Massive progress we have made in DataFromSky! As we keep on further development of DataFromSky solution, we can now boast great successes in the field of automatic detection from a moving UAV!
Imagine that you can directly detect and count the vehicles in traffic congestions while flying a drone over the traffic node! Impossible? Not anymore – since this moment, DataFromSky can detect and analyze vehicles from a flying UAV and provide information about the total number of vehicles and its category. Knowing such valuable information in real time, one can immediately predict the length of delay in the traffic congestion or the length of the tailback itself.
The success rate of the detection exceeds 95,67 % and the detection works on 12 828 CUDA cores – the image is evaluated in 200 ms.
See the video below, showing the raw data output from the detector – the data is not additionally modified in any way (no tracking, no filtration), however, even with the raw output we reached such a high accuracy of detection. Further data processing (tracking, etc.) will make the output even more accurate!
New traffic analysis from Mexico city.
The visualization video is showing capabilities of DataFromSky in very high traffic flow, where the density reaches 9 000 vehicles per hour!
Video displays IDs and categories of tracked objects (car, medium and heavy vehicle, bus, motocycle), current speed and acceleration, stationary time, traffic counting through gates, trajectories visualization and heatmaps based on average speed and acceleration.
Recently, DataFromSky has performed traffic analysis of another video from United Arab Emirates, Dubai! This huge multilevel intersection in the middle of the arabic desert, which you can see in the video below, was a challenging task due to the large area covered and the weather and climate environment with very dusty and sandy conditions. The object of the analysis was monitoring of traffic behaviour in this vast area with traffic density reaching almost 8 500 vehicles per hour!
As you may see in the video, large area of one busy cloverleaf near the city of Dubai was analyzed – the average length of one trajectory in the video reaches 800 meters.
If you regurarly follow our website, you may know, that we are still working on development of mobile solution, which will be capable of collecting and analyzing traffic data in real time. This smart UAV analyzer will be a subsystem of Intelligent Transportation System (ITS). The main objective is to design the cutting edge technology: an UAV based solution for the purpose of active traffic management over a road network.
See also our latest video analyzed in UAE by clicking on this link.Read more
New research paper called “How accurate are small drones for measuring microscopic traffic parameters?” published in Transportation Letters Journal can provide an answer. Adam Babinec, one of the main IT developers standing behind DataFromSky solution, co-author of the article conducted by University of Athens and Brno University of Technology, says:
„With this paper, we wanted to examine the potential of using sUAV as part of the ITS infrastructure as a way of extracting naturalistic trajectory data from aerial video footage from a low volume intersection and a pedestrian passage. Moreover, we have examined the accuracy of speed data collected from a drone compared to data collected from an On-Board Diagnostics II (OBD-II) device.”
The results show, that the distortion correction and calibration phase of the video are of crucial importance for high level of accuracy, in terms of stabilizing and geo-referencing the video. In order to extract accurate data from drone video footage, each aerial video must be stabilized to compensate for camera movement, since sudden vibrations or small wind gusts can lead to large errors on the ground.
Picture: a lightweight hexacopter (left), study area – low volume intersection in University of Athens campus (right)
The research paper is accessible on this page.
The accuracy of DataFromSky system itself was examined in another research paper, created in collaboration with Faculty of Information Technology and Faculty of Civil Engineering at Brno University of Technology and posted in International Journal of Transportation Science and Technology, for more information click here.Read more
How to apply multicriteria selection for all trajectories and choose only the ones you want? With DataFromSky Viewer – sofisticated software for advanced traffic data interpretation and visualization, you are able to create not only the simple Origin Destination Matrix, but with advanced setttings, the multi-criteria selection for the trajectories can be applied.
Following gif presents detailed process, how to do it:
Usecase shown in the gif
If you want to track particular path and see how many vehicles have used that combination, a selection tool can be used.
In step 1, selection criteria for Entry and Exit gate were defined and all the trajectories, which have not used this combination were removed. After this selection, 4 trajectories remain.
Step 2 – As we wanted to know, how many cars leaving the service station in the South exit will turn right on the crossroad, another selection critera was applied. This condition specifies that only trajectories, which have used the combination Entry gate: south exit of the service station and Exit gate: turning right on the traffic lights, will be shown. Finally, only 2 trajectories left, which fulfill all the above defined criteria.
Why is the multi-criteria selection useful?
– show only the trajectories you want
– monitor and analyze traffic behavior in a defined way
– monitor time cars spend in service station / facilities
– find traffic patterns, identify problematic zones