Deep-learning (DL)-based object detection algorithms can greatly benefit the community at large in fighting fires, advancing climate intelligence, and reducing health complications caused by hazardous smoke particles. Existing DL-based techniques, which are mostly based on convolutional networks, have proven to be effective in wildfire detection. However, there is still room for improvement. First, existing methods tend to have some commercial aspects, with limited publicly available data and models. In addition, studies aiming at the detection of wildfires at the incipient stage are rare. Smoke columns at this stage tend to be small, shallow, and often far from view, with low visibility. This makes finding and labeling enough data to train an efficient deep learning model very challenging. Finally, the inherent locality of convolution operators limits their ability to model long-range correlations between objects in an image. Recently, encoder–decoder transformers have emerged as interesting solutions beyond natural language processing to help capture global dependencies via self- and inter-attention mechanisms. We propose Nemo: a set of evolving, free, and open-source datasets, processed in standard COCO format, and wildfire smoke and fine-grained smoke density detectors, for use by the research community. We adapt Facebook’s DEtection TRansformer (DETR) to wildfire detection, which results in a much simpler technique, where the detection does not rely on convolution filters and anchors. Nemo is the first open-source benchmark for wildfire smoke density detection and Transformer-based wildfire smoke detection tailored to the early incipient stage. Two popular object detection algorithms (Faster R-CNN and RetinaNet) are used as alternatives and baselines for extensive evaluation. Our results confirm the superior performance of the transformer-based method in wildfire smoke detection across different object sizes. Moreover, we tested our model with 95 video sequences of wildfire starts from the public HPWREN database. Our model detected 97.9% of the fires in the incipient stage and 80% within 5 min from the start. On average, our model detected wildfire smoke within 3.6 min from the start, outperforming the baselines.
more »
« less
FIgLib & SmokeyNet: Dataset and Deep Learning Model for Real-Time Wildland Fire Smoke Detection
The size and frequency of wildland fires in the western United States have dramatically increased in recent years. On high-fire-risk days, a small fire ignition can rapidly grow and become out of control. Early detection of fire ignitions from initial smoke can assist the response to such fires before they become difficult to manage. Past deep learning approaches for wildfire smoke detection have suffered from small or unreliable datasets that make it difficult to extrapolate performance to real-world scenarios. In this work, we present the Fire Ignition Library (FIgLib), a publicly available dataset of nearly 25,000 labeled wildfire smoke images as seen from fixed-view cameras deployed in Southern California. We also introduce SmokeyNet, a novel deep learning architecture using spatiotemporal information from camera imagery for real-time wildfire smoke detection. When trained on the FIgLib dataset, SmokeyNet outperforms comparable baselines and rivals human performance. We hope that the availability of the FIgLib dataset and the SmokeyNet architecture will inspire further research into deep learning methods for wildfire smoke detection, leading to automated notification systems that reduce the time to wildfire response.
more »
« less
- Award ID(s):
- 2100237
- NSF-PAR ID:
- 10322779
- Date Published:
- Journal Name:
- Remote Sensing
- Volume:
- 14
- Issue:
- 4
- ISSN:
- 2072-4292
- Format(s):
- Medium: X
- Sponsoring Org:
- National Science Foundation
More Like this
-
-
null (Ed.)Due to Wildfire's huge destructive impacts on agriculture and food production, wildlife habitat, climate, human life and ecosystem, timely discovery of fires enable swift response to fires before they go out of control, in order to minimize the resulting damage and impacts. One of the emerging technologies for fire monitoring is deploying Unmanned Aerial Vehicles, due to their high flexibility and maneuverability, less human risk, and on-demand high quality imaging capabilities. In order to realize a real-time system for fire detection and expansion analysis, fast and high-accuracy image-processing algorithms are required. Several studies have shown that deep learning methods can provide the most accurate response, however the training time can be prohibitively long, especially when using online learning for constant refinement of the developed model. Another challenge is the lack of large datasets for training a deep learning algorithm. In this respect, we propose to use a pretrained mobileNetV2 architecture to implement transfer learning, which requires a smaller dataset and reduces the computational complexity while not compromising the accuracy. In addition, we conduct an effective data augmentation pipeline to simulate some extreme scenarios, which could promise the robustness of our approach. The testing results illustrate that our method maintains a high identification accuracy in different situations - original dataset (99.7%), adding Gaussian blurred (95.3%), and additive Gaussian noise (99.3%).more » « less
-
Current forest monitoring technologies including satellite remote sensing, manned/piloted aircraft, and observation towers leave uncertainties about a wildfire’s extent, behavior, and conditions in the fire’s near environment, particularly during its early growth. Rapid mapping and real-time fire monitoring can inform in-time intervention or management solutions to maximize beneficial fire outcomes. Drone systems’ unique features of 3D mobility, low flight altitude, and fast and easy deployment make them a valuable tool for early detection and assessment of wildland fires, especially in remote forests that are not easily accessible by ground vehicles. In addition, the lack of abundant, well-annotated aerial datasets – in part due to unmanned aerial vehicles’ (UAVs’) flight restrictions during prescribed burns and wildfires – has limited research advances in reliable data-driven fire detection and modeling techniques. While existing wildland fire datasets often include either color or thermal fire images, here we present (1) a multi-modal UAV-collected dataset of dual-feed side-by-side videos including both RGB and thermal images of a prescribed fire in an open canopy pine forest in Northern Arizona and (2) a deep learning-based methodology for detecting fire and smoke pixels at accuracy much higher than the usual single-channel video feeds. The collected images are labeled to “fire” or “no-fire” frames by two human experts using side-by-side RGB and thermal images to determine the label. To provide context to the main dataset’s aerial imagery, the included supplementary dataset provides a georeferenced pre-burn point cloud, an RGB orthomosaic, weather information, a burn plan, and other burn information. By using and expanding on this guide dataset, research can develop new data-driven fire detection, fire segmentation, and fire modeling techniques.more » « less
-
more » « less
Abstract Past studies reported a drastic growth in the wildland–urban interface (WUI), the location where man‐made structures meet or overlap wildland vegetation. Fighting fire is difficult in the WUI due to the combination of wildland and structural fuels, and therefore, WUI areas are characterized by frequent damage and loss of structures from wildfires. Recent wildland fire policy has targeted fire prevention, evacuation planning, fuel treatment, and home hardening in WUI areas. Therefore, it is important to understand the occurrence of wildfire events relative to the location of the WUI. In this work, we have reported the occurrences of wildfires with respect to the WUI and quantified how much of the WUI is on complex topography in California, which intensifies fire behavior and complicates fire suppression. We have additionally analyzed the relative importance of WUI‐related parameters, such as housing density, vegetation density, and distance to wildfires, as well as topographic factors, such as slope, elevation, aspect, and surface roughness, on the occurrence of large and small wildfires and the burned area of large wildfires near the WUI. We found that a very small percentage of wildfire ignition points and large wildfire‐burned areas (>400 ha or 1000 acres) were located in the WUI areas. A small percentage of large wildfires were encountered in WUI (3%), and the WUI area accounted for only 4% of the area burned, which increased to 5% and 56%, respectively, outside WUI (5‐km buffer from WUI). Similarly, 66% of fires ignited outside WUI, whereas only 3.6% ignited within WUI. Results from this study have implications for fuel management and infrastructure hardening, as well as for fire suppression and community response.
-
The objective of this study was to assess feasibility of integrating a coupled fire-atmosphere model within an air-quality forecast system to create a multiscale air-quality modeling framework designed to simulate wildfire smoke. For this study, a coupled fire-atmosphere model, WRF-SFIRE, was integrated, one-way, with the AIRPACT air-quality modeling system. WRF-SFIRE resolved local meteorology, fire growth, the fire plume rise, and smoke dispersion, and provided AIRPACT with fire inputs. The WRF-SFIRE-forecasted fire area and the explicitly resolved vertical smoke distribution replaced the parameterized BlueSky fire inputs used by AIRPACT. The WRF-SFIRE/AIRPACT integrated framework was successfully tested for two separate wildfire events (2015 Cougar Creek and 2016 Pioneer fires). The execution time for the WRF-SFIRE simulations was <3 h for a 48 h-long forecast, suggesting that integrating coupled fire-atmosphere simulations within the daily AIRPACT cycle is feasible. While the WRF-SFIRE forecasts realistically captured fire growth 2 days in advance, the largest improvements in the air quality simulations were associated with the wildfire plume rise. WRF-SFIRE-estimated plume tops were within 300-m of satellite-estimated plume top heights for both case studies analyzed in this study. Air quality simulations produced by AIRPACT with and without WRF-SFIRE inputs were evaluated with nearby PM 2 . 5 measurement sites to assess the performance of our multiscale smoke modeling framework. The largest improvements when coupling WRF-SFIRE with AIRPACT were observed for the Cougar Creek Fire where model errors were reduced by ∼50%. For the second case (Pioneer fire), the most notable change with WRF-SFIRE coupling was that the probability of detection increased from 16 to 52%.more » « less
- Free Publicly Accessible Full Text
-
Accepted Manuscript1.0
- Journal Article:
- https://doi.org/10.3390/rs14041007
