Using drone mapping software to make a map of a 1,300+ acre park


Recommended Posts

Hi all,

CAVU Media created a 1,300+ acre map of New Orleans City Park. They were able to achieve higher resolution on the imagery w/ drone mapping software compared to traditional aerial photography. I would love to know your thoughts on this study as I've gotten some interesting feedback and challenges from posting this article.

Link to post
Share on other sites

Interesting project. The 1300 acres is a little daunting, especially with a DJI product. The team was lucky to be able to operate under the Class B airspace and that the Class C and D airspace to the west cut off where it does. I only wish that I was so lucky. They also got lucky in that they were able to fly at max altitude (400 ft AGL) instead of being limited by the surrounding airspace that is so common in urban areas.

I have found that higher resolution sells vs manned flights. The stock sets we were having flown before we purchased our own UAS was running in the $12-14,000 range per set. I can fly one set with my Inspire and it pays for the total cost of our program to date and then some change left over. We control what we get and we are able to update the standard footage that the local COG flies every year on an as-needed basis whenever the base map changes. We provide higher resolution imagery. We can update the area on our own timeframe.

The cons are crowd control that manned aircraft does not have to deal with. Everything we fly has to be coordinated in very fine detail with administration, maintenance, campus PD, local PD in some instance, ect... far in advance. Time required to fly the set. A manned flight can do this in a day or less. With an Inspire, to fly an entire package would require many months and Mother Nature taking a serious vacation. And then, of course, there is the processing of all the imagery. Granted, after setting up the GCPs its just the punch of the enter key and coming back in a few days maybe?....never tried something that larger in scope.

I would really like to do something on this scale myself with a commercial UAS that has more legs than the prosumer products. 

Link to post
Share on other sites

Hi Kara,

the article is fine.  This is a very difficult project and understanding the logistics is important.  But this article mostly reads like propaganda for DroneDeploy and DJI.  

What is not mentioned is how much it would cost to do this sort of mission.  How long did it take to process 6000 images?  How much does that processing cost?  How accurate it the DEM?

I know everyone wants to believe you can buy a $1400 drone, use a cell phone app and produce an accurate DEM/3D model, but there's no free lunch here.  To develop an accurate model would require all the resolution that tiny P4Pro camera could muster and if that many images are not compressed then it would probably take more than three weeks to render.  

Also in the article the amount of coverage seemed to be gated by the amount of flight time, hence the plug for the Phantom4.  I'm not sure DroneDeploy wants to encourage operators to be flying beyond line of sight?  Not saying they did but there was no mention of how they accomplished that?

 

Link to post
Share on other sites

Hi Chuck, this was a case study and direct account from CAVU Media. Ian, the owner, has a lot of high-profile clients (Red Bull, PBS). Given his reputation, I wouldn't think he'd be making false claims about the cost, accuracy, or any other factors that went into this project. Granted the guy has decades of experience but to say it's propaganda when Ian is educating people on the possibilities of the software isn't accurate.

Link to post
Share on other sites

Kara, I'm guessing you work for DroneDeploy?  Your reading something into my post that isn't there.

I didn't suggest that he made any false claims, nor did I call into question his reputation. I'm sure he's a great guy.  However there was no mention of GCP's, no reference to accuracy, how long it took to process, or how much would it cost to process this job?

Whether you call it a case study or propaganda its all marketing that suggests that anyone who purchases a P4 and uses DroneDeploy can do this. It matters very little what DroneDeploy says about the positional accuracy of its service. A combination of factors affects the positional accuracy of DEM, or other derivatives of remotely sensed data. Poor operation of the best drone can impair the positional accuracy of a deliverable, accuracy is dependent on a variety of factors.  This is true for all photogrammetry applications.  

Good for him for sharing his story, but he's hardly the only company out there doing this. You said in your reply that he has decades of experience, not in the commercial drone industry, the article pointed out he started in 2013.  Prior to that he was a website designer, not a civil engineer, land surveyor, GIS professional.     

This article fosters the misperception that “anyone can do it” in a forum where people are looking for advice starting a commercial drone business.  Lacking the knowledge of the fundamentals is certain to cause anyone who took this article at face value considerable financial loss and compromises to public safety.    

http://realitymodeling-pw.bentley.com/a3D/Construct Timelapse/models/4-12/App/?scene=http://realitymodeling-pw.bentley.com/a3D/Glendale/Production_1/Scene/Production_1.3mx#%2F

This scan was done for the DHS, the VP gave a speech at a park in downtown Glendale and they needed a very accurate 3D model for one square mile around the park.  The accuracy in that area is very good, the further you go from the center the less accurate it becomes.  This was shot with a Canon 5DS (50MP) camera on a 3-axis gimbal.  There is no "off the shelf" drone we could have used to accomplish this.  I can't tell you how many shots we had, but it was way less than 6000 and it took a week to render.     

To say that Ian is trying to "educate" people about the accuracy of DroneDeploy is a bit disingenuous.

Link to post
Share on other sites

I'm mostly curious on how he dealt with the line-of-sight issue. To me, this is the biggest hurdle to tackling a large area like this. The article mentions that it's a 1 mile wide area, and it's heavily wooded, which adds more complication to line-of-sight.

Link to post
Share on other sites
12 hours ago, SactoSteve said:

I'm mostly curious on how he dealt with the line-of-sight issue. To me, this is the biggest hurdle to tackling a large area like this. The article mentions that it's a 1 mile wide area, and it's heavily wooded, which adds more complication to line-of-sight.

I would imagine a team of V/Os.

Link to post
Share on other sites
  • 2 months later...

Now that it is possible to fly the P4 with RTK onboard the need for GCP's is reduced. I have just completed a mission where the RMSE is 0.35m from the photos with no GCPs and agreement with ground verification points is less than 30cm in all cases.

NO GCP processing, just carry the sensor and run a few scripts prior to your normal processing workflow.

File 21-8-17, 09 35 10_v1.jpg

Link to post
Share on other sites
13 hours ago, Spatial Analytics said:

Now that it is possible to fly the P4 with RTK onboard the need for GCP's is reduced. I have just completed a mission where the RMSE is 0.35m from the photos with no GCPs and agreement with ground verification points is less than 30cm in all cases.

NO GCP processing, just carry the sensor and run a few scripts prior to your normal processing workflow.

13.78" may not require GCPs if you are doing a photoshoot but for mapping or survey purposes, that is abysmal. 30cm or 11.8" is not that much better. If you truly want the ability to fly without GCPs, you are going to need to spend ~$23K + for a UAS capable of meeting the requirements. The P4 is not it.

Link to post
Share on other sites

Here in Australia we have just had a change to the commercial drone laws. Sub 2Kg (4.4lb) drones need company registration, but no organisational or personal pilot training. This means the sub 2KG class will bloom and many tasks which were in the realm of the registered, certified and trained drone operating company will be undertaken in house by the engineers, farmers, realtors.  I am working with a number of these to develop solutions that fit their positioning requirements. In these cases rather than mandate over precise (and therefore expensive) positioning requirements I work backwards from their end product.  In many cases 30cm really delivers and allows a business to afford to fly daily or weekly missions to monitor change.

Link to post
Share on other sites

Organizations that make money through the use of certified data won't have an issue with operator certification.  They will either build a department with certified operators or they will out source to organizations who can provide the certified operators.   

If they have no need for accuracy what's the value of the data?  Why do they need you?

Sure there are lots of applications that don't require 2cm accuracy but there are few customers willing to pay for data with less accuracy.  

 

Link to post
Share on other sites
  • 3 weeks later...

Kara is active on other forums and usually posts information that depicts DroneDeply in favorable light as most of her references are direct links back to DD webpages. Often her actual understanding of the content that is posted is less than expert. While I have often clicked on her links and have learned much information from her posting of them, her practice is usually to offer several different webpage links across several subject topics. The one thing they almost all have in common is that they all refer to a DD generated discussion.

To hold her accountable for indepth knowledge on so many topics that she posts about is unrealistic.

Just my 2 cents and might not even be worth that, it is only my opinion based on observations, no intent of disrespect.

Link to post
Share on other sites

I suspect you're right.  No one has a problem with her posting links to articles about commercial workflows etc., likewise, I don't think anyone is trying to hold her to account for the in-depth knowledge.  It's helpful to know if people have an affiliation with the organization, in this case, DroneDeply, so we can decide for ourselves the value of the link.

Increasingly we're living in a world where everyone knows how to do everything but hardly anyone has actually done anything. 

Link to post
Share on other sites

Archived

This topic is now archived and is closed to further replies.