Fire Spread Rate (FSR) can indicate how fast a fire is spreading, which is especially helpful for wildfire rescue and management. Historically, images obtained from sun-orbiting satellites such as Moderate Resolution Imaging Spectroradiometer (MODIS) were used to detect active fire and burned area at the large spatial scale. However, the daily revisit cycles make them inherently unable to extract FSR in near real-time (hourly or less). We argue that the Himawari-8, a next generation geostationary satellite with a 10-min temporal resolution and 0.5–2 km spatial resolution, may have the potential for near real-time FSR extraction. To that end, we propose a novel method (named H8-FSR) for near real-time FSR extraction based on the Himawari-8 data. The method first defines the centroid of the burned area as the fire center and then the near real-time FSR is extracted by timely computing the movement rate of the fire center. As a case study, the method was applied to the Esperance bushfire that broke out on 17 November, 2015, in Western Australia. Compared with the estimated FSR using the Commonwealth Scientific and Industrial Research Organization (CSIRO) Grassland Fire Spread (GFS) model, H8-FSR achieved favorable performance with a coefficient of determination (R2) of 0.54, mean bias error of –0.75 m/s, mean absolute percent error of 33.20% and root mean square error of 1.17 m/s, respectively. These results demonstrated that the Himawari-8 data are valuable for near real-time FSR extraction, and also suggested that the proposed method could be potentially applicable to other next generation geostationary satellite data.
Forest fires can destroy forest and inflict great damage to the ecosystem. Fortunately, forest fire detection with video has achieved remarkable results in enabling timely and accurate fire warnings. However, the traditional forest fire detection method relies heavily on artificially designed features; CNN-based methods require a large number of parameters. In addition, forest fire detection is easily disturbed by fog. To solve these issues, a lightweight YOLOX-L and defogging algorithm-based forest fire detection method, GXLD, is proposed. GXLD uses the dark channel prior to defog the image to obtain a fog-free image. After the lightweight improvement of YOLOX-L by GhostNet, depth separable convolution, and SENet, we obtain the YOLOX-L-Light and use it to detect the forest fire in the fog-free image. To evaluate the performance of YOLOX-L-Light and GXLD, mean average precision (mAP) was used to evaluate the detection accuracy, and network parameters were used to evaluate the lightweight effect. Experiments on our forest fire dataset show that the number of the parameters of YOLOX-L-Light decreased by 92.6%, and the mAP increased by 1.96%. The mAP of GXLD is 87.47%, which is 2.46% higher than that of YOLOX-L; and the average fps of GXLD is 26.33 when the input image size is 1280 × 720. Even in a foggy environment, the GXLD can detect a forest fire in real time with a high accuracy, target confidence, and target integrity. This research proposes a lightweight forest fire detection method (GXLD) with fog removal. Therefore, GXLD can detect a forest fire with a high accuracy in real time. The proposed GXLD has the advantages of defogging, a high target confidence, and a high target integrity, which makes it more suitable for the development of a modern forest fire video detection system.
7 July 2020 in IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing
As one of the important coastal cities in China, Tianjin has been urbanized dramatically over the past 40 years, and the urbanization rate has been up to 83.15% by 2018. In this study, we used the continuous change detection and classification algorithm to comprehensively understand the urban expansion processes in Tianjin based on the Landsat time series from 1985 to 2018 with 30-m resolution. Specially, we applied the c-factor approach with the Ross Thick-LiSparse-R model to correct the bidirectional reflectance distribution function (BRDF) effect for each Landsat image and calculated a spatial line density feature for improving the change detection and the classification. Based on the study in Tianjin, we found that BRDF correction can substantially improve the change detection (9.00% higher overall accuracy) and classification (1.08% higher overall accuracy); and the line density is also beneficial to classification (0.48% higher overall accuracy), especially for impervious surface (1.70% less commission errors and 1.49 % less omission errors). By analyzing the imperious surface change processes, we observed that Tianjin has undergone rapid urban expansion in the past decades, and the urban area was mainly transformed from cropland around the central area before 2005 and later from the coast.
The spatial pattern and temporal variation in paddy rice areas captured by remote sensing imagery provide an effective way of performing crop management and developing suitable agricultural policies. However, fragmented and scattered rice paddies due to undulating and varied topography, and the availability and quality of remote sensing images (e.g., frequent cloud coverage) pose significant challenges to accurate long-term rice mapping, especially for traditional pixel and phenological methods in subtropical monsoon regions. This study proposed a superpixel and deep-learning-based time series method to analyze Landsat time series data for paddy rice classification in complex landscape regions. First, a superpixel segmentation map was generated using a dynamic-time-warping-based simple non-iterative clustering algorithm with preprocessed spectral indices (SIs) time series data. Second, the SI images were overlaid onto the superpixel map to construct mean SIs time series for each superpixel. Third, a multivariate long short-term memory full convolution neural network (MLSTM-FCN) classifier was employed to learn time series features of rice paddies to produce accurate paddy rice maps. The method was evaluated using Landsat imagery from 2000 to 2020 in Cengong County, Guizhou Province, China. Results indicate that the superpixel MLSTM-FCN achieved a high performance with an overall accuracy varying from 0.9547 to 0.9721, which presents an 0.17–1.23% improvement compared to the random forest method. This study showed that combining spectral, spatial, and temporal features with deep learning methods can generate accurate paddy rice maps in complex landscape regions.
Rice false smut (RFS) is a late-onset fungal disease that primarily affects rice panicle in recent years. Severe RFS can decrease the yield by 20–30% and severely affect rice quality. This research used hyperspectral remote sensing data from unmanned aerial vehicles (UAV). On the basis of genetic algorithm combined with partial least squares to select the feature bands, this paper creates a new method to use the Pearson correlation coefficient method and Instability Index between Classes (ISIC) method to further select characteristic bands, which further eliminated 27.78% of the feature bands when the model monitoring accuracy was improved overall. The prediction accuracy of the Gradient Boosting Decision Tree model and Random Forest model was the best, which were 85.62% and 84.10%, respectively, and the monitoring accuracy was improved by 2.22% and 2.4% compared with that before optimization. Then, based on the UAV hyperspectral data and the combination of characteristic bands selected by the three band optimization methods, the sensitive band ranges of rice false smut monitoring were determined, which were 698–800 nm and 974–997 nm. This paper provides an effective method of selecting characteristic bands of hyperspectral data and a method of monitoring crop diseases’ using unmanned aerial vehicles.
Rice false smut (RFS), caused by Ustilaginoidea virens, is a significant grain disease in rice that can lead to reduced yield and quality. In order to obtain spatiotemporal change information, multitemporal hyperspectral UAV data were used in this study to determine the sensitive wavebands for RFS identification, 665–685 and 705–880 nm. Then, two methods were used for the extraction of rice false smut-infected areas, one based on spectral similarity analysis and one based on spectral and temporal characteristics. The final overall accuracy of the two methods was 74.23 and 85.19%, respectively, showing that the second method had better prediction accuracy. In addition, the classification results of the two methods show that the areas of rice false smut infection had an expanding trend over time, which is consistent with the natural development law of rice false smut, and also shows the scientific nature of the two methods.