Challenges of an Autonomous Wildfire Geolocation System Based on Synthetic Vision Technology

Abstract

Thermographic imaging has been the preferred technology for the detection and tracking of wildfires for many years. Thermographic cameras provide some very important advantages, such as the ability to remotely detect hotspots which could potentially turn into wildfires if the appropriate conditions are met. Also, they can serve as a key preventive method, especially when the 30-30-30 rule is met, which describes a situation where the ambient temperature is higher than 30 ∘C, the relative humidity is lower than 30%, and the wind speed is higher than 30 km/h. Under these circumstances, the likelihood of a wildfire outburst is quite high, and its effects can be catastrophic due to the high-speed winds and dry conditions. If this sort of scenario actually occurs, every possible technological advantage shall be used by firefighting teams to enable the rapid and efficient coordination of their response teams and to control the wildfire following a safe and well-planned strategy. However, most of the early detection methods for wildfires, such as the aforementioned thermographic cameras, lack a sufficient level of automation and usually rely on human interaction, imposing high degrees of subjectivity and latency. This is especially critical when a high volume of data is required in real time to correctly support decision-making scenarios during the wildfire suppression tasks. The present paper addresses this situation by analyzing the challenges faced by a fully autonomous wildfire detection and a tracking system containing a fully automated wildfire georeferencing system based on synthetic vision technology. Such a tool would provide firefighting teams with a solution capable of continuously surveilling a particular area and completely autonomously identifying and providing georeferenced information on current or potential wildfires in real time.

Analyzing Multispectral Satellite Imagery of South American Wildfires Using Deep Learning

“…Some 10-band images had a large number of pixels that were 0, the reasons for which were unknown. 2 After elimination of these images, the data set was narrowed down to 14,274 "fire" images. "Non-fire" images were not included as data for the wildfire detection model, for reasons described in Section 3.2.3.…”

“…The F 2 is a weighted harmonic mean of the precision and recall, with more weight given to the recall. Theoretically, maximizing the recall should be prioritized over maximizing the precision, since in scenarios where wildfire prediction software will be used, it is more important to minimize false negatives than false positives (Arana-Pulido et al 2018) [2]. When a region is mistakenly assessed as wildfire-free, fire-extinguishing resources will not be allocated to that region, meaning the fire could expand while rescuers are not aware.…”

“…Though Landsat 8 Operational Land Imager (OLI) and Thermal Infrared Sensor (TIRS) images have 11 bands, the Panchromatic band with 15 meters resolution was excluded at the discretion of the researchers who processed the data 2. This might have been due to the fact that they were acquired at the borders of WRS path-row grids.…”

Arquitectura de Big Data para la gestión de incendios forestales en la región de La Araucanía

“…Finalmente, se utilizó un criterio de experto simulado para la determinación de los índices de peligro, extraído de la literatura. Se utilizó la conocida como "regla del 30-30-30" mencionada por Arana-Pulido et al (2018), que según los consultores asociados al equipo es una de las reglas aplicadas dentro de la propia CONAF. Esta regla hace referencia a que existe una alta posibilidad de propagación de un incendio si la humedad está bajo el 30 %, la temperatura alcanza 30 °C o más, y el viento tiene una velocidad de 30 km/h (Arana- Pulido et al, 2018).…”

Introducing Sustainability in a Robotic Engineering Degree: A Case Study

“…BomberBot: The operation of this robot is focused on helping in the prevention and extinguishing of forest fires by using the so-called 30-30-30 rule when predicting fire danger. This rule states that if the wind exceeds 30 kilometers per hour in a given location, there is a humidity of 30 percent or less, and a temperature above 30 degrees, there is a high risk of fire [40]. The robot includes sensors to implement this rule and to predict if there is a high probability of fire in an area (see Figure 3c).…”