Carrying out tests for image analysis data collection

Eye In The Sky - News

From 13 to 15 of April of 2021, several tests were carried out at Lousã Aerodrome.

These tests consisted of the acquisition of thermal and visible image data from static platforms and a mobile platform, namely a multirotor drone.

These tests had as main objective the analysis of the variation of the spatial resolution of the objects of interest in the images as a function of the vertical and horizontal distance.

To complement the study carried out previously, these tests had a greater focus on situations with multiple fire ignitions.

The results of these tests will be made available in the near future.

Portugal Space reference to the Eye in the Sky Project

Portugal Space reference to Eye in the Sky Project

In December 2020, an article about Eye in the SKy was published in the Square Kilometer Array’s CONTACT magazine (no. 6) in the IMPACT section, authored by Domingos Barbosa and Miguel Bergano. This article was subsequently published in January 2021 on Portugal Space’s Linkedin social network.

This article demonstrates the application of space sciences and radio astronomy in other areas such as rural firefighting, remote locations without access to telecommunications infrastructures, etc.

The work presents the applicability of space sciences in other fields of endeavor.

Original article: https://issuu.com/ska_telescope/docs/contact_-_issue_06/6

The article on Portugal Space’s Linkedin social network can be viewed here:
https://www.linkedin.com/embed/feed/update/urn:li:ugcPost:6758037681453846528

Note: The Portuguese Space Agency – Portugal Space, is a private non-profit association, responsible for promoting and implementing the “Portugal Space 2030” Strategy and coordinating the management of the several national Space-related programs by encouraging investment, creating qualified jobs and providing services related to Space science and technology within the European Space Agency (ESA) and the process of development of the International Centre for Research in the Atlantic (AIR Center).

Testing for image analysis in uplift experiments

cover news 2021-01-13 Eye In The Sky

On 13 January 2021, several tests were held at Lousã Aerodrome. These tests consisted in the vertical lifting of a drone up to about 250m in height.

This drone was carrying several cameras in the infrared and visible range that recorded the images of a small burning on the ground, on the vertical axis of the drone’s lifting.

These tests had as main objective the analysis of the variation of the images’ resolution as a function of the vertical distance.

We will make the results of these tests available in the near future.

Participation in workshop on international conference NeurIPS

On December, 11 of 2020, an contribution authored by Maria João Sousa, Alexandra Moutinho and Miguel Almeida [1], was presented in the proposal track of the peer-reviewed international workshop, Tackling Climate Change with Machine Learning at the conference Neural Information Processing Systems (NeurIPS).

The proposal addresses the development of expert-in-the-loop systems that combine the benefits of semi-automated data annotation with relevant domain knowledge expertise, towards the generation of large-scale image databases for relevant wildfire tasks to empower the application of machine learning techniques in wildfire intelligence in real scenarios.

The proposal manuscript is openly available online at the workshop webpage (https://www.climatechange.ai/papers/neurips2020/90) alongside the presentation slides, which are also made available on this page.

Reference:
[1] M.J. Sousa, A. Moutinho and M. Almeida, “Expert-in-the-loop Systems Towards Safety-critical Machine Learning Technology in Wildfire Intelligence,” NeurIPS 2020 workshop: Tackling Climate Change with Machine Learning.

News in the Square Kilometer Array CONTACT magazine, issue nº 6

cover news ska magazine EITS

On December 6, 2020, an article was published in the Square Kilometer Array magazine, CONTACT, written by Domingos Barbosa and Miguel Bergano. This article explains the application of the technologies used by SKA, in the monitoring and fire-fighting in Portugal.

The work presented proposes the performance increase of data transmissions carried out through radio frequency channels, usually, applied in emergency communications by the rescue and security forces. The strategy for the development of this work is in line with the tasks carried out by SKA to guarantee the transmission of data from antennas installed in remote and difficult to access locations. The adequacy of the work to be carried out, with certain tasks of the SKA telescope, is aimed at promoting the efficiency of the transfer of information to all who fight the fire, applying Earth Observation techniques.

The communications system to be developed consists of the development of a repeater for voice communications, reception and transmission of telemetric data, commands and images.

The proposal manuscript is openly available online at the webpage (https://issuu.com/ska_telescope/docs/contact_-_issue_06).

Reference:
[1] Domingos Barbosa, Miguel Bergano, “Monitoring wildfires in Portugal with SKA-related technology”, SKAO Magazine, Contact, Issue 6, December 2020, página 7. Available at: https://issuu.com/ska_telescope/docs/contact_-_issue_06

Article published in international journal Sensors (MDPI)

cover news 2020-11-28

On November, 28 of 2020, an article authored by Maria João Sousa, Alexandra Moutinho and Miguel Almeida [1], was published in a peer-reviewed international journal, namely Sensors from the Multidisciplinary Digital Publishing Institute (MDPI).

The article presents analyses of the sensing capabilities and the image processing pipelines of thermal imaging sensors for fire detection applications, paving the way for the development of autonomous systems for early warning and monitoring of fire events.

The article is available online , published under an open-access licence (https://www.mdpi.com/1424-8220/20/23/6803).

 

Reference:

[1] M. J. Sousa, A. Moutinho and M. Almeida, “Thermal Infrared Sensing for Near Real-Time Data-Driven Fire Detection and Monitoring Systems”, Sensors, 2020, 20(23):6803. doi: 10.3390/s20236803

Master thesis in the context of the Eye in the Sky project

In October 2020, Vasco Cepêda completed his master’s degree in Mechanical Engineering presenting his work entitled “Simulation of a latex balloon with a hydrogen generation system”.

This work was supervised by professors Alexandra Moutinho and Rui Costa Neto, being developed in the context of the Eye in the Sky Project.

This work can be accessed via the link: https://fenix.tecnico.ulisboa.pt/cursos/memec/dissertacao/1409728525632951

Participation in Robotics Summer School

From 14 to 18 of September, Maria João Sousa and Carlos Viegas participated in the 2020 IEEE RAS Multi-robot systems Summer School at Czech Technical University in Prague. The summer school featured lectures, as well as participant presentations, including the following themes:

– “Decentralized Multi-UAV Fleets for Wildfire Detection and Monitoring Systems”, presented by Maria João Sousa (IDMEC)

– “Robots for Fire Management”, presented by Carlos Viegas (ADAI)

The event also included practical tutorials and two student competitions with tasks concerning drone path planning and formation control (leader-follower). Maria and Carlos participated as a team in the Leader Follower task and won second place in the competition.

Balloon launch in wildfire event

Within the scope of the Eye in the Sky project, a wildfire monitoring test was carried out in the context of the fire event that started in Conqueiros/Proença-a-Nova (Portugal) by 11:30 am on September 13th.

In this test, a balloon equipped with three cameras operating in the visible range was launched from Montalvão, in the municipality of Nisa, at 7 pm on 14 September. The balloon reached altitudes greater than 12km, covering the path shown in Figure 1.

Figure 1 - Balloon trajectory (in blue ‘o', launch site‘⭑’, payload landing site ‘■’) and fire event region (outlined in green) that led to the burned area (in yellow) in Cunqueiros-Sobreira Formosa

This test had as main objective the test of image acquisition with several cameras operating in the visible range, during the night. The test did not produce satisfactory results, either because the balloon did not fly over the fire due to the limitation in the definition of the launch zone and the direction of the existing winds, or due to the disturbance caused by the smoke column and the excess of artificial lights existing in the captured area. Some of the acquired images can be seen in Figure 2, allowing us to notice the variation in luminosity along the way up.

Figure 2 – Some images acquired in the balloon launch on September 14th near Montalvão.

Presentation of scientific article in international conference

On July, 22 of 2020, an article authored by Maria João Sousa, Alexandra Moutinho and Miguel Almeida [1], was presented at the World Congress on Computational Intelligence (WCCI), a flagship biennial international conference that joins conferences of the three branches of computational intelligence, namely: fuzzy systems (FUZZ-IEEE), neural networks (IJCNN) and evolutionary computation (CEC).

The article proposes a data-driven optimization solution for the dimensioning and design of configurations of fleets of unmanned aerial vehicles, based on an estimated demand for the need to perform aerial missions in a given region. The case study analyzed focuses on the application of this method for situations of monitoring of forest fires.

The article is available online (https://ieeexplore.ieee.org/document/9177642), and also available on this page.

Link to the file on ResearchGate:
https://www.researchgate.net/publication/343894818_Decentralized_Distribution_of_UAV_Fleets_Based_on_Fuzzy_Clustering_for_Demand-driven_Aerial_Services

Reference:

[1] M. J. Sousa, A. Moutinho and M. Almeida. (2020). “Decentralized Distribution of UAV Fleets Based on Fuzzy Clustering for Demand-driven Aerial Services” in IEEE International Conference on Fuzzy Systems (FUZZ–IEEE), Glasgow, United Kingdom, 2020, pp.1–8, doi: 10.1109/FUZZ48607.2020.9177642.

Presentation of scientific article in international conference