon, 

Bob 

---

From: icy-so...@googlegroups.com <icy-so...@googlegroups.com> on behalf of Jeremy Cullen <jezcu...@gmail.com>

Sent: Thursday, September 3, 2020 1:48 PM
To: icy-so...@googlegroups.com <icy-so...@googlegroups.com>
Subject: Re: [icy] completely out of my depth here please help

 

Hi thanks for the advice. I have no experience whatsoever this is all new to me. Basically I spend about 4 hours every week walking over my whole farm measiung how much grass I have. So I'm trying to develop a way to simply have my drone fly over the farm and fo it for me. I know i will need to continue doing this to correlate the readings in the paddock to what the computer analysis comes up with but I do have access to 3 other farms so potentially 300 plus reading every week. I would like to long term turn this into something that I could develop and potentially release to the market as the closest there is at the moment is something called LIC space which works under the same principles but is very unreliable as it relies on images taken by a satellite and if there is cloud cover when it flies over you don't get any readings. 

On Fri, 4 Sep 2020, 3:56 am Robert Smith, <rsmit...@live.com> wrote:

Hello Jeromy,

Actually you may be better off using a open-source software called 'Paraview' by Kitware and used by millions in the scientific community. It is fairly simple to use with amazing use. Many national labs and dept.s of government also use it so you should be able to handle it well.

If you don't mind my asking, what in paparticular are you trying to research, and do you have knowledge of the spectrometry useful to regular RBG images for analysis of plant matter? I may be able to advise on this.

Good Luck,

Bob Smith

---

From: icy-so...@googlegroups.com <icy-so...@googlegroups.com> on behalf of Jeremy Cullen <jezcu...@gmail.com>
Sent: Wednesday, September 2, 2020 7:55 PM
To: Icy imaging <icy-so...@googlegroups.com>
Subject: [icy] completely out of my depth here please help

 

Hi so i have recently been talking with a professor in the uk who has been developing a paper about using a drone to measure biomass in pasture using a drone and photogrammetry software. now i am just a dairy farmer but im not too shabby with a computer but so far im stumped i know i need to do the following.

First one needs to process the photographs via something called 'structure-from-motion photogrammetry' (with software like Agisoft Metashape or Pix4D, other free alternatives that are a bit less user friendly), in order to generate 3D point clouds.

Then one needs to analyse the point cloud, to quantify plant canopy height across the area of interest

please someone help me out or direct me to a tutorial on how to use this software

--
You received this message because you are subscribed to the Google Groups "Icy imaging" group.
To unsubscribe from this group and stop receiving emails from it, send an email to icy-software...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/icy-software/a6f09c2c-3158-486c-9f55-16a3af4dcd74n%40googlegroups.com.

--
You received this message because you are subscribed to the Google Groups "Icy imaging" group.
To unsubscribe from this group and stop receiving emails from it, send an email to icy-software...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/icy-software/CH2PR19MB371996822BA593FEE38F9E1ABD2C0%40CH2PR19MB3719.namprd19.prod.outlook.com.

--
You received this message because you are subscribed to the Google Groups "Icy imaging" group.
To unsubscribe from this group and stop receiving emails from it, send an email to icy-software...@googlegroups.com.

To view this discussion on the web visit https://groups.google.com/d/msgid/icy-software/CABkJw9hrt3xVKjoyHkBfV5vtUuJQn2TJJFXeHyGEh5dAo_zDRw%40mail.gmail.com.

[Jeremy Cullen]

 100MEDIA.rar

thanks again bob this information is greatly appreciated. ill be using the stock standard RGB camera that comes with the drone which is a phantom 4 advance. as to the weather conditions i really couldnt say but the drone is rated upto wind speeds of 45 knots but i know people have flown these models in higher wind speeds. and i'm not really to sure on the altitude i wasn't sure how well it would work upto 400 ft but if it did then i would definitely be able to cover our farm relatively quickly as its only 137 hectares.

the ultimate goal here for me is to be able to break the data down into kilograms of dry matter per hectare which is how we measure our grass. A quick example 

paddock 12 reads at 3100KgDM/h the cows will eat that down to 1500KgDM/h which means there is a total of 1600KgDM/ha available for the cows.

i would consider getting a multi spectrum sensor in the future if it works out well. because i know that rbg works with the light refraction which will be different at different times of the day etc. also next time i do a farm walk (which is when i measure my grass) i will fly over the whole farm and capture the images. in the mean time ill send you the photos of just 1 paddock i did as a sample to see if i could do it ( which i couldnt work it out) these were taken at 50m and the pasture reading for that paddock was 2894Kg dry matter per hectare 

To view this discussion on the web visit https://groups.google.com/d/msgid/icy-software/CH2PR19MB3719425FFB3EBFFC8E20219FBD2C0%40CH2PR19MB3719.namprd19.prod.outlook.com.

[Robert Smith]

Jeromy Cullen,

A couple of things to note. Can you obtain these images in a continous video format, not single images?.

Also the type of images you aquire can only be used to determine type of vegetation, health of vegetation, areas, etc. etc.. In order to determine the height of the grass you would need information from a third dimention, such as LIDAR or RADAR, or the ability to triangulate using shadows from the height of known objects. In order to triangulate requires details of time of day, cloud cover, and direction of motion such as from a GPS unit and a consistant altitude reading.

The point cloud approach the professor mentions is only applicable to a surround flight of a particular object if using a conventional camera.

If there is a distinct visual difference in the height of the grass you could make estimates of the relative heights, not to accurate height measurements.

In either case much post-processing of the images is required to optimize the data.

I am sending you some samples of what can be done readily with your images. You decide if that is what you wish to achieve.

Good day,

Bob

---

From: icy-so...@googlegroups.com <icy-so...@googlegroups.com> on behalf of Jeremy Cullen <jezcu...@gmail.com>

Sent: Thursday, September 3, 2020 11:06 PM

To view this discussion on the web visit https://groups.google.com/d/msgid/icy-software/CABkJw9jC3-jrTagdwfygEfjgWesJGkM0C-4CH1r5jHgWfUXD3A%40mail.gmail.com.