DataFromSky delivers the data for the models of tactical behavior of cyclists communicating their intentions.
The research project focuses on modeling the behavior of bicyclists at intersections under complex traffic situations and interactions with motor vehicle traffic.
Bicycling is an important part of the urban transport system and traffic control measures are necessary to control traffic flow. The capacity and traffic quality of all traffic participants at signalized intersections can be highly influenced depending on the volume of bicycles.
Video data for this project were collected by setting up high definition cameras on the top of the buildings (to obtain suitable aerial view for the DataFromSky Aerial analysis) near several intersections in the cities of Munich and Freiburg in Germany. Explicit and implicit behavior patterns of cyclists at intersections in the video are observed. Explicit behaviors include bicyclists’ body language to reveal their intentions like hand gestures or head movement. Implicit behaviors such as speed change, view change, or path change, indicate the lateral movement of cyclists indirectly. As for lane changing behavior and driving features, such as speed and positions, extracted by DataFromSky, the cyclist’s path selection behavior at intersections is analytically described.
The analyzed video data results in a model that predicts the behavior of cyclists at intersections based on their explicit and implicit behavior patterns in specific traffic situations. The resulting model can be used for example for supporting tactical and operational functions of automated vehicles in specific traffic situations or as an initial analysis that can give traffic and urban planners a very good overview for proposing some traffic adjustments.
This research project, with the use of DataFromSky, is being carried out by Technical University of Munich in Germany, Department of Civil, Geo and Environmental Engineering, lead by Dr. rer. nat. Andreas Keler, Dr.-Ing. Matthias Spangler, created by M.Sc. Georgios Grigoropoulos and Niu Ruixin.
Traffic flow based on trajectries: green – cars, red – cyclists, blue – pedestrians
DataFromSky Team was asked by the Slovak TV to answer the question on how can traffic analysis using AI help to improve the worsening traffic situation in many large cities? Intelligent traffic analysis can help to identify main transportation problems and the causes of frequent bottlenecks or traffic conflicts. Nowadays, transport engineers have a very powerful tool for planning and detecting these shortcomings = artificial intelligence platform DataFromSky.
DataFromSky Aerial analysis is, for its high precision and detailed telemetry data about each traffic participant, utilized by many municipalities, transportation departments, traffic engineers, urban and city planners or universities and research institutions worldwide. Our Aerial analysis is often used as an input for various traffic simulation models, or in behavioral studies of drivers for self-driving vehicles.
Have a look at the full reportage in Slovak TV
(play the video from 0:35 – 5:15, available only in Czech/Slovak language).
We have moved from the Sky to the Ski!
With our newly released DataFromSky Light, we are testing the software in unusual conditions and scenarios. Counting how many skiers are on the slopes? Not a problem at all. With our software, one can easily analyze how many people are currently on the slope, how many are waiting by the ski lifts, how many of them just stopped by a ski bar or restaurant or who is skiing outside marked slopes and much more!
You still have a chance to try the software on your own video, free of charge. It could be another atypical record, but keep in mind the recommended parameters for the input video (to achieve the best results with the highest accuracy):
resolution: 1280×720 or more (min. PAL)
bitrate: 5 Mb/s or more
framerate: 25 or more
min. traffic object size: 32px in one dimension
continuous video without cuts
no camera movement
Upload your video to WeTransfer and send it to firstname.lastname@example.org – if you need more information, follow the instructions in this post.
We are proud to announce that a new version of the DataFromSky Viewer is now available and brings lots of new amazing features for the advanced Safety Analysis. Do you want to evaluate how many risky situations and dangerous driving events occurred on the road? Use our DataFromSky intelligence! We can count the milliseconds to the near accident and predict the accident before it happened!
Traffic safety is a very important subject in the traffic industry and therefore our focus in this version was mainly on the new Safety Analysis feature that detects traffic conflicts like Time to Collision, Post-Encroachment Time or Heavy Braking in the scene. The DataFromSky Viewer now becomes a first tool that allows the user to detect and interactively analyse various traffic conflict situations between different traffic objects. Besides the standard ways of detecting the preemtive traffic conflicts, we used an innovative approach to simulate traffic object movement. Due to the huge interest in the traffic Safety Analysis, you can look forward to new features regarding this topic in the near future.
Enjoy a video showing an example usage of the new Safety Analysis feature:
The development of DataFromSky Safety Analysis features is done in collaboration with FAST VUT Brno and is supported by Technological Agency of Czech Republic (TAČR, project TH02010882).
A very good summarizing article about the capabilities of DataFromSky and how the advanced image data processing is contributing to optimize the urban mobility has been published on LinkedIn. Several use cases in which the data from the sky might be helpful are described by this article:
– possibility of analyzing the traffic node in terms of traffic safety – based on the detection of risky situations, near accidents and collisions
– analysis of interaction of vehicles, cyclists and pedestrians in public areas such as in surroundings of schools, city centers, bus and train stations etc.
– studies of parking lots utilization, behavior of vehicles in a carpark
– and many other applications which you can think about
To summarize, the greatest benefits of traffic research from the aerial video using advanced software for image processing are:
A) Accurate and detailed data about traffic from the bird’s eye perspective:
– complete trajectory and moving patterns of each traffic participant
– complex OD- matrix – vehicle counts, turning movements
– speed and acceleration profile of each vehicle
– vehicle classification into categories (car / heavy vehicle / bus, / motorcycle etc.)
– measurement of travel time and stationary time (queuing time)
– advanced heatmaps based on the velocity, acceleration, traffic flow
– evaluation of traffic safety (detection of near accidents based on calculation of Time to Collision [TTC], detection of risky situations or inappropriate driving behavior, heavy breaking)
B) Reliable, detailed research without influencing traffic behavior (inconspicuous solution).
C) Ultimate traffic research, almost without a human intervention – based on the machine learning and DNN.
D) Powerful tool which can improve traffic safety and urban planning of many Smart cities.
Do you want to know more? Read the whole article – click on this link.
Those of you who are following us regularly know that we are constantly working on improving and creating new possibilities of use for our DataFromSky Aerial Artificial Intelligence platform. DataFromSky can help to save lives not only in the transportation area, but also as a very powerful SAR (Search and Rescue) system!
We are currently developing a unique body recognition application – a software that recognizes human bodies from aerial pictures to find missing people much easier and faster. DataFromSky AAI Rescue system reduces the time required to rescue lost people or help with finding victims of natural disasters or some unexpected situations. Together with a greek UAV partner, we are working on the concept that can save lives in cases of natural disasters such as happened recently in the Greek resort of Mati.
Given the concerns inherent in large regions due to limited rescue distribution, UAV-based frameworks are a promising platform for providing aerial imaging. The primary role of such intelligent drone is to assist in search & rescue, structure and commercial building fires, wildland fires, and help document natural disasters while helping to find the victims faster! With a thermal imaging camera, there is a high chance to detect fires already at its beginnings and possibly to prevent a disaster.
DataFromSky team is constantly working on developing and upgrading our software tool for advanced data visualization and interpretation DataFromSky Viewer. We have recently released a new enhanced version of this tool, including many improvements, modifications and new features! The new version of the Viewer is much more user-friendly, we have simplified the structure of the Menu and the main toolbar and make the UI easier to work with.
What are the most important news and features added?
– possibility to export the data in WGS84 coordinates
– measurement of headways both in seconds and in meters (exporting these data to CSV is a matter of course)
– possibility to export the data for all configured gates into one single file,
– possibility to export a video with anonymization regions
– video position is shown also in milliseconds from the start of the video
– added information about the number of currently occurring trajectories in the frame
– video rotation
We have also created a completely new DataFromSky Demo package for you! You can download a free trial version of DataFromSky Viewer with the full functionality directly on our website (go on our homepage and click on “Try and See” at the bottom)! The demo package contains one short sample video of a roundabout and presents what is DataFromSky capable of in its full version:
– creating and definition of counting gates by the user
– classification of all objects (including bicycles or pedestrians)
– advanced export functions into CSV, XLS, AutoCad enabled
– OD matrix/turn count in defined directions – color overlay, excel report for all configured gates
– speeds and acceleration measurement anywhere in the scene – displayed number or colored overlay –> creation of heat maps
– advanced traffic analysis: Time to Follow (Tf) or Gap Time analysis (Tg)
– stationary time measurement and export
– headways measurement and export in seconds and meters
Take a look at this video, where the basic system functionalities are presented to provide you quick and easy introduction to the software environment. Click on this link.
Naturally, we do not rest on our laurels and we are already working on a new version of the Viewer, which includes the advanced Safety analysis and detection of collisions and near accidents. Follow our progress!
We are proud to announce that we have connected DataFromSky with a simulation model VISSIM! PTV Vissim is a microscopic multi-modal traffic flow simulation software package developed by PTV Group, first developed in 1992 and nowadays, Vissim is a global market leader in microscopic simulation models! The simulation displays all road users and their interactions in one model.
We have created a promo video demonstrating how to export the results from DataFromSky and use it as an input in Vissim. This possibility opens so many opportunities how to utilize the real data from DataFromSky analysis and how to increase the accuracy of your simulation model!
What kind of data can be exported and what are the plans for further interconnection?
– traffic counts
– vehicle category (number of cars, heavy vehicles, buses, motorcycles etc.)
– turning movements – route decisions and routes
– speeds for model calibration
– travel times (between two defined gates)
– headways between two vehicles following each other (gap in seconds) – advanced gap analysis and headways data provided only by DataFromSky!
– queue lengths (will be added)
The video plays also the role of the very first User manual video of DataFromSky Viewer – it describes the process of creating counting gates in order to export the Origin-destination data and import them into Vissim. Other short instructional videos will follow soon!
Enjoy the video:
Another great move and improvement of DataFromSky – take a look what the software is now capable of! We can detect and track people even in very crowded environment, reaching the accuracy of detection over 99 %!
The following video is a part of a project supported by EU programme Horizon 2020 called “Study of the individual and collective behavior of people in large-scale events”. The objective is to extract the trajectories of all individuals and to create a pedestrian behavior model for the crowded environment based on the real data from DataFromSky analysis.
This project opens new channels and brings other new possibilities in using DataFromSky in UAV monitoring:
– utilization in perimeter security – active and real-time aerial surveillance in mass events (sport events, concerts, parades, demonstrations etc.)
– detection of non-standard behavior, dangerous situations, suspicious gathering (in demonstrations and crowds)
– studies of pedestrians movement behavior and its patterns
and much more!
Keep a close eye on our website and social pages to follow our progress (Twitter, Facebook, LinkedIn and Vimeo)!
Another great success!
We are proud to announce, that DataFromSky is mentioned in the Drone Market Environment Map for 2018, among the companies and people that are set to have the biggest impact on the drone market in this year! The map includes around 1000 companies and organizations in 12 main sections, that will truly drive and define the drone industry. You can find DataFromSky in the Software section – Computer Vision/AI & Navigation. To read the article or see the whole map, click on this link.