Below you will find instructions for some common use cases.
Creating High Quality Orthophotos¶
Without any parameter tweaks, ODM chooses a good compromise between quality, speed and memory usage. If you want to get higher quality results, you need to tweak some parameters:
--orthophoto-resolutionis the resolution of the orthophoto in cm/pixel. Decrease this value for a higher resolution result.
--ignore-gsdis a flag that instructs ODM to skip certain memory and speed optimizations that directly affect the orthophoto. Using this flag will increase runtime and memory usage, but may produce sharper results.
--texturing-nadir-weightshould be increased to
29-32in urban areas to reconstruct better edges of roofs. It should be decreased to
0-6in grassy / flat areas.
--texturing-data-termshould be set to area in forest areas.
--mesh-sizeshould be increased to
--mesh-octree-depthshould be increased to
10-11in urban areas to recreate better buildings / roofs.
Calibrating the Camera¶
Camera calibration is a special challenge with commodity cameras. Temperature changes, vibrations, focus, and other factors can affect the derived parameters with substantial effects on resulting data. Automatic or self calibration is possible and desirable with drone flights, but depending on the flight pattern, automatic calibration may not remove all distortion from the resulting products. James and Robson (2014) in their paper Mitigating systematic error in topographic models derived from UAV and ground‐based image networks address how to minimize the distortion from self-calibration.
Bowling effect on point cloud over 13,000+ image dataset collected by World Bank Tanzania over the flood prone Msimbasi Basin, Dar es Salaam, Tanzania.
To mitigate this effect, there are a few options but the simplest are as follows: fly two patterns separated by 20°, and rather than having a nadir (straight down pointing) camera, use one that tilts forward by 5°.
As this approach to flying can be take longer than typical flights, a pilot or team can fly a small area using the above approach. OpenDroneMap will generate a calibration file called cameras.json that then can be imported to be used to calibrate another flight that is more efficiently flown.
Alternatively, the following experimental method can be applied: fly with much lower overlap, but two crossgrid flights (sometimes called crosshatch) separated by 20° with a 5° forward facing camera.
Crossgrid overlap percentages can be lower than parallel flights. To get good 3D results, you will require 68% overlap and sidelap for an equivalent 83% overlap and sidelap.
To get good 2D and 2.5D (digital elevation model) results, you will require 42% overlap and sidelap for an equivalent 70% overlap and sidelap.
Vertically separated flight lines also improve accuracy, but less so than a camera that is forward facing by 5°.
From James and Robson (2014), CC BY 4.0
Creating Digital Elevation Models¶
By default ODM does not create DEMs. To create a digital terrain model, make sure to pass the
--dtm flag. To create a digital surface model, be sure to pass the
For DTM generation, a Simple Morphological Filter (smrf) is used to classify points in ground vs. non-ground and only the ground points are used. The
smrf filter can be controlled via several parameters:
--smrf-scalarscaling value. Increase this parameter for terrains with lots of height variation.
--smrf-slopeslope parameter, which is a measure of “slope tolerance”. Increase this parameter for terrains with lots of height variation. Should be set to something higher than 0.1 and not higher than 1.2.
--smrf-thresholdelevation threshold. Set this parameter to the minimum height (in meters) that you expect non-ground objects to be.
--smrf-windowwindow radius parameter (in meters) that corresponds to the size of the largest feature (building, trees, etc.) to be removed. Should be set to a value higher than 10.
Changing these options can affect the result of DTMs significantly. The best source to read to understand how the parameters affect the output is to read the original paper An improved simple morphological filter for the terrain classification of airborne LIDAR data (PDF freely available).
--smrf-threshold option has the biggest impact on results.
SMRF is good at avoiding Type I errors (small number of ground points mistakenly classified as non-ground) but only “acceptable” at avoiding Type II errors (large number non-ground points mistakenly classified as ground). This needs to be taken in consideration when generating DTMs that are meant to be used visually, since objects mistaken for ground look like artifacts in the final DTM.
Two other important parameters affect DEM generation:
--dem-resolutionwhich sets the output resolution of the DEM raster (cm/pixel)
--dem-gapfill-stepswhich determines the number of progressive DEM layers to use. For urban scenes increasing this value to 4-5 can help produce better interpolation results in the areas that are left empty by the SMRF filter.
Example of how to generate a DTM:
docker run -ti --rm -v /my/project:/datasets/code <my_odm_image> --project-path /datasets --dtm --dem-resolution 2 --smrf-threshold 0.4 --smrf-window 24
Measuring stockpile volume¶
Weather conditions modify illumination and thus impact the photography results. Best results are obtained with evenly overcast or clear skies. Also look for low wind speeds that allow the camera to remain stable during the data collection process. In order to avoid shadows which on one side of the stockpile can obstruct feature detection and lessen the number of resulting points, always prefer the flights during the midday, when the sun is at the nadir so everything is consistently illuminated. Also ensure that your naked eye horizontal visibility distance is congruent with the planned flight distances for the specific project, so image quality is not adversely impacted by dust, fog, smoke, volcanic ash or pollution.
Most stockpile measurement jobs does not require a crosshatch pattern or angled gimbal as the resting angle of stockpile materials allows the camera to capture the entire stockpile sides. Only some special cases where erosion or machinery operations causes steep angles on the faces of the stockpile would benefit of the crosshatch flight pattern and angled camera gimbal but consider that these additional recognized features come at a cost, (in field labor and processing time) and the resulting improvements are sometimes negligible.
In most of the cases a lawn mower flight pattern is capable of producing highly accurate stockpile models.
Recommended overlap would be between 75% and 80% with a sidelap in the order of 65% to 70%. It is also recommended to slightly increase overlap and sidelap as the flight height is increased.
Flight height can be influenced by different camera models, but in a general way and in order to ensure a balance between image quality and flight optimization, it is recommended to be executed at heights 3 to 4 times the tallest stockpile height. So for a 10 meter stockpile, images can be captured at a height of 40 meters. As the flight height is increased, it is also recommended to increase overlap, so for a 40 meter height flight you can set a 65% sidelap and 75% overlap, but for a planned height of 80 meters a 70% sidelap and 80% overlap allowing features to be recognized and properly processed.
To achieve accuracy levels better than 3%, the use of GCP’s is advised. Typically 5 distributed GCP are sufficient to ensure accurate results. When placing or measuring GCP, equipment accuracy should be greater than the GSD. Survey grade GNSS and total stations are intended to provide the required millimetric accuracy.
For further information on the use of GCPs, please refer to the Ground Control Points section.
A highly accurate model can be achieved using WebODM high resolution predefined settings. Then you can further adjust some parameters as necessary.
If using ODM, these this reference values can help you configure the process settings.
As almost 50% of the material will be found in the first 20% of the stockpile height, special care should be taken in adequately defining the base plane.
In WebODM Dashboard, clic on “view map” to start a 2D view of your project.
Once in the 2D map view, clic on the “Measure volume, area and length” button.
then clic on “Create a new measurement”
Start placing the points to define the stockpile base plane
Clic on “Finish measurement” to finish the process.
Dialog box will show the message “Computing …” for a few seconds, and after the computing is finished the volume measurement value will be displayed.
If you are using the command line OpenDroneMap you can use the dsm files to measure the stockpile volumes using other programs.
Also consider that once the limits of the stockpile are set in software like QGis, you will find there are some ways to determine the base plane. So for isolated stockpiles which boundaries are mostly visible, a linear approach can be used. While for stockpiles set in slopes or in bins, the base plane is better defined by the lowest point. Creation of a triangulated 3D surface to define the base plane is advised for large stockpiles. This is also valid for stockpiles paced on irregular surfaces.
For carefully planned and executed projects, and specially when GSD is less than 1 cm, the expected accuracy should be in the range of 1% to 2%. The resulting accuracy is comparable to the commercially available photogrammetry software and the obtained using survey grade GNSS equipment.
Since many users employ docker to deploy OpenDroneMap, it can be useful to understand some basic commands in order to interrogate the docker instances when things go wrong, or we are curious about what is happening. Docker is a containerized environment intended, among other things, to make it easier to deploy software independent of the local environment. In this way, it is similar to virtual machines.
A few simple commands can make our docker experience much better.
Listing Docker Machines¶
We can start by listing available docker machines on the current machine we are running as follows:
> docker ps CONTAINER ID IMAGE COMMAND CREATED STATUS PORTS NAMES 2518817537ce opendronemap/odm "bash" 36 hours ago Up 36 hours zen_wright 1cdc7fadf688 opendronemap/nodeodm "/usr/bin/nodejs /va…" 37 hours ago Up 37 hours 0.0.0.0:3000->3000/tcp flamboyant_dhawan
If we want to see machines that may not be running but still exist, we can add the -a flag:
> docker ps -a CONTAINER ID IMAGE COMMAND CREATED STATUS PORTS NAMES 2518817537ce opendronemap/odm "bash" 36 hours ago Up 36 hours zen_wright 1cdc7fadf688 opendronemap/nodeodm "/usr/bin/nodejs /va…" 37 hours ago Up 37 hours 0.0.0.0:3000->3000/tcp flamboyant_dhawan cd7b9585b8f6 opendronemap/odm "bash" 3 days ago Exited (1) 37 hours ago nostalgic_lederberg e31010c00b9a opendronemap/odm "python /code/run.py…" 3 days ago Exited (2) 3 days ago suspicious_kepler c44e0d0b8448 opendronemap/nodeodm "/usr/bin/nodejs /va…" 3 days ago Exited (0) 37 hours ago wonderful_burnell
Accessing logs on the instance¶
Using either the CONTAINER ID or the name, we can access any logs available on the machine as follows:
> docker logs 2518817537ce
This is likely to be unwieldy large, but we can use a pipe | character and other tools to extract just what we need from the logs. For example we can move through the log slowly using the more command:
> docker logs 2518817537ce | more [INFO] DTM is turned on, automatically turning on point cloud classification [INFO] Initializing OpenDroneMap app - Mon Sep 23 01:30:33 2019 [INFO] ============== [INFO] build_overviews: False [INFO] camera_lens: auto [INFO] crop: 3 [INFO] debug: False [INFO] dem_decimation: 1 [INFO] dem_euclidean_map: False ...
Pressing Enter or Space, arrow keys or Page Up or Page Down keys will now help us navigate through the logs. The lower case letter Q will let us escape back to the command line.
We can also extract just the end of the logs using the tail commmand as follows:
> docker logs 2518817537ce | tail -5 [INFO] Cropping /datasets/code/odm_orthophoto/odm_orthophoto.tif [INFO] running gdalwarp -cutline /datasets/code/odm_georeferencing/odm_georeferenced_model.bounds.gpkg -crop_to_cutline -co NUM_THREADS=8 -co BIGTIFF=IF_SAFER -co BLOCKYSIZE=512 -co COMPRESS=DEFLATE -co BLOCKXSIZE=512 -co TILED=YES -co PREDICTOR=2 /datasets/code/odm_orthophoto/odm_orthophoto.original.tif /datasets/code/odm_orthophoto/odm_orthophoto.tif --config GDAL_CACHEMAX 48.95% Using band 4 of source image as alpha. Creating output file that is 111567P x 137473L. Processing input file /datasets/code/odm_orthophoto/odm_orthophoto.original.tif.
The value -5 tells the tail command to give us just the last 5 lines of the logs.
Command line access to instances¶
Sometimes we need to go a little deeper in our exploration of the process for OpenDroneMap. For this, we can get direct command line access to the machines. For this, we can use docker exec to execute a bash command line shell in the machine of interest as follows:
> docker exec -ti 2518817537ce bash root@2518817537ce:/code#
Now we are logged into our docker instance and can explore the machine.
Cleaning up after Docker¶
Docker has a lamentable use of space and by default does not clean up excess data and machines when processes are complete. This can be advantageous if we need to access a process that has since terminated, but carries the burden of using increasing amounts of storage over time. Maciej Łebkowski has an excellent overview of how to manage excess disk usage in docker.
Using ODM from low-bandwidth location¶
What is this and who is it for?¶
Via Ivan Gayton’s repo.
OpenDroneMap can’t always be effectively set up locally—it takes a fairly powerful machine to process large datasets—so a cloud machine can sometimes be the answer for people in the field. However, bandwidth is a problem in many low-income settings. This constraint can’t be solved completely, but the following method does a reasonable job of reducing the bandwidth needed to process drone imagery datasets on the cloud from African locations.
Here we present a tricky but workable process to create an OpenDroneMap cloud machine (not CloudODM, mind you, just a cloud-based instance of ODM that you run from the command line) and use it to remotely process large photo sets. It requires familiarity with Unix command line use, ssh, a Digital Ocean account (Amazon AWS would work as well, possibly with slight differences in the setup), and a moderate level of general computer literacy. If you aren’t fairly computer-savvy and willing to fuss with a slightly tricky setup, CloudODM is what you should be looking at.
The whole process is mostly targeted at someone flying substantial missions in an African or similar location looking to process data ASAP while still in a field setting. Therefore it emphasizes a workflow intended to reduce bandwidth/data transfer, rather than just the simplest way of running ODM.
Create a Digital Ocean droplet with at least 4GB of RAM. That’ll cost about $20/month. Less than 4GB of RAM and the install will probably fail. When we actually run the ODM process we’ll resize it to a much larger—and more expensive—cloud machine, but between runs you can downsize it between runs to the second-cheapest droplet which costs only $10/month (the cheapest droplet, at $5/month, comes with such a small drive that you can’t downsize back to it).
Should be an Ubuntu 18.04 instance to ensure dependency compatibility
Create a user with sudo privileges. Digital Ocean’s insanely good documentation can help you figure this out. In our case we set up a user called
odm, so connecting to it is via the command
ssh email@example.com(where the x’s stand for the IPv4 address of your server). If you want to follow this example closely, do use the username
odm; then your install path will be
/home/odm/ODM/and will match all of the examples in this document.
Go ahead and execute
sudo apt updateand
sudo apt upgradeto ensure your server isn’t dangerously without updates. Make sure to stay with Ubuntu 18.04.
Download and install ODM on it from the ODM Github (regular, not WebODM) with the following commands:
git pull https://github.com/OpenDroneMap/ODM.git cd ODM bash configure.sh install
If you do this from the default home folder of your user (i.e.
odm) the path to the install will be
There are some environmental variables that need to be set. Open the ~/.bashrc file on your machine and add the following 3 lines at the end (From the ODM github). The file can be opened with
nano ~/.bashrc(or whatever text editor you use in lieu of nano). Be sure to replace
/home/odm/with the correct path to the location where you extracted OpenDroneMap if you didn’t do everything exactly as in our example (for example if you used a different username in your server setup):
export PYTHONPATH=$PYTHONPATH:/home/odm/ODM/SuperBuild/install/lib/python2.7/dist-packages export PYTHONPATH=$PYTHONPATH:/home/odm/ODM/SuperBuild/src/opensfm export LD_LIBRARY_PATH=$LD_LIBRARY_PATH:/home/odm/ODM/SuperBuild/install/lib
Note that the ODM github readme contains a slight error, the install directory name will be ODM, not OpenDroneMap (you’ll see this if you compare the above instructions to the ones on the ODM GitHub).
In order to prevent a crash wherein the split-merge process fails to locate its own executable, we add the following lines to
~/.bashrc(adjust paths if you’ve set things up differently from our example):
export PYTHONPATH=$PYTHONPATH:/home/odm/ODM/ export PATH=$PATH:/home/odm/ODM/
Now you’ll need a second cloud hard drive (a “Volume” in Digital Ocean jargon) big enough to manage your project. Rule of thumb seems to be 10 times the size of your raw image set; we’ve got a 100GB image set and set up a 1000GB volume (once the run is done you should be able to get rid of most of this expensive drive capacity, but it’s needed to complete the process). Set up the volume, attach it to your droplet, and configure its mount point (in this example we’re setting it to
Prep data and project¶
Now push your images onto the server. You can use Secure Copy (scp) like so:
scp -r /path/to/my/imagefolder firstname.lastname@example.org:/mnt/odmdata/.
ODM requires the directories on the machine to be set up just so. The
critical bits are the install folder (if you installed as above, it’s
/home/odm/ODM/) and the project folder
ODM’s settings.yaml file specifies a single parent directory containing all projects. This is what goes in the project path line of the settings.yaml file (slightly confusingly, this is actually the parent directory of the individual project directories, which are specified by the project name parameter when calling ODM). Edit settings.yaml and set the project_path parameter to (as per our example setup)
/mnt/odmdata/, which in this case points to the Volume we created. Individual project directories are created within that.
Individual project directories, i.e.
/mnt/odmdata/myproject/contain the gcp_list.txt file, the image_groups.txt file, and the images folder for each project```
The images folder, i.e.
/mnt/odmdata/myproject/images/contains all of the images. If you set it up like this, the images don’t get re-copied because they’re already in the directory that ODM wants them in.
Modify settings.yaml to specify the parent directory of the project folder (in this case the Volume we created,
/mnt/odmdata/). Make sure the images are in the correct spot, i.e.
/mnt/odmdata/myproject/imagesand the other ancillary files (gcp_list.txt and image_groups.txt) are in the root folder
if you have the images in separate folders for individual AOI blocks or flights (which you will if your flight management was organized), you can create an image_groups.txt file with the incantations
for i in *; do cd $i; for j in *; do echo "$j $i" >> ../$i.txt; done; cd ../; done;and
for i in myproject/*.txt; do cat $i >> image_groups.txt; done;. That should create a file with the correct structure: a list of all image files and a “group name” after each one (which in this case will simply be the name of the folder it came from). Then move all of the image files into a single directory called images in the project root dir (so
/mnt/odmdata/myproject/images/). The image_groups.txt file will allow ODM to keep track of which images belong to the same batch, even though they’re all in a single directory.
Resize droplet, pull pin, run away¶
Shut down and resize your machine to an appropriately monstrous number of CPUs and amount of memory. I use the memory-optimized machine with 24 dedicated vCPUs and 192GB of RAM (which costs about $1.60/hr—which adds up fast, it’s over $1000/month). Restart, and get to work quickly so as not to waste expensive big-droplet time.
Launch the ODM process via ssh using nohup (so that if you’re cut off, processing will continue)
Alternately you can use GNU screen to launch the process from a screen session which won’t stop if your connection is interrupted; launch
screen, and use
<ctrl> a <ctrl> dto detach,
screen -rto re-attach. But using screen won’t get you a log file of all of the console output unless you do something specific to capture that, while nohup gives you a file with all of the console output, including error messages, for free.
Note: as of 2020-03 the normal incantation
python run.py -i /path/to/image/folder project_nameseems not to work; the
--imageparameter causes a weird error. So we drop the -i parameter, and rely on the project directory line in the settings.yaml file to direct ODM to the right place. Now using (including a split-merge):
nohup python run.py myproject --split 1 --split-overlap 0 --ignore-gsd --depthmap-resolution 1000 --orthophoto-resolution 5 --dem-resolution 15 --pc-las --dsm
This points ODM at the folder (in this example)
/mnt/odmdata/myproject/. Provided the image_groups.txt and gcp_list.txt are in this folder, the images are in
/mnt/odmdata/myproject/images/, and the project path in settings.yaml is
/mnt/odmdata/it will not waste time and space copying images.
Note that this assumes you have an image_groups.txt file. If not, this
-split-overlap 0will probably fuck things up, and the
--split 1is literally a random number that will be ignored after the image_groups.txt file is loaded (I think it normally controls how many groups it splits a set of images into, but in our case we’re assuming the images are already grouped sensibly). If you don’t have a large dataset (>1000 images), omit the
Follow the progress using tail (so that you’ll know when it’s done)
tail -f nohup.out
You may want to keep an eye on htop (to get a sense of the resource usage so that in future you can only spin up a machine as large as necessary)
After it finishes (assuming you survive that long)¶
As soon as processing is done, shut down the machine and resize it back down to the inexpensive minimum capacity.
Start the machine back up, and log in via ssh.
If you want to save download bandwidth, you can compress the orthophoto using GDAL. Don’t add overviews, do that on your local machine to avoid making the file bigger before downloading it.
gdal_translate -co COMPRESS=JPEG -co PHOTOMETRIC=YCBCR -co TILED=YES -b 1 -b 2 -b 3 -mask 4 --config GDAL_TIFF_INTERNAL_MASK YES /path/to/original/filename.extension /path/to/output.tif
Download using scp:
scp email@example.com:/mnt/odmdata/myproject/odm_orthophoto/odm_orthophoto.tif(or grab the compressed version you created in the last step)
Once you get the file on your local computer, you can use QGIS to add overviews (“pyramids”) or use the GDAL command
gdaladdo -r average /path/to/image.tif 2 4 8 16 32 64 128 256 512 1024.
You can archive the odm_texturing, odm_georeferencing, and odm-dem folders using tar to make them easier to download in one piece (and maybe smaller).
tar -zcvf archivename /path/to/folder