Technology to support search for missing persons

To arrive on time

Search and Rescue with Unmanned Aerial Vehicle


Innovative IT system supporting the search for missing persons in the wilderness.

Planning a drone mission based on human behavior models.

Automatic human detection finalised with accurate locations of people.

Real opportunity to reduce costs of search operations and risk taken by rescuers.

Spin-off company
under the auspices of University of Wroclaw

University of Wroclaw



SARUAV is a software that fits the existing search procedures.


High detection rate in all seasons.

Short calculation time.

Potential locations of missing people available quickly during the process of detection.

Detection finalised with exact coordinates of the potential locations.


System for gathering, archiving and reporting information obtained as the search proceeds.

SARUAV is able to work directly in the field with no Internet access.

Map application that does not require extensive GIS training.

Step by step guideance to help to work under stress.

Easy access to important fragments of photos for quick visual verification of potential location.


Prof. Tomasz Niedzielski


He is a mathematician and geographer. He holds doctoral degrees in technical sciences in the field of geodesy and cartography as well as in Earth sciences in the field of geography. He is a habilitated doctor of Earth sciences (geography) and a full Professor of Earth sciences. Tomasz Niedzielski is an expert in modelling and forecasting hydrological processes. He works on elaborating systems for processing data acquired by unmanned aerial vehicles, in particular for the automated human detection. He gained research experience at the University of Wrocław (Poland), University of Aberdeen (UK), Space Research Centre of Polish Academy of Sciences, Wrocław Centre for Networking and Supercomputing of Wrocław University of Technology (Poland). He served as a principal investigator of numerous research projects financed by domestic and international resources. He is an author or co-author of several dozen of papers published in high-quality international journals, a European patent as well as over 200 presentations at conferences and seminars.

dr Miroslawa Jurecka

Member of the board

She is a geographer who graduated both in the field of physical geography as well as GIS and cartography. She is a Ph.D. candidate at the University of Wrocław. Her doctoral dissertation focuses on the use of geoinformation methods in searching for lost people. Her expertise is mathematical modelling of lost person behaviour in the geographical space, based on ring and mobility models. She also works on the analysis of aerial imagery acquired by unmanned aerial vehicles, in particular she focuses on image processing methods for person detection. Mirosława Jurecka has a few years experience in geospatial industry. She was awarded a fellowship of the Foundation for Polish Science and was an investigator in the research project of the Ministry of Science and Higher Education. She is a co-author of a few papers published in international peer-reviewed scientific journals.

Dr Bartlomiej Mizinski

Member of the board

He is a mathematician and computer scientist. He holds a Ph.D. degree in Earth Sciences in the field of geography, gained in the aftermath of a dissertation on designing automated modelling systems in hydroinformatics. His interests focus on methods for modelling and forecasting time series, stochastic processes, statistics and image processing techniques. Bartłomiej Miziński gained research experience at the University of Wrocław (Poland) and Wrocław Centre for Networking and Supercomputing of Wrocław University of Technology (Poland). He was an investigator of four research projects and was awarded a fellowship of the Foundation for Polish Science. His M.Sc. thesis has been recognized by Polish Mathematical Society. He is a co-author of several publications in high-quality international scientific journals and of a European patent.

Frequently asked questions

How is the SARUAV license constructed?

The company SARUAV Ltd. grants an annual and spatially-limited license. This is both a legal (license conditions) and technical limitation: for each implementation, it is necessary to prepare dedicated spatial data, and the automatic detection of people works in the area specified by the customer.

How is the price of the annual license determined?

When calculating the price of the annual license, several factors are taken into account, e.g. the size of the area and the number of computers on which the system is to be installed.

Each implementation requires a separate assessment, taking into account the customer's expectations regarding the number of workstations with the SARUAV system and the terms of the warranty and service.

How does the implementation process look like?

At present, each implementation is carried out personally by a client. The system is installed on the client's computer with our remote support.

What warranty and service do you provide?

Due to the fact that the selection and purchase of a drone and computer remain the client’s responsibility, during the license period we only service software developed by SARUAV Ltd.

It is done on the terms specified in the license agreement:
"In the event of a failure of the application - the SARUAV system, the Licensee undertakes to notify the Licensor of this fact no later than ................... after receiving information about the event. Within ....................... after receiving the notification, the Licensor undertakes to remove the failure or other event (response time), provided that the Licensee delivered computer hardware to the Licensor within this period in order to remove the failure, unless the scope of the failure does not require interference with the hardware on which the application will be installed. The decision in this regard is made by the Licensor after preliminary analysis of the reported problem."

What parameters must the computer meet for the system to operate with?

The laptop requirements are as follows:

minimal recommended
1660 6GB
2060 6GB
CPU Intel i5
or AMD Ryzen 5
Intel i7
or AMD Ryzen 7
HDD depending on the size of the implementation
(at least 10GB of disk space
for an area the size of a voivodeship
or 40GB for the country)
SOFT WARE Windows 10
current graphics card drivers
+ CUDA Toolkit software
+ Microsoft Visual C++ Redistributable libraries

What are the recommendations for the drone-mounted camera so that the detection can be properly performed?

In order for the detection to be as effective as possible and to receive the most accurate estimation of the coordinates of the indicated persons, the camera should enable the observation of the area in the nadir view, i.e. camera looking vertically down. Camera parameters and flight altitude should be adjusted to ensure the ground sampling distance (GSD) between 1 and 2.5 cm/px. In simple terms, it means that the width/height of one pixel in the photo corresponds to the distance of 1-2.5 cm in the field.

For example, the system works with cameras installed on DJI multirotors (Phantom 4 Advanced, Phantom 4 Pro V2.0, Mavic Enterprise), Autel multirotors, Yuneec multirotors or eBee fixed-wing drones. A camera with a 20 Mpx matrix, when flying at a height of about 60 meters, provides a good photo resolution and GSD at a level below 2 cm/px.

What are the flight recommendations?

Aerial photos should be nadir images of terrain, the camera should point vertically downwards and the lens should be set vertically towards the terrain. Acquisition of aerial photos should be carried out in accordance with the recommendations commonly accepted in low-altitude photogrammetry, i.e. side and front overlap of 60/80%, so that each point is recorded in several photos. Such a large coverage ensures that each point is captured from different directions and angles, so that the possibility of overlooking a person is minimized. High coverage does not significantly affect the duration of the analysis.

Moreover, the altitude of the flight should be adjusted to the parameters of the camera mounted onboard a UAV, so that the obtained photos are of high ground resolution. In order for the detection to be as effective as possible, the parameters of the camera and the flight altitude should be adapted to acquire an image with GSD of 1 to 2.5 cm/px. In simple terms, it means that the width/height of one pixel in the photo corresponds to the distance of 1-2.5 cm in the field.

Is it possible to detect people from video recordings using the system?

No. Videos are not processed by the system. The system operates only with nadir aerial images taken in accordance with commonly accepted recommendations in close-range photogrammetry (side and front overlap at the level of 60/80%).

Does the system use infrared photos and videos?

No. Our two detection algorithms work only on nadir RGB images. Tests with near infrared (NIR) imagery were performed, however, the detection results were found to be less satisfactory.

We also do not work on images acquired in far infrared - thermovision.

Does the system work at night?

Our detection algorithms operate on high-resolution RGB images captured in daylight, photos captured at night are associated with a certain limitation. Even if it is possible to perform night flights from a legal perspective, the photos obtained during such a flight are often of poor quality for the detection to be effective.

The exception may be areas very well lit by artificial light, or performing flights with an additional source of strong light, mounted onboard an unmanned aerial vehicle. Such a reflector could be directed downwards to illuminate the photographed area, and flights should take place at a height that allows sufficient lighting, while maintaining all safety conditions. SARUAV performed some tests regarding this subject, but only preliminary scientific results are available so far.

What is the data processing output and how long does it take to receive the results of the analysis?

Maps of the theoretical range of human movement, as well as maps with automatically indicated locations of objects resembling people are created in real time on a fast laptop, operating completely offline, without access to the Internet.

The detection of people in the photos is possible after the unmanned aerial vehicle has landed and the images have been downloaded to the laptop on which the system is installed. For example, detection on approximately 100 aerial imagery along with full visualization is carried out in 2-3 minutes.

Does the system automatically generate the drone mission?

No. The terrain for the flight can be based on the recommendations obtained in the system, but the final decision on where the drone should be sent rests on the people responsible for planning the search. They have the appropriate management experience and are responsible for the optimal allocation of resources in the field.

The responsibility for planning the drone route and performing the flight appropriately lies with the drone operator who uses the external flight planning software, independent of the SARUAV system, dedicated to specific unmanned aerial vehicles.

Is it possible for the system to work in real time?

The default operation of the detection module is as follows: the drone takes photos (the SARUAV system does not control the drone, and the flight is carried out by the operator), it lands, the memory card is removed and inserted to the laptop or the drone is connected to the laptop using a cable, downloading photos takes place and the detection begins. At the moment, the image processing takes place on a laptop in the field, at the search action site, completely offline, but according to the above-mentioned way (let us say it is near-real time, taking into account the duration of subsequent flights of one drone).

If the customer has a drone that transfers photos (in good quality) to the computer during the flight, we are quite easily able to adapt our application to work in this mode. However, this is not feasible in the current version.

Can the SARUAV system be adapted to detect animals, fire sources or other objects?

Such works were not carried out by us. The human detection system was developed within the last seven years, and during this period, both conceptual and programming work as well as field experiments were carried out. After such a long period of research, our detectors guarantee high efficiency in detecting people, but we cannot directly translate their potential into the identification of other objects.


Click the link to open the article

Przeglad Uniwersytecki


pl. Uniwersytecki 1
50-137 Wroclaw, Poland

NIP: 897-186-20-10
KRS: 0000761532
REGON: 381994896

Share capital: 9 400, 00 PLN
We are on Facebook

Our partners