This paper proposes a new machine learning based system for forest fire earlier detection in a low-cost and accurate manner. Accordingly, it is aimed to bring a new and definite perspective to visual detection in forest fires. A drone is constructed for this purpose. The microcontroller in the system has been programmed by training with deep learning methods, and the unmanned aerial vehicle has been given the ability to recognize the smoke, the earliest sign of fire detection. The common problem in the prevalent algorithms used in fire detection is the high false alarm and overlook rates. Confirming the result obtained from the visualization with an additional supervision stage will increase the reliability of the system as well as guarantee the accuracy of the result. Due to the mobile vision ability of the unmanned aerial vehicle, the data can be controlled from any point of view clearly and continuously. System performance are validated by conducting experiments in both simulation and physical environments.
Active fire detection in satellite imagery is of critical importance to the management of environmental conservation policies, supporting decision-making and law enforcement. This is a well established field, with many techniques being proposed over the years, usually based on pixel or region-level comparisons involving sensor-specific thresholds and neighborhood statistics. In this paper, we address the problem of active fire detection using deep learning techniques. In recent years, deep learning techniques have been enjoying an enormous success in many fields, but their use for active fire detection is relatively new, with open questions and demand for datasets and architectures for evaluation. This paper addresses these issues by introducing a new large-scale dataset for active fire detection, with over 150,000 image patches (more than 200 GB of data) extracted from Landsat-8 images captured around the world in August and September 2020, containing wildfires in several locations. The dataset was split in two parts, and contains 10-band spectral images with associated outputs, produced by three well known handcrafted algorithms for active fire detection in the first part, and manually annotated masks in the second part. We also present a study on how different convolutional neural network architectures can be used to approximate these handcrafted algorithms, and how models trained on automatically segmented patches can be combined to achieve better performance than the original algorithms - with the best combination having 87.2% precision and 92.4% recall on our manually annotated dataset. The proposed dataset, source codes and trained models are available on Github (https://github.com/pereira-gha/activefire), creating opportunities for further advances in the field
Wildfires are one of the costliest and deadliest natural disasters in the US, causing damage to millions of hectares of forest resources and threatening the lives of people and animals. Of particular importance are risks to firefighters and operational forces, which highlights the need for leveraging technology to minimize danger to people and property. FLAME (Fire Luminosity Airborne-based Machine learning Evaluation) offers a dataset of aerial images of fires along with methods for fire detection and segmentation which can help firefighters and researchers to develop optimal fire management strategies. This paper provides a fire image dataset collected by drones during a prescribed burning piled detritus in an Arizona pine forest. The dataset includes video recordings and thermal heatmaps captured by infrared cameras. The captured videos and images are annotated and labeled frame-wise to help researchers easily apply their fire detection and modeling algorithms. The paper also highlights solutions to two machine learning problems: (1) Binary classification of video frames based on the presence [and absence] of fire flames. An Artificial Neural Network (ANN) method is developed that achieved a 76% classification accuracy. (2) Fire detection using segmentation methods to precisely determine fire borders. A deep learning method is designed based on the U-Net up-sampling and down-sampling approach to extract a fire mask from the video frames. Our FLAME method approached a precision of 92% and a recall of 84%. Future research will expand the technique for free burning broadcast fire using thermal images.
In many forest fire incidences, late detection of the fire has lead to severe damages to the forest and human property requiring more resources to gain control over the fire. An early warning and immediate response system can be a promising solution to avoid such massive losses. This paper considers a network consisting of multiple wireless sensors randomly deployed throughout the forest for early prompt detection of fire. We present a framework to model fire propagation in a forest and analyze the performance of considered wireless sensor network in terms of fire detection probability. In particular, this paper models sensor deployment as a Poisson point process (PPP) and models the forest fire as a dynamic event which expands with time. We also present various insights to the system including required sensor density and impact of wind velocity on the detection performance. We show that larger wind velocity may not necessarily imply bad sensing performance or the requirement of a denser deployment.
Automatic visual fire detection is used to complement traditional fire detection sensor systems (smoke/heat). In this work, we investigate different Convolutional Neural Network (CNN) architectures and their variants for the non-temporal real-time bounds detection of fire pixel regions in video (or still) imagery. Two reduced complexity compact CNN architectures (NasNet-A-OnFire and ShuffleNetV2-OnFire) are proposed through experimental analysis to optimise the computational efficiency for this task. The results improve upon the current state-of-the-art solution for fire detection, achieving an accuracy of 95% for full-frame binary classification and 97% for superpixel localisation. We notably achieve a classification speed up by a factor of 2.3x for binary classification and 1.3x for superpixel localisation, with runtime of 40 fps and 18 fps respectively, outperforming prior work in the field presenting an efficient, robust and real-time solution for fire region detection. Subsequent implementation on low-powered devices (Nvidia Xavier-NX, achieving 49 fps for full-frame classification via ShuffleNetV2-OnFire) demonstrates our architectures are suitable for various real-world deployment applications.
Artificial intelligence has been applied in wildfire science and management since the 1990s, with early applications including neural networks and expert systems. Since then the field has rapidly progressed congruently with the wide adoption of machine learning (ML) in the environmental sciences. Here, we present a scoping review of ML in wildfire science and management. Our objective is to improve awareness of ML among wildfire scientists and managers, as well as illustrate the challenging range of problems in wildfire science available to data scientists. We first present an overview of popular ML approaches used in wildfire science to date, and then review their use in wildfire science within six problem domains: 1) fuels characterization, fire detection, and mapping; 2) fire weather and climate change; 3) fire occurrence, susceptibility, and risk; 4) fire behavior prediction; 5) fire effects; and 6) fire management. We also discuss the advantages and limitations of various ML approaches and identify opportunities for future advances in wildfire science and management within a data science context. We identified 298 relevant publications, where the most frequently used ML methods included random forests, MaxEnt, artificial neural networks, decision trees, support vector machines, and genetic algorithms. There exists opportunities to apply more current ML methods (e.g., deep learning and agent based learning) in wildfire science. However, despite the ability of ML models to learn on their own, expertise in wildfire science is necessary to ensure realistic modelling of fire processes across multiple scales, while the complexity of some ML methods requires sophisticated knowledge for their application. Finally, we stress that the wildfire research and management community plays an active role in providing relevant, high quality data for use by practitioners of ML methods.
Due to industry deployment and extension of urban areas, early warning systems have an essential role in giving emergency. Fire is an event that can rapidly spread and cause injury, death, and damage. Early detection of fire could significantly reduce these injuries. Video-based fire detection is a low cost and fast method in comparison with conventional fire detectors. Most available fire detection methods have a high false-positive rate and low accuracy. In this paper, we increase accuracy by using spatial and temporal features. Captured video sequences are divided into Spatio-temporal blocks. Then a saliency map and combination of color and texture features are used for detecting fire regions. We use the HSV color model as a spatial feature and LBP-TOP for temporal processing of fire texture. Fire detection tests on publicly available datasets have shown the accuracy and robustness of the algorithm.
In this work we explore different Convolutional Neural Network (CNN) architectures and their variants for non-temporal binary fire detection and localization in video or still imagery. We consider the performance of experimentally defined, reduced complexity deep CNN architectures for this task and evaluate the effects of different optimization and normalization techniques applied to different CNN architectures (spanning the Inception, ResNet and EfficientNet architectural concepts). Contrary to contemporary trends in the field, our work illustrates a maximum overall accuracy of 0.96 for full frame binary fire detection and 0.94 for superpixel localization using an experimentally defined reduced CNN architecture based on the concept of InceptionV4. We notably achieve a lower false positive rate of 0.06 compared to prior work in the field presenting an efficient, robust and real-time solution for fire region detection.
This paper proposes a vision-based fire and smoke segmentation system which use spatial, temporal and motion information to extract the desired regions from the video frames. The fusion of information is done using multiple features such as optical flow, divergence and intensity values. These features extracted from the images are used to segment the pixels into different classes in an unsupervised way. A comparative analysis is done by using multiple clustering algorithms for segmentation. Here the Markov Random Field performs more accurately than other segmentation algorithms since it characterizes the spatial interactions of pixels using a finite number of parameters. It builds a probabilistic image model that selects the most likely labeling using the maximum a posteriori (MAP) estimation. This unsupervised approach is tested on various images and achieves a frame-wise fire detection rate of 95.39%. Hence this method can be used for early detection of fire in real-time and it can be incorporated into an indoor or outdoor surveillance system.
Remote Sensing applications can benefit from a relatively fine spatial resolution multispectral (MS) images and a high revisit frequency ensured by the twin satellites Sentinel-2. Unfortunately, only four out of thirteen bands are provided at the highest resolution of 10 meters, and the others at 20 or 60 meters. For instance the Short-Wave Infrared (SWIR) bands, provided at 20 meters, are very useful to detect active fires. Aiming to a more detailed Active Fire Detection (AFD) maps, we propose a super-resolution data fusion method based on Convolutional Neural Network (CNN) to move towards the 10-m spatial resolution the SWIR bands. The proposed CNN-based solution achieves better results than alternative methods in terms of some accuracy metrics. Moreover we test the super-resolved bands from an application point of view by monitoring active fire through classic indices. Advantages and limits of our proposed approach are validated on specific geographical area (the mount Vesuvius, close to Naples) that was damaged by widespread fires during the summer of 2017.
Fire disasters typically result in lot of loss to life and property. It is therefore imperative that precise, fast, and possibly portable solutions to detect fire be made readily available to the masses at reasonable prices. There have been several research attempts to design effective and appropriately priced fire detection systems with varying degrees of success. However, most of them demonstrate a trade-off between performance and model size (which decides the model's ability to be installed on portable devices). The work presented in this paper is an attempt to deal with both the performance and model size issues in one design. Toward that end, a `designed-from-scratch' neural network, named FireNet, is proposed which is worthy on both the counts: (i) it has better performance than existing counterparts, and (ii) it is lightweight enough to be deploy-able on embedded platforms like Raspberry Pi. Performance evaluations on a standard dataset, as well as our own newly introduced custom-compiled fire dataset, are extremely encouraging.
Video smoke detection is a promising fire detection method especially in open or large spaces and outdoor environments. Traditional video smoke detection methods usually consist of candidate region extraction and classification, but lack powerful characterization for smoke. In this paper, we propose a novel video smoke detection method based on deep saliency network. Visual saliency detection aims to highlight the most important object regions in an image. The pixel-level and object-level salient convolutional neural networks are combined to extract the informative smoke saliency map. An end-to-end framework for salient smoke detection and existence prediction of smoke is proposed for application in video smoke detection. The deep feature map is combined with the saliency map to predict the existence of smoke in an image. Initial and augmented dataset are built to measure the performance of frameworks with different design strategies. Qualitative and quantitative analysis at frame-level and pixel-level demonstrate the excellent performance of the ultimate framework.
Dynamic texture is a field of research that has gained considerable interest from computer vision community due to the explosive growth of multimedia databases. In addition, dynamic texture is present in a wide range of videos, which makes it very important in expert systems based on videos such as medical systems, traffic monitoring systems, forest fire detection system, among others. In this paper, a new method for dynamic texture characterization based on diffusion in directed networks is proposed. The dynamic texture is modeled as a directed network. The method consists in the analysis of the dynamic of this network after a series of graph cut transformations based on the edge weights. For each network transformation, the activity for each vertex is estimated. The activity is the relative frequency that one vertex is visited by random walks in balance. Then, texture descriptor is constructed by concatenating the activity histograms. The main contributions of this paper are the use of directed network modeling and diffusion in network to dynamic texture characterization. These tend to provide better performance in dynamic textures classification. Experiments with rotation and interference of the motion pattern were conducted in order to demonstrate the robustness of the method. The proposed approach is compared to other dynamic texture methods on two very well know dynamic texture database and on traffic condition classification, and outperform in most of the cases.
In this article, a video base early fire alarm system is developed by monitoring the smoke in the scene. There are two major contributions in this work. First, to find the best texture feature for smoke detection, a general framework, named Histograms of Equivalent Patterns (HEP), is adopted to achieve an extensive evaluation of various kinds of texture features. Second, the \emph{Block based Inter-Frame Difference} (BIFD) and a improved version of LBP-TOP are proposed and ensembled to describe the space-time characteristics of the smoke. In order to reduce the false alarms, the Smoke History Image (SHI) is utilized to register the recent classification results of candidate smoke blocks. Experimental results using SVM show that the proposed method can achieve better accuracy and less false alarm compared with the state-of-the-art technologies.
Emergency events involving fire are potentially harmful, demanding a fast and precise decision making. The use of crowdsourcing image and videos on crisis management systems can aid in these situations by providing more information than verbal/textual descriptions. Due to the usual high volume of data, automatic solutions need to discard non-relevant content without losing relevant information. There are several methods for fire detection on video using color-based models. However, they are not adequate for still image processing, because they can suffer on high false-positive results. These methods also suffer from parameters with little physical meaning, which makes fine tuning a difficult task. In this context, we propose a novel fire detection method for still images that uses classification based on color features combined with texture classification on superpixel regions. Our method uses a reduced number of parameters if compared to previous works, easing the process of fine tuning the method. Results show the effectiveness of our method of reducing false-positives while its precision remains compatible with the state-of-the-art methods.
Social media could provide valuable information to support decision making in crisis management, such as in accidents, explosions and fires. However, much of the data from social media are images, which are uploaded in a rate that makes it impossible for human beings to analyze them. Despite the many works on image analysis, there are no fire detection studies on social media. To fill this gap, we propose the use and evaluation of a broad set of content-based image retrieval and classification techniques for fire detection. Our main contributions are: (i) the development of the Fast-Fire Detection method (FFDnR), which combines feature extractor and evaluation functions to support instance-based learning, (ii) the construction of an annotated set of images with ground-truth depicting fire occurrences -- the FlickrFire dataset, and (iii) the evaluation of 36 efficient image descriptors for fire detection. Using real data from Flickr, our results showed that FFDnR was able to achieve a precision for fire detection comparable to that of human annotators. Therefore, our work shall provide a solid basis for further developments on monitoring images from social media.
In this study, an early fire detection algorithm has been proposed based on low cost array sensing system, utilizing gas sensors, dust particles and ambient sensors such as temperature and humidity sensor. The odor or smell-print emanated from various fire sources and building construction materials at early stage are measured. For this purpose, odor profile data from five common fire sources and three common building construction materials were used to develop the classification model. Normalized feature extractions of the smell print data were performed before subjected to prediction classifier. These features represent the odor signals in the time domain. The obtained features undergo the proposed multi-stage feature selection technique and lastly, further reduced by Principal Component Analysis (PCA), a dimension reduction technique. The hybrid PCA-PNN based approach has been applied on different datasets from in-house developed system and the portable electronic nose unit. Experimental classification results show that the dimension reduction process performed by PCA has improved the classification accuracy and provided high reliability, regardless of ambient temperature and humidity variation, baseline sensor drift, the different gas concentration level and exposure towards different heating temperature range.
Colour analysis is a crucial step in image-based fire detection algorithms. Many of the proposed fire detection algorithms in a still image are prone to false alarms caused by objects with a colour similar to fire. To design a colour-based system with a better false alarm rate, a new colour-differentiating conversion matrix, efficient on images of high colour complexity, is proposed. The elements of this conversion matrix are obtained by performing K-medoids clustering and Particle Swarm Optimisation procedures on a fire sample image with a background of high fire-colour similarity. The proposed conversion matrix is then used to construct two new fire colour detection frameworks. The first detection method is a two-stage non-linear image transformation framework, while the second is a direct transformation of an image with the proposed conversion matrix. A performance comparison of the proposed methods with alternate methods in the literature was carried out. Experimental results indicate that the linear image transformation method outperforms other methods regarding false alarm rate while the non-linear two-stage image transformation method has the best performance on the F-score metric and provides a better trade-off between missed detection and false alarm rate.
Heretofore, global burned area (BA) products are only available at coarse spatial resolution, since most of the current global BA products are produced with the help of active fire detection or dense time-series change analysis, which requires very high temporal resolution. In this study, however, we focus on automated global burned area mapping approach based on Landsat images. By utilizing the huge catalog of satellite imagery as well as the high-performance computing capacity of Google Earth Engine, we proposed an automated pipeline for generating 30-meter resolution global-scale annual burned area map from time-series of Landsat images, and a novel 30-meter resolution global annual burned area map of 2015 (GABAM 2015) is released. GABAM 2015 consists of spatial extent of fires that occurred during 2015 and not of fires that occurred in previous years. Cross-comparison with recent Fire_cci version 5.0 BA product found a similar spatial distribution and a strong correlation ($R^2=0.74$) between the burned areas from the two products, although differences were found in specific land cover categories (particularly in agriculture land). Preliminary global validation showed the commission and omission error of GABAM 2015 are 13.17% and 30.13%, respectively.
현대 영상 감시 시스템은 인간 운영자가 효과적으로 모니터링할 수 있는 수준을 훨씬 초과하는 영상 스트림을 생성하므로, 보안 이벤트의 신속한 탐지를 위해 자동화된 분석이 필수적이다. 본 논문은 공유 GPU 플랫폼에서 구역 기반 인가를 포함한 얼굴 인식, 자동 번호판 인식, 무기 탐지, 화재 및 연기 탐지, 인간 행동 인식을 동시에 수행하는 통합 멀티태스크 심층 학습 프레임워크를 제시한다. 공개 데이터셋에서 충분히 다루어지지 않는 시나리오를 해결하기 위해, 본 연구에서는 두 가지 태스크별 심층 학습 모델을 제안한다: 병합 및 재레이블링된 데이터셋에서 파인튜닝된 단일 클래스 무기 탐지기([email omitted] = 0.947 달성)와, 614개의 영상 클립으로 구성된 자체 구축 반달리즘 데이터셋에서 학습된 SlowFast-R50 행동 인식 모델(94.33% 분류 정확도 달성)이다. 연속 영상에서의 견고성을 향상시키기 위해, 모든 탐지 모듈은 다중 프레임 확인, 신뢰도 가중 투표, 연쇄 필터링에 기반한 시간적 이벤트 검증 아키텍처에 통합되어 프레임 수준의 예측을 신뢰할 수 있는 보안 이벤트로 변환한다. 각 모듈은 공개 데이터셋(LFW, D-Fire, FIRESENSE, UCF-Crime)에서 독립적으로 평가된 후, 통합 엔드투엔드 시스템 평가가 수행된다. 제안된 시간적 검증 전략은 화재 및 연기의 오경보율을 52%에서 4%로 낮추고, 영상 번호판 정확 일치 정확도를 66.7%에서 81.8%로 향상시키며, 전체 프레임워크는 일반 하드웨어에서 프레임당 지연 시간 100ms 미만으로 실시간 동작을 유지한다. 이러한 결과는 특화된 심층 학습 모델과 시간적 이벤트 검증의 결합이 신뢰할 수 있는 실시간 지능형 영상 감시를 위한 효과적이고 실용적인 솔루션임을 입증한다.