GOAT logo

Join the conversation! The forum activity is now at GOATeach.org!  We are working to cross pollinate our conversations. Document and share tools at farm hack and talk at GOAT!  Also join GOAT riot and introduce yourself and your projects!

Posted 3-D image of plot on tool page

Topic Type: 
Information

I just posted a link to the 3D image of the plot site on the tool forum http://www.farmhack.net/tools/covercrop-remote-sensing-imaging#forum

Chris Fastie's picture

I selected 149 of the photos from the regular camera on the dual IR/visible rig balloon flights over Dorn's study plots and submitted them to Hypr3D. It used only 49 of the photos and the model covered fewer than half of the study plots. So I borrowed 40 of the photos from Jeff's higher elevation flight and added them for a total of 189 photos. I downsampled all the photos to 1200 pixels across first. The resulting model used 150 photos and covered an area similar to that of Dorn's first model. The texture mapped over the 3D model seems to include some high and some low resolution images. The modeled topography definitely reflects the relief of the mown paths, and the house and yard trees look pretty cool.

Is it time to try this in 123D Catch?

Dorn's picture

I will download this into meshlab - that seems to provide a little higher resolution too. It would be interesting to compare to 123D Catch output.

Chris Fastie's picture

Dorn, did you get my latest model to display in Meshlab? I tried it for the first time, and the gray model displays but I can't figure out how to make it display the texture map.

Dorn's picture

I had the same issue. Mashlab seems to indicate that is is loading texture, but doesn't ever complete. I was wondering if I needed a little more computing horsepower to get it to display...

Chris Fastie's picture

That's good to know. I think something is weird with the texture jpeg. I will run the Hypr3D process again. The Hypr server took more than 5 hours to complete the processing the first time. Before I submit it again I might sort through the photos and maybe lighten or darken some so they match better. I should put the images at Dropbox now that there is more room there. I will let everyone know when they are there.

jywarren's picture

Dorn - Chris posted a bit more and I was speculating about how we might calculate biomass from the hi-res Meshlab data. I thought you might have some input:

http://publiclaboratory.org/notes/cfastie/5-31-2012/3d-model-meshlab

Also, I was wondering if you'd done any ground-based biomass measurements, or how they might be done (barring, i suppose, cutting down all your veg and weighing it... :-P )

Dorn's picture

We will actually be cutting two one meter samples in each plot, sorting the biomass,drying it and then weighing it. Which is why if we could find a remote way to calculate these values, we could save many many hours of labor.

I think that we might be able to get some rough information on biomass based on growth height which could probably be extracted based on the known scale of the plots and the heights of the plots above the pathways which is a known ground level. It would be interesting to see what kind of resolution we do have.

We will be doing the ground based measurements within the next two weeks, so we could correlate the on the ground data anytime after that is entered.

Albert Rogers's picture

Dorn, did you get my latest model to display in Meshlab? I tried it for the first time, and the gray model displays but I can't figure out how to make it display the texture map.

Sunny Slope Orchard's picture

Albert: I am Bill Spurlock, farm hack user Sunny Slope Orchard. I happened to notice that when I click on your name in this forum MY user profile came up, even though the URL says "http://farmhack.net/users/albert-rogers"

Somehow your profile got replaced by mine.