Hello everyone, just received my level 1sUAS certification, have my part 107/and small drone business. I'm looking to add the 640T to my fleet, from what I see after downloading autel IR tool, it appears similar to flir tools, i.e. your able to add emissivity; etc. Has anyone out there utilized autel's tool yet for making reports? anyone using the 640T yet?
Hello everyone, just received my level 1sUAS certification, have my part 107/and small drone business. I'm looking to add the 640T to my fleet, from what I see after downloading autel IR tool, it appears similar to flir tools, i.e. your able to add emissivity; etc. Has anyone out there utilized autel's tool yet for making reports? anyone using the 640T yet?
Empire Drones stated they are working on a review of the sensor and software suite that should be available soon. Hopefully they update us here, but keep an eye on their YouTube page
Autel has just released the Evo 2 Dual 640R with a radiometric Flir Boson in 640 resolution. This will work in the more mature Flir Tools. Price $10,500 and should be FTC Made in the USA so it's grant friendly. I now have a bit of buyer's remorse with my Mavic Enterprise Advanced. But Autel still does not offer an Enterprise Shield type protection plan so I would not purchase it without insurance.
Hello all. I just recieved the 640R V1. Please see the below link for an explanation of what I have found with it so far. More details and photos to come as I practice with it more. Mahalo! Have a great weekend. Annnddd...congrats @JDL on your certification! Did you do ITC?
I have now saw the Evo 2 Dual 640R for sale on a few vendor's sites. 640 resolution Flir Boson radiometric sensor. Best news $500 more than the regular Dual. The Boson had an extremely long journey to becoming radiometric but it's here and this sensor will mature into a fine product. Flir also...
How do you like the flight performance? I just had a demo ...it looks really shaky in the wind ....and landing it took a few bunny hops for it to land (yes, I had the throttle down all the way) -- the rep said that was normal ...the auto land system isnt working right ....anyone else experience that? Also the Autel tools like a bit better then the DJI tools ...deciding whether going to the R for FLIR tools is a good reason or not ...
I have not experienced any of those issues. My feels rock solid flying, never had it bunny hop landing either. The Autel software for analysis is nice. The T is just less expensive, but the side by side comparison shows they T and R version are too close for me to spend the extra money.
Nice. What are you using for map processing software? Are you using a third party for flying the mission or mapping tool native to Autel? I can even see several possible Anomalies in the school building Ortho bottom right structure PV array. Great job.
Nice. What are you using for map processing software? Are you using a third party for flying the mission or mapping tool native to Autel? I can even see several possible Anomalies in the school building Ortho bottom right structure PV array. Great job.
Does Autel allow you to capture RBG and Thermal image at same time while flying mission? Most of my Clients ask for that so currently using Flir XT2 and works nice, but the price and quality of the Autel is definitely something I am going to keep an eye on. What are your mission settings. Front/side lap around 80% and somewhere about 140 ft above target? Also, what are the file formats that Autel can take in? rJpeg, Tiff, Jpeg??
For processing of a non-radiometric orthomap try Reality Capture. It does not need the high overlap that Pix4D does and with its pay as you play purchasing its costs are very low due to the low resolution thermal images (As compared to 20MP images from a Phantom 4 Pro).
Here is my limited experience with different software and hardware.
Drone Deploy
Autel Evo 640T: Not radiometric at this time. Processes as an RGB image, but the details are extremely good! This is from a member on here and Autel Pilots that has been very generous with sharing his in depth testing of the 640T as well as his RTK testing of the Evo 2 RTK aircraft.
The M2EA will process in Drone Deploy and the output will be radiometric. The image set I ran though had inaccurate temperatures. The detail of the ortho was not that great. I am sure Drone Deploy being the innovator that it is will address and fix this. I think this is more of a camera calibration issue, especially since the M2EA's images are distorted.
Mavic 2 Enterprise Advanced low overlap images. This ortho is radiometric
Pix4D needs high overlap to process correctly. At this time you will have to process as an RGB ortho.
BUT, there is software now available to convert the DJI proprietary RJPEG to radiometric TIFF and you can then use this in Pix4D to create a true radiometric ortho. Here is a link to a thread on the sister site Mavic Pilots where a member shares his Radiometric output in Pix4D. This conversion software is pricey though.
I finally had time to go out and obtain the images to make a thermal map with the Enterprise Advanced. Pix4D wanted 90 front and 90 side overlap per the directions and I found this to make the aircraft fly at only 2-3 mph. I do think that DJI Pilot does not have the correct information about...
mavicpilots.com
Pix4D Output in ArcMap
Reality Capture
In my testing this was a big welcome surprise. Low overlap images that would not process in Pix4D, processed fine in Reality Capture. This is non-radiometric. The pay when you export pricing works great in that you only pay to download the outputs when happy with the results. On such low resolution images the pricing is pennies for small areas.
Reality Capture high Overlap Mavic 2 Enterprise Advanced
Reality Capture images that did not have enough overlap to process in Pix4D of a solar panel array. With 90/90 overlap I think the results would of been better.
This is from a member on here and Autel Pilots that has been very generous with sharing his in depth testing of the 640T as well as his RTK testing of the Evo 2 RTK aircraft.
For processing of a non-radiometric orthomap try Reality Capture. It does not need the high overlap that Pix4D does and with its pay as you play purchasing its costs are very low due to the low resolution thermal images (As compared to 20MP images from a Phantom 4 Pro).
Here is my limited experience with different software and hardware.
Drone Deploy
Autel Evo 640T: Not radiometric at this time. Processes as an RGB image, but the details are extremely good! This is from a member on here and Autel Pilots that has been very generous with sharing his in depth testing of the 640T as well as his RTK testing of the Evo 2 RTK aircraft.
The M2EA will process in Drone Deploy and the output will be radiometric. The image set I ran though had inaccurate temperatures. The detail of the ortho was not that great. I am sure Drone Deploy being the innovator that it is will address and fix this. I think this is more of a camera calibration issue, especially since the M2EA's images are distorted.
Mavic 2 Enterprise Advanced low overlap images. This ortho is radiometric
Pix4D needs high overlap to process correctly. At this time you will have to process as an RGB ortho.
BUT, there is software now available to convert the DJI proprietary RJPEG to radiometric TIFF and you can then use this in Pix4D to create a true radiometric ortho. Here is a link to a thread on the sister site Mavic Pilots where a member shares his Radiometric output in Pix4D. This conversion software is pricey though.
I finally had time to go out and obtain the images to make a thermal map with the Enterprise Advanced. Pix4D wanted 90 front and 90 side overlap per the directions and I found this to make the aircraft fly at only 2-3 mph. I do think that DJI Pilot does not have the correct information about...
In my testing this was a big welcome surprise. Low overlap images that would not process in Pix4D, processed fine in Reality Capture. This is non-radiometric. The pay when you export pricing works great in that you only pay to download the outputs when happy with the results. On such low resolution images the pricing is pennies for small areas.
Reality Capture high Overlap Mavic 2 Enterprise Advanced
Reality Capture images that did not have enough overlap to process in Pix4D of a solar panel array. With 90/90 overlap I think the results would of been better.
I had some time to analyze Drone Deploy's outputs and the results as far as radiometric temps is much better for this example. A major problem is that Drone Deploy's inverted teardrop makes it hard to position it on small targets. I used DJI Thermal Analysis Tool for this comparison.
Solar Panel Array Anomaly
Drone Deploy
DJI Thermal Analysis Tool
So 113*F vs 123 *F I left the parameters in DJI TAT at default.
Next Example is the hot Pavement which is R2 in the above image.
126*F vs 124*F, so it is much closer.
I would really like to obtain that DJI RJPEG to radiometric TIFF app, as I am curious as to how Pix4D would fare with a true radiometric ortho in this comparison.
Another note that is disappointing is that DJI seems to be watering down the capabilities of the M2EA's thermal as there are less adjustable parameters in DJI TAT for the M2EA than for the H20T sensor in the Matrice 300 RTK. I understand that the flagship should have the best options, but neutering capabilities seems counterproductive when Autels small 640 thermal package seems to be a better choice more and more.
I think if Autel added an Enterprise Care insurance and software to make their thermal images into radiometric geotifs, they would have a handsdown clear small package 640 thermal winner. As it stands though, it is one heck of a package and I would have an Enterprise 640T package with RTK module if they offered Enterprise Care to protect my expensive investment. The M2EA won out for my personal use due to me needing the 2 year crash protection.
For processing of a non-radiometric orthomap try Reality Capture. It does not need the high overlap that Pix4D does and with its pay as you play purchasing its costs are very low due to the low resolution thermal images (As compared to 20MP images from a Phantom 4 Pro).
Here is my limited experience with different software and hardware.
Drone Deploy
Autel Evo 640T: Not radiometric at this time. Processes as an RGB image, but the details are extremely good! This is from a member on here and Autel Pilots that has been very generous with sharing his in depth testing of the 640T as well as his RTK testing of the Evo 2 RTK aircraft.
The M2EA will process in Drone Deploy and the output will be radiometric. The image set I ran though had inaccurate temperatures. The detail of the ortho was not that great. I am sure Drone Deploy being the innovator that it is will address and fix this. I think this is more of a camera calibration issue, especially since the M2EA's images are distorted.
Mavic 2 Enterprise Advanced low overlap images. This ortho is radiometric
Pix4D needs high overlap to process correctly. At this time you will have to process as an RGB ortho.
BUT, there is software now available to convert the DJI proprietary RJPEG to radiometric TIFF and you can then use this in Pix4D to create a true radiometric ortho. Here is a link to a thread on the sister site Mavic Pilots where a member shares his Radiometric output in Pix4D. This conversion software is pricey though.
I finally had time to go out and obtain the images to make a thermal map with the Enterprise Advanced. Pix4D wanted 90 front and 90 side overlap per the directions and I found this to make the aircraft fly at only 2-3 mph. I do think that DJI Pilot does not have the correct information about...
In my testing this was a big welcome surprise. Low overlap images that would not process in Pix4D, processed fine in Reality Capture. This is non-radiometric. The pay when you export pricing works great in that you only pay to download the outputs when happy with the results. On such low resolution images the pricing is pennies for small areas.
Reality Capture high Overlap Mavic 2 Enterprise Advanced
Reality Capture images that did not have enough overlap to process in Pix4D of a solar panel array. With 90/90 overlap I think the results would of been better.