Spatio-Temporal Data Model for Early Wildfire Detection

Forest fires require rapid and precise early smoke detection to minimize damage. This study focuses on employing smoke recognition methods for early warning systems in forest fire detection, identifying smoke as the primary indicator. A significant hurdle lies in the absence of a large-scale dataset for real-world early forest fire smoke detection. Early smoke videos present characteristics such as smoke plumes being small, slow-moving, and/or semi-transparent in color, and include images where there is background interference, posing critical challenges for practical recognition algorithms. To address these issues, this paper introduces a real-world early smoke monitoring video dataset as a foundational resource. The proposed 4D attention-based motion target enhancement network includes an important frame sorting module which adaptively selects essential frame sequences to improve the detection of slow-moving smoke targets. Additionally, a 4D attention-based motion target enhancement module is introduced to mitigate interference from smoke-like objects and enhance recognition of light smoke during the initial stages. Moreover, a high-resolution multi-scale fusion module is presented, incorporating a small target recognition layer to enhance the network’s ability to detect small smoke targets. This research represents a significant advancement in early smoke detection for forest fire surveillance, with practical implications for enhancing fire management.

DETECT: Development of Technologies for Early HCC Detection

Aiming at the low detection accuracy of existing fire early smoke target detection models, this paper designs a fire early fire smoke detection model based on YOLOv5. First, check the data to understand the fire-prone scenes and divide them into two categories: indoor and outdoor; Second, according to different scenes, considering the influence of environmental factors such as light and scale, manually collect and label the smoke pictures in the early stage of the fire; Third, The Focal Loss loss function is used instead to alleviate the problem of unbalanced classification in the data set. Experiments on the self-made early fire smoke dataset show that the improved network mAP value is 2.3% higher than that of YOLOv5. The experimental results verify the effectiveness of the algorithm.

Early wildfire smoke detection faces challenges such as limited datasets, small target sizes, and interference from smoke-like objects. To address these issues, we propose a novel approach leveraging Efficient Channel and Dilated Convolution Spatial Attention (EDA). Specifically, we develop an experimental dataset, Smoke-Exp, consisting of 6016 images, including real-world and Cycle-GAN-generated synthetic wildfire smoke images. Additionally, we introduce M-YOLO, an enhanced YOLOv5-based model with a 4× downsampling detection head, and MEDA-YOLO, which incorporates the EDA mechanism to filter irrelevant information and suppress interference. Experimental results on Smoke-Exp demonstrate that M-YOLO achieves a mean Average Precision (mAP) of 96.74%, outperforming YOLOv5 and Faster R-CNN by 1.32% and 3.26%, respectively. MEDA-YOLO further improves performance, achieving an mAP of 97.58%, a 2.16% increase over YOLOv5. These results highlight the potential of the proposed models for precise and real-time early wildfire smoke detection.

BackgroundForests are invaluable resources, and fire is a natural process that is considered an integral part of the forest ecosystem. Although fire offers several ecological benefits, its frequent occurrence in different parts of the world has raised concerns in the recent past. Covering millions of hectares of forest land, these fire incidents have resulted in the loss of human lives, wild habitats, civil infrastructure, and severe damage to the environment. Around 90% of wildland fires have been caused by humans intentionally or unintentionally. Early detection of fire close to human settlements and wildlife centuries can help mitigate fire hazards. Numerous artificial intelligence-based solutions have been proposed in the past decade that prioritize the detection of fire smoke, as it can be caught through remote sensing and provide an early sign of wildland fire. However, most of these methods are either computationally intensive or suffer from a high false alarm rate. In this paper, a lightweight deep neural network model is proposed for fire smoke detection in images captured by satellites or other remote sensing sources.ResultsWith only 0.6 million parameters and 0.4 billion floating point operations per second, the hybrid network of convolutional and vision transformer blocks efficiently detects smoke in normal and foggy environmental conditions. It outperforms seven state-of-the-art methods on four datasets, including a self-collected dataset from the “Moderate Resolution Imaging Spectroradiometer” satellite imagery. The model achieves an accuracy of more than 99% on three datasets and 93.90% on the fourth dataset. The t-distributed stochastic neighbor embedding of extracted features by the proposed model demonstrates its superior feature learning capabilities. It is remarkable that even a tiny occurrence of smoke covering just 2% of the satellite image area is efficiently detected by the model.ConclusionsWith low memory and computational demands, the proposed model works exceedingly well, making it suitable for deployment in resource constrained devices for forest surveillance and early fire smoke detection.

Disambiguating past events: Accurate source memory for time and context depends on different retrieval processes

Smoke plumes are the first things seen from space when wildfires occur. Thus, fire smoke detection is important for early fire detection. Deep Learning (DL) models have been used to detect fire smoke in satellite imagery for fire detection. However, previous DL-based research only considered lower spatial resolution sensors (e.g., Moderate-Resolution Imaging Spectroradiometer (MODIS)) and only used the visible (i.e., red, green, blue (RGB)) bands. To contribute towards solutions for early fire smoke detection, we constructed a six-band imagery dataset from Landsat 5 Thematic Mapper (TM) and Landsat 8 Operational Land Imager (OLI) with a 30-metre spatial resolution. The dataset consists of 1836 images in three classes, namely “Smoke”, “Clear”, and “Other_aerosol”. To prepare for potential on-board-of-small-satellite detection, we designed a lightweight Convolutional Neural Network (CNN) model named “Variant Input Bands for Smoke Detection (VIB_SD)”, which achieved competitive accuracy with the state-of-the-art model SAFA, with less than 2% of its number of parameters. We further investigated the impact of using additional Infra-Red (IR) bands on the accuracy of fire smoke detection with VIB_SD by training it with five different band combinations. The results demonstrated that adding the Near-Infra-Red (NIR) band improved prediction accuracy compared with only using the visible bands. Adding both Short-Wave Infra-Red (SWIR) bands can further improve the model performance compared with adding only one SWIR band. The case study showed that the model trained with multispectral bands could effectively detect fire smoke mixed with cloud over small geographic extents.

Early smoke and flame detection based on transformer

This paper presents a video system, operating in the visible spectrum range, for early smoke detection in passenger trains. The main idea is integrating standard smoke sensors with the results of a smoke detection processing, which exploits video surveillance cameras already available on-board the train. To this aim a novel video processing flow is proposed exploiting temporal, spatial and chromatic characteristics of the reference scenario. The proposed algorithm has been successfully verified with several video set and its implementation complexity fully characterized

A method of early smoke detection in video using swaying and diffusion feature is presented in this paper. Firstly, in view of early smoke's swaying feature, choquet fuzzy integral was adopted to e...

A method of early smoke detection in video using swaying and diffusion feature is presented in this paper. Firstly, in view of early smoke's swaying feature, choquet fuzzy integral was adopted to extract dynamic regions from video frames, and then, a swaying identification algorithm based on centroid calculation was used to distinguish candidate smoke region from other dynamic regions. Secondly, smoke diffusion makes different textures between the bottom region and the top region of smoke. This unique feature was used to differentiate smoke from other candidate smoke regions by Gray Level Co-occurrence Matrix. Experiments show that the proposed method is effective, robust, and has a performance of earlier smoke alarm. The processing rate of the smoke detection method achieves 25 frames per second with an image size of 320 × 240 pixels.

The early smoke detection in outdoor spaces concerns to people's life and safety tasks. Such technique is necessary in video surveillance systems near building, on bridges, ships, into tunnels, in landscape monitoring systems, etc. We have suggested a novel video–based method of smoke detection by spatio–temporal clustering which involves three developing stages. The first stage connects with motion detection, the second stage is based on a colour and texture analysis of moving regions and the third stage is enhanced by a spatio–temporal clustering of moving regions with a turbulence parameter. A spatio–temporal volume data permits effectively dynamic track of smoke propagation in outdoor spaces using the designed real–time software. Experimental results show that the proposed set of spatial and temporal features always permits to detect the smoke and non–smoke moving objects in outdoor scenes with a complex background; the percent of true detected smoke elements in test video sequences is up to 89%.

In recent decades, early smoke detection in outdoor environment is a hot topic due to its practical importance for a fire safety. Many researchers have contributed to this area since the 1990s. The chapter aims to follow the evolution of conventional image processing and machine learning methods based on the motion, semi-transparent, color, shape, texture and fractal features to deep learning solutions using various deep network architectures. Our experimental researches in this area have been conducted since 2010. This chapter reflects the original techniques of early smoke detection in complex outdoor scenes.KeywordsSmoke recognitionSmoke detectionSmoke segmentationMachine learningDeep learningFeaturesOutdoor spaceComplex scene

Routine action sequences can share a great deal of similarity in terms of their stimulus response mappings. As a consequence, their correct execution relies crucially on the ability to preserve contextual and temporal information. However, there are few empirical studies on the neural mechanism and the brain areas maintaining such information. To address this gap in the literature, we recently recorded the blood-oxygen level dependent (BOLD) response in a newly developed coffee-tea making task. The task involves the execution of four action sequences that each comprise six consecutive decision states, which allows for examining the maintenance of contextual and temporal information. Here, we report a reanalysis of this dataset using a data-driven approach, namely multivariate pattern analysis, that examines context-dependent neural activity across several predefined regions of interest. Results highlight involvement of the inferior-temporal gyrus and lateral prefrontal cortex in maintaining temporal and contextual information for the execution of hierarchically organized action sequences. Furthermore, temporal information seems to be more strongly encoded in areas over the left hemisphere.

Shaking leaves has been the biggest interference for early forest smoke video detection. Moving average method, Gaussian mixture method and its improved methods are often used to update background, but the performance is not good through background subtraction. Codebook algorithm is applied to extract foreground for early forest smoke detection, quantization techniques are used to obtain background model from time series, then getting foreground image through background subtraction. Through multiple video tests, the experimental results that the filtering performance, anti-noise performance and accuracy are better than other methods above.