Video Flight Following (FMV)

I would like to see the ability to upload a pdf map of an area and use it as the base map while flying and recording video. The ability to mark annotations on the base map either while in flight or from the desktop after uploading video. This would allow for faster reaction to events and the ability to send ground truthing crews to check the annotations created on the base map.

Could you expand upon what the end goal is for what you are doing?

What generates the PDF map and what is on it?

Why not map it? You can use Live Map to get a pretty good map in 30 minutes…

I think most of this is accomplishable, but details would help us figure out the best direction.

We use drones in planning and during the harvest stage of our forestry operations. the flights pre harvest are used to verify creeks and riparian areas and to look for points of interest. We collect data from other land users and it is sometimes just a stakeholder pointing to an area on a map where they think their cabin is or some other feature that they would like us to protect during harvest. this would allow us to do quick 5min flight to verify location of these features rather than spending an hour or more mapping these areas to analyze later. The photos that we have to use for planning our harvest blocks is sometimes over a decade old and a lot of things can change in that time, beavers may have flooded portions of the land or high waters may have diverted a stream through a new location. the quick flight with video would trim a substantial amount of time off our assessments when compared to mapping these areas which are at times in excess of 500ha.
We also fly to assess harvest areas during harvest to ensure all merchantable products are collected. doing a quick video flight allows us to point out to equipment operators where missed product is that may have been missed by the night shift.
This would also allow us to perform pre-construction flights of planned roads to identify and inspect locations that should be diverted around. An example would be creek crossings locations that may have unstable banks making the risk to installing crossing at planned locations a sediment issue, when there may be a more suitable crossing only a few hundred meters away.
Being able to fly and annotate these areas on a map that can be used by field staff to ground truth would be a huge time saver. This would allow for more environmentally sound planning and aid in reporting pre construction and post construction findings to the government.

This may sound a bit like I am rambling but hopefully you see what I would like to achieve if you need any more clarification let me know.

Not rambling at all, that was a great summation. Just in my opinion these are all things that could be accomplished with much more concrete identifiable, locatable and quantifiable information on a map. You don’t have to plan the entire area just create smaller plans for the best-guess areas. If you use Live Map you will quickly get an orthomosaic on the screen that can then be used to fly to those exact points for further video documentation that can then also be uploaded to the map. You can also pause a mission when you come across features that you want to video and/or take manual aerials of and then continue it when you are done.

Oh, and just FYI that 500ha can be setup to fly in 2-2.5 hours with a Phantom 4 Pro. Caveats being that it would take 9 batteries and 2800 images.

yeah I understand your point and the maps are great they are just not the best option for all our circumstances. I did a lot of mapping and field verifications when I was a field supervisor and the maps are helpful and are great when you have the time to spend flying and producing them. The way most of our drone data is collected right now is through video with planners making estimates of where things are located. The ability to draw these annotations with some relative accuracy while flying or back in the office while reviewing video would be a game changer for forestry.

We have looked at other software platforms (ESRI) and tried to get our drones to work with their software and discovered that the DJI outputs were missing a few pieces of data to make it run smoothly. I brought the idea to DD and they said put it up here and see if there is enough interest for them to take a serious look at it. If it is not something that can be done here then my only other options is to look at a different drone with appropriate outputs to use in the ESRI software and make the video captured data give me what I want.

Yeah, I have made some large maps in the past pushing my DJI drones to the limits and packing tons of batteries. Always made for fun days. I am guessing that you get a bit more life out of a battery than i do up here in the winter.

1 Like

Another option might be 360 panos… They can use the inspection process for more detailed information. Those, aerials and video can all be a “map” without the actual orthomosaic as a background - just the Google Earth. You can also make a “map” out of a raster image in QGIS or I would assume ESRI by georeferencing it and then importing into DroneDeploy as a base map. The one thing that DroneDeploy really needs to get on is the sharing aspect of videos. Markups of frames would be a nice option as well.

Just did some more calculations and you can fly half of a crosshatch mission which provides obliques and it would take just over an hour for 500ha. 1500 images and 5-6 batteries.