Search

Forest fires have been a major threat to forest ecosystems and its biodiversity, as well as the environment in general, particularly in the Mediterranean regions. To mitigate fire spreading, this study aims at finding a fire-break solution for territories prone to fire occurrence. To the effect, here follows a model to map and predict phase transitions in fire regimes (spanning fires vs. penetrating fires) based on terrain morphology. The structure consists of a 2-scale network using site percolation and SIR epidemiology rules in a cellular automata to model local fire Dynamics. The target area for the application is the region of Serra de Ossa in Portugal, due to its wildfire incidence. The study considers the cases for a Moore neighbourhood of warm cells of radius 1 and 2 and also considers a heterogeneous terrain with 3 classes of vegetation. Phase transitions are found for different combinations of fire risk for each of these classes and use these values to parametrize the resulting landscape network.

In this paper, we propose an efficient algorithm to directly restore a clear image from a hazy input, which can be adapted for nighttime image dehazing. The proposed algorithm hinges on a trainable neural network realized in an encoder–decoder architecture. The encoder is exploited to capture the context of the derived input images, while the decoder is employed to estimate the contribution of each input to the final dehazed result using the learned representations attributed to the encoder. The constructed network adopts a novel fusion-based strategy which derives three inputs from an original input by applying white balance (WB), contrast enhancing (CE), and gamma correction (GC). We compute pixel-wise confidence maps based on the appearance differences between these different inputs to blend the information of the derived inputs and preserve the regions with pleasant visibility. The final clear image is generated by gating the important features of the derived inputs. To train the network, we introduce a multi-scale approach to avoid the halo artifacts. Extensive experimental results on both synthetic and real-world images demonstrate that the proposed algorithm performs favorably against the state-of-the-art dehazing for nighttime images.