New enhanced version of our software tool for advanced data interpretation and visualization – DataFromSky Viewer – comes up with few very useful and special features! We have added new functionality: Object presence alert.
By a polygon you can define a zone with guarded objects – you can even create conditions under which the alert will appear. This functionality transforms DataFromSky to a perfect mobile solution for monitoring of areas of interest such as airports, military establishments, border lines and other areas with increased demands for flexible protection against penetration of intruders. If you want to get more information about the current state of development, please do not hesitate to contact us using any of our contact details!
Our solution can be used for tracking not only cars and various vehicles, but also for tracking other objects and traffic participants, for example pedestrians or bicycles. Our partner from Denmark – traffic engineering company COWI – is currently studying and analyzing the behaviour of traffic participants in surroundings of schools and kindergartens in Denmark by using DataFromSky for the analysis. It would provide a new and systematic overview of the situation during the morning peak hour, when many students arrive to school at the same time. More detailed, they are searching for some inappropriate and dangerous behaviour of road users and other participants of the traffic, which includes all the parents driving their children to schools, school buses, bikes, motorbikes, other vehicles passing by the neighbourhood and all the pedestrians. Based on this analysis it is easier to formulate some proposals of changes which should be adopted by schools to protect the safety of schoolchildren. This analysis can also lead to better decision making done by municipalities, whether to build a new crossing or add some safety features etc.
If you want to read more about this analysis and its application, visit this website, where you can find the whole article: https://cowicitycreators.wordpress.com/2017/02/22/hvor-farlig-er-skolevejen-naar-vi-ser-den-fra-luften/#more-2643
An example of mapping the behaviour of traffic users is shown in the pictures below, where different user types have different color and each line shows the traffic flow for one trajectory.
Picture: Before (1.) and after (2.) changes: Each line shows one trajectory. Pedestrians (blue), cyclists (red), drivers (green).
DataFromSky will participate in the International Congress on Transport infrastructure and systems (TIS) held in Rome in April 10.to 12.! You have a great chance to see DataFromSky in action and assess its capabilities. Don’t hesitate and visit the TIS conference in Rome to find out more!
One of the main topic discussed on the conference will be the emissions caused by traffic and their impact on human health. The EU Commission is preparing a strategy for clean transport, which should become effective after 2020 and it is in compliance with the strategy proposal for low-emission transport. One of the significant impulses is the increasing air pollution in cities and a high share (up to one third) of road transport in the creation of greenhouse gases. Nowadays, the total share of transport in the creation of greenhouse gases is around 23% (according to data collected by EUROSTAT).
What we are going to present on the Congress, is the possible approaches to solve this topical issue by using data obtained from the DataFromSky software for modelling of intensities and emission from traffic. This tool uses real traffic information about the monitored traffic area, such as the type of the passing vehicles, speed, acceleration profile or traffic density. Using these mentioned data, application of the DataFromSky software to measure the emission load in cities is currently being dealt with. The aim is to create a supporting tool for traffic control, which is in accordance with the Smart Cities concept and leads to reducing the emission load from traffic. We actively collaborate on this project with top researchers from Institute of Forensic Engineering of Brno University of Technology (doc. Ing. Vladimir Adamec, Csc. & Ing. Barbora Schullerova, Ph.D.) . This issue opens the door for further research and implementation in traffic research.
Other participant of the conference will be Andrea Marella, engineer at TrafficLab, our service partner from Italy. His paper “Implementing traffic simulation models with aerial traffic survey” descibes the method how to obtain both a complex set of data of OD matrix and detailed human driver behaviours data in order to set a specific scenario simulation. DataFromSky tool was involved and used in this research as well.
The whole program of the conference and the list of participants can be found here
We are happy to announce that we have succeeded with our proposed project and get supported in Programme Epsilon 2 conducted by Technological Agency of Czech Republic:
System for Preemptive Safety Analysis of Road Network Nodes and Traffic Flow
The aim of the project is development of a preemptive quantitative system for analysis of road network nodes in terms of safety. The system will be
based on analysis of time-space information about each individual traffic member. The information will be automatically extracted with high precision
and frequency from video data collected by UAVs. The trajectory data extracted from video will be analyzed in terms of traffic safety, based on
detection of non-standard and aggressive behaviour, conflict situations, incomfort during passage etc.
We are proud to announce, that we become a member of NVIDIA Inception Program! This virtual accelerator program led by NVIDIA nurtures dedicated and exceptional startup companies, who are revolutionizing industries with advances in AI and data science, and help them in all stages of product development, prototyping, and deployment. Using NVIDIA Pascal GPUs and Caffe’s deep learning framework, DataFromSky is able to do the automatic extraction of time-space-trajectories of moving objects in aerial video data captured by drones and/or balloons. The overall system is divided into three main parts: preprocessing, vehicle detection, and tracking. In the preprocessing step, the acquired image gets undistorted and geo-registered against a user-selected reference frame. The current solution for detection and tracking utilizes machine powered by NVidia GTX 1080 cards.
We designed a highly optimized multi-layer detector and trained it on a hand annotated training dataset with 10.2M positive and 10.2M negative samples per class. The detector is capable of learning from its errors and improving over time. Due to the outstanding performance of detection and tracking algorithms, DataFromSky can provide high-quality data and guarantee a hit rate of over 96%. See the raw output from DFS deep neural networks below: