最新刊期
- “辽宁省朝阳市森林病虫灾害监测取得新进展，专家利用卫星数据和算法提升监测精度，为森林保护提供数据支持。”
摘要：Objective Due to the combined effects of climate change and human activities, the frequency and scale of forest pest disturbances have been continuously increasing, significantly affecting the structure and services of forest ecosystems. Accurately identifying regional forest pest disturbances and analyzing their spatiotemporal characteristics of outbreaks are of great significance for the protection of forest ecosystems.Method In this study, based on Landsat 8 satellite annual time series data, with Chaoyang City in Liaoning Province as the study area, we comprehensively analyzed the separability of forest canopy temporal spectral characteristics for fire, logging, and forest pest disturbances. Adjusting the control parameters of the LandTrendr algorithm improved the "sensitivity" of extracting weak forest disturbance information, and ultimately, the random forest algorithm was used to extract the spatiotemporal information of forest pest disturbances from 2013 to 2023.Result The results showed that: (1) The temporal spectral characteristics of medium-resolution satellite images can effectively distinguish forest pest disturbances from fire and logging in Chaoyang City, providing a basis for identifying regional forest pest disturbances. (2) Temporal satellite images can effectively extract spatiotemporal information of forest disturbances and be used for forest pest disturbance identification. The overall accuracy of forest disturbance identification and pest disturbance monitoring in this study were 0.893 and 0.891, respectively, with Kappa coefficients of 0.785 and 0.850. (3) Forest disturbances in Chaoyang City are mainly due to pest infestations, primarily occurring in Jianping County and Lingyuan City in the west, accounting for 67.97% of the total pest disturbance area in the city. The forest pest disturbances in Chaoyang City exhibit an "intermittent" outbreak phenomenon in the temporal dimension.Conclusion The study results can provide data support for forest management and offer methodological references for the classification of different forest disturbances and the spatiotemporal monitoring of forest pest disturbances.关键词：Forest pest disaster;time series data;spectral analysis;LandTrendr algorithm;random forest algorithm- 4
- |
- 0
- |
- 0
更新时间：2024-09-27 - “在森林资源监测领域，塞罕坝林场研究展示了高光谱与LiDAR数据结合的树种分类策略，为森林资源高效监测提供了新方法。”
摘要：ObjectiveThis research aims to examine the key factors influencing the accuracy of tree species classification using airborne hyperspectral data combined with Light Detection and Ranging (LiDAR) in forest environments. Accurate identification of individual tree species is essential for effective forest resource monitoring, management, ecosystem assessment, and biodiversity conservation. While many small-scale studies have explored tree species classification in forests with diverse species compositions and complex age structures, achieving this over larger areas remains a significant challenge. This study focuses on evaluating the effects of spectral consistency correction, canopy height information, and individual tree canopy segmentation on classification accuracy. Saihanba Mechanical Forest Farm, a large-scale artificial plantation, was selected as the study site to explore these factors.MethodTo assess the impact of different factors on tree species classification accuracy, the research utilized a Random Forest classification algorithm and developed four distinct classification strategies. The first strategy used vegetation indices derived from multi-flightline images without applying Bidirectional Reflectance Distribution Function (BRDF) correction. The second strategy incorporated BRDF correction into the multi-flightline images before deriving vegetation indices. The third approach integrated canopy height information, specifically the Canopy Height Model (CHM), with the BRDF-corrected vegetation indices. The fourth and final strategy combined BRDF-corrected vegetation indices, CHM, and individual tree canopy segmentation data. The classification accuracy of each strategy was systematically compared to quantify the contribution of each factor toward improving tree species classification precision.ResultThe results indicated that individual tree canopy segmentation significantly reduced misclassification errors arising from the mixing of multiple species within a single canopy, leading to a notable 10.74% improvement in classification accuracy. Using the Random Forest model’s feature importance ranking, individual tree segmentation emerged as the most critical factor, followed by BRDF correction, and then the canopy height model. Although BRDF correction reduced spectral reflectance variability caused by differing sun-observation geometries across flight strips, it only led to a modest improvement in classification accuracy of 3.48%. The introduction of the Canopy Height Model (CHM) yielded minimal gains in accuracy, contributing just 0.67%, particularly in areas with uniform vertical forest structures or species spanning multiple age cohorts.ConclusionThis study demonstrates that integrating airborne hyperspectral data with LiDAR holds substantial promise for enhancing tree species classification in large-scale artificial plantations. Among the factors analyzed, individual tree segmentation proved to be the most impactful in improving accuracy. In contrast, the relatively minor influence of BRDF correction and canopy height features underscores the need for further refinement and optimization. Overall, the findings emphasize the importance of considering multiple factors in remote sensing workflows to enhance the efficiency and accuracy of forest resource monitoring, management, and other forestry-related applications, especially in expansive forest environments. These insights provide a valuable theoretical foundation and practical recommendations for future forest management and ecological monitoring efforts.关键词：Tree Species Classification;airborne hyperspectral data;BRDF correction;LIDAR data;individual tree segmentation;Random Forest;vegetation indices;Saihanba mechanized forest farm- 51
- |
- 19
- |
- 0
更新时间：2024-09-25 - “在海面高度反演领域，研究者提出了一种结合神经网络与注意力机制的误差补偿模型，有效提升了星载GNSS-R反演精度。”
摘要：Objective In the current research on sea surface height inversion from satellite-borne GNSS reflected signals, classical algorithms are usually used to invert sea surface height. However, due to the existence of multiple complex errors such as inaccurate receiver orbit, system error, ionosphere error, troposphere error, etc., the results inverted using classical algorithms are mostly of low accuracy. Therefore, an error model is needed to correct the inversion results. Classic error models generally improve the accuracy of sea surface height inversion by correcting common errors such as tropospheric error, ionosphere error, and antenna baseline attitude error, but there are still large errors that cannot be corrected. To address this problem, this paper proposes an error compensation model based on the combined training of neural networks and attention mechanisms to correct the sea surface height inversion results.Method This paper designs a CNN-AM training method that combines a Convolutional Neural Network (CNN) model with an Attention Mechanism (AM) to accurately train the error of sea surface height inversion from satellite-borne GNSS reflection signals, generate an error compensation model to replace the classical error model, and improve the accuracy of sea surface height inversion.Result The proposed model was compared with the classic error model, CNN model, and random forest model, and tested on about 2 million DDM (Delay-Doppler Mapping) data of the FY-3E dataset. The evaluation indicators used MAE (Mean Absolute Error) and RMSE (Root Mean Square Error). The MAE of the GPS (Global Positioning System) reflected signal data corrected by the error compensation model was 1.74 meters, and the RMSE was 2.25 meters; the MAE of the BDS (Beidou Navigation Satellite System) reflected signal data corrected by the error compensation model was 0.97 meters, and the RMSE was 2.16 meters. Compared with the classic error model, the correction accuracy was improved by about 80%; compared with the random forest model and CNN model, the accuracy was also slightly improved.Conclusion This paper proposes an error compensation model based on the training of neural network and attention mechanism to correct the sea surface height inversion results. Experiments show that the proposed error compensation model effectively corrects the sea surface height inversion error of space-borne GNSS-R.关键词：GNSS-R;neural network;satellite-based;FY-3E;sea surface height inversion;Error;DDM;Beidou- 17
- |
- 43
- |
- 0
更新时间：2024-09-24 - “白鹤滩水电站边坡监测新进展，采用改进InSAR技术，提升了变形监测精度。”
摘要：(Objective)Long-term monitoring of reservoir bank slopes in hydropower stations is essential for early warning of slope instability in the dam area. Interferometric synthetic aperture radar (InSAR) technology has been proven to provide long-term and efficient monitoring of reservoir bank slopes. The engineering experience shows that the surface deformation of the bank slope has a strong response to the reservoir water level change and rainfall, and the cyclic deformation is significant. However, the traditional InSAR deformation estimation method can only obtain the linear deformation rate of the monitored object, and ignoring the effect of reservoir water level change and rainfall on the slope will lead to large residual deformation, which seriously affects the accuracy of the InSAR deformation estimation of the reservoir slope.(Method)In this study, the cycle model and precipitation influence factor are introduced into the time series InSAR deformation modeling, and the InSAR deformation model taking into account the reservoir water level change and rainfall influence is constructed to replace the traditional mathematical empirical model, so as to better describe the deformation evolution law of slope of the reservoir bank in the process of InSAR deformation modeling. By directly estimating the unknown model parameters through the singular value decomposition (SVD) method and using the model parameters to calculate the deformation components caused by water level change and rainfall in the bank slope, the monitoring accuracy of the bank slope deformation can be improved and the prediction of bank slope deformation can be realized.(Result)Based on the improved method in this paper, the time-series deformation results of the Dawanzi-Qiluogou section of Baihetan Hydropower Station were obtained for a period of 31 months. The results show that the deformation in this area is dominated by a linear trend, and the maximum cumulative deformation reaches -155 mm during the period from January 2020 to July 2022, along with the change of the reservoir water level. The RMS of the residual high-pass deformation shows that the modeling accuracy of the improved model is improved by 12.5% compared with that of the traditional InSAR model. The in situ GNSS monitoring results show that the external accuracy of the deformation obtained by this method is ±2.9 mm.(Conclusion)In this paper, an improved time-series InSAR deformation estimation method is proposed for slopes in the bank area. Based on the new method, the time-series deformation results are obtained for a period of 31 months in the section of Dawanzi-Qiluogou. It is found that the slope deformation in the near-river area is larger than that in the far-river area, and the deformation is cyclic along with the change of reservoir water level, and there is a lag effect of 2 months relative to the date of the change of dry and rainy seasons. The method of this paper can replace the traditional InSAR method, and provide an important reference for the identification, monitoring and early warning of potential instability zones in the construction and operation of hydropower stations and other large-scale projects.关键词：InSAR;deformation monitoring;Bank Slopes;Unstable;Time series modeling- 18
- |
- 27
- |
- 0
更新时间：2024-09-24 -
Progress in the Research of Volcano Hotspot Identification Algorithm using Satellite Infrared Data AI导读
“卫星红外技术监测火山活动，专家梳理算法分类与进展，为火山监测提供新视角。”
摘要：Volcano monitoring is essential for predicting volcanic eruptions and taking early warning measures. Traditional ground-based monitoring methods cannot fully cover all volcanoes. Satellite remote sensing technology, with its advantages of global coverage and high temporal and spatial resolution, is an important complement for near-real-time monitoring of volcanic activities, especially the detection of lava flows and volcanic thermal anomalies.This study introduces the current status of typical sensors for infrared remote sensing of volcanic hotspots and summarizes the methodology for detecting volcanic hotspots using satellite infrared data. Firstly, the history of thermal infrared satellite data monitoring and satellite system development is summarized, and various types of algorithms and satellite systems have been applied to make the monitoring of volcanic activities on a global scale more efficient and accurate. Secondly, the development of volcanic hotspot identification algorithms is analyzed, and the existing volcanic hotspot identification algorithms are classified into four categories according to the different characteristics of the volcano used and its surrounding features (spatial/temporal): spatial feature algorithms, temporal feature algorithms, comprehensive feature algorithms and artificial intelligence algorithms. The spatial feature algorithms are categorized into fixed threshold method and dynamic threshold method based on different methods of threshold selection (fixed threshold/dynamic threshold). Based on the above classification, we describe the current status of each type of volcanic hotspot identification algorithms and summarize their data, scope of application, and application limitations, which provides a comprehensive classification and assessment for understanding and improving volcano hotspot detection technology, and is of great significance for the development of future volcano thermal remote sensing theories and technologies.Subsequent research should improve the adaptability of the algorithms in different volcanic environments, combine the advantages of traditional algorithms and artificial intelligence, and utilize historical data and time-series analyses to identify volcanic hotspots more accurately. In addition, fusion of high-resolution and multispectral satellite data will improve the spatial and spectral resolution of volcanic activity monitoring, thus capturing the micro features of volcanoes more accurately. These improvements will enhance the comprehensiveness and accuracy of volcanic hotspot monitoring and provide more reliable support for monitoring, early warning and prevention of geologic hazards.关键词：volcanic lava flows;thermal remote sensing;infrared satellite data;volcano monitoring;thermal anomalies;hotspot automatic detection;algorithm classification;disaster prevention and reduction- 10
- |
- 28
- |
- 0
更新时间：2024-09-24 - “在沼泽湿地NPP估算领域，本研究利用MODIS数据和核函数构建的kNDVI，提高了估算精度，为湿地保护提供数据支持。”
摘要：Objective Swampy wetlands (forest swamps, scrub swamps and herbaceous swamps) are one of the most important carbon reservoirs on earth and play a pivotal role in the global carbon cycle. The proportion of marshy wetlands in China is nearly 40% of the total wetland area, which is of great significance for maintaining regional biodiversity and ecosystem carbon balance. Net primary productivity (NPP) of vegetation refers to the amount of organic matter accumulated by green plants through photosynthesis minus the remaining part of autotrophic respiration per unit of time and per unit of space, and it is one of the most important indicators of the carbon sequestration potential of marsh wetlands, which plays an important role in reflecting the ecological changes of vegetation in the context of climate change.Method Aiming at the relatively weak research on NPP estimation in China's swampy wetlands and the saturation problem in the process of NPP estimation, this study estimated the NPP of China's swampy wetlands in the last 20 years based on MODIS remote sensing data products (MOD13Q1 and MCD12Q1) using the kernel normalized vegetation index (kNDVI) constructed by the kernel function (RBF) with the CASA model. Additionally, the spatiotemporal evolution of China's swampy wetlands and its driving mechanism from 2001 to 2020 were quantitatively analyzed and discussed.Results The results of the study showed that the coefficient of determination (R2) of NPP_kNDVI estimated using the kNDVI index with the measured value of NPP was 0.854, and the root-mean-square error (RSME) was 14.46 g C/m2month, which was closer to the real NPP value compared with NPP_NDVI. Compared with the saturation phenomenon of NDVI in highly vegetated areas, the kNDVI vegetation index mitigates the saturation effect of the vegetation index itself, especially adapts to both densely and sparsely vegetated areas, and improves the accuracy of the estimation of net primary productivity (NPP) of the vegetation to a certain extent. The regional pattern of multi-year NPP mean values in China's swampy wetlands is obvious, showing a decreasing and then increasing trend from low latitude to high latitude in terms of latitude, which is the result of a combination of factors, such as the distribution of swampy wetlands, air temperature, precipitation, and solar radiation. The annual mean change in NPP in the study area from 2001 to 2020 ranged from 162.73 to 189.34 g C/m2a showed a fluctuating upward trend, with a growth rate of 1.215 g C/m2a (R2=0.82) and a mean value of 177.17 g C/m2a. Between 2001 and 2020, the proportions of areas with increasing and decreasing NPP trends in China's swampy wetlands were 72.96 % and 26.27 %, respectively, and were mainly concentrated in the northeastern plains, the northeastern and southwestern parts of Qinghai Province, and the northern part of Sichuan. Compared with human activities, climate change is the main driving factor affecting the spatial and temporal evolution of China's swampy wetlands, with 66.23% and 33.76% of the influence area respectively.关键词：Swampy wetlands;net primary productivity;Spatiotemporal evolution;Climate change;Human activities;CASA;kNDVI;China- 58
- |
- 43
- |
- 0
更新时间：2024-09-13 - “国家卫星气象中心发布风云三号D星雪深产品，经伊春市林区实测数据验证，精度显著提升，为林区雪深反演算法改进提供参考。”
摘要：Objective Snow depth (SD) and SWE (snow water equivalent) are crucial parameter to describe snow cover information. High precision SD and SWE data plays an important role in the study of weather forecast, hydrology, surface processes and other applications. Passive microwave remote sensing is an effective means of observing SD and SWE. Since April 2019, the National Satellite Meteorological Center has released passive microwave global SD and SWE products of Microwave Radiation Imager (MWRI) aboard the Fengyun-3 D Satellite (FY-3D). Compared with FY-3B SD retrieval algorithm, the operational algorithm of FY-3D introduces fractional forest cover for performing empirical correction on forest impact in Northeast China. This study investigates the performance of the improved FY-3D SD and SWE operational algorithm and verify the accuracy of the corresponding products in the forest area in Northeast China.Method This article obtained the situation of SD in the study area over the years through observation data from meteorological stations in Yichun, Heilongjiang Province. The FY-3D SD and SWE operational products are validated through measured snow course data and SD data observed by meteorological stations in the forest areas. Moreover, the uncertainty of FY-3D SD products and the representativeness of meteorological stations are analyzed.Result The results indicate that there is strong temporal heterogeneity in SD distribution in the Yichun region. The verification results indicate that the FY-3D SD product exhibits an overall underestimation, and the RMSE is 5cm and 13.2cm, respectively, when compared with the measurements of snow course and the observations of meteorological station. While the RMSE between the FY-3D SWE product and the snow course data is 2.1mm. FY-3D SD operational algorithm, as a semi-empirical algorithm, cannot eliminate the influence of forests on microwave radiation brightness temperature. Although forest radiometric correction can enhance the correlation between brightness temperature gradient and SD, the empirical nature of forest radiometric correction also increases the uncertainty of snow depth inversion results.Conclusion Analysis shows that the FY-3D algorithm has a lag in response to sudden snow drops due to its lack of response to new snow with an exponential correlation length of 0.11mm. At the beginning of the snow season, when the snow depth remains below 5cm, the change in brightness temperature gradient caused by soil freezing can be misjudged by the inversion algorithm, leading to overestimation of snow depth during this period. In the preliminary exploration of site representativeness, the analysis of the differences between point and surface combined with field observations show that snow in forest areas is deeply influenced by various factors, leading to strong local spatial heterogeneity. This work can provide reference for improving the SD inversion algorithm in forest regions based by domestic FY-3D brightness temperature data in the future.关键词：FY-3D/MWRI;snow depth;snow water equivalent;product validation;forest region- 40
- |
- 29
- |
- 0
更新时间：2024-09-13 - “在植物营养监测领域，研究者利用高光谱技术对薄壳山核桃叶片氮素含量进行无损估算，通过分数阶微分和两波段光谱指数，显著提升了估算精度，为植物养分监测提供了新思路。”
摘要：Objective Nitrogen is not only a component element of protein and chlorophyll, but also plays a key role in plant growth and development. Obtaining and analyzing the nitrogen content in plants can reveal their nutritional status and growth changes. Non-destructive and efficient estimation of plant physiological and biochemical indicators using hyperspectral technology can provide a reliable data collection method for the evaluation of nutrient levels and health status during plant growth and development.Methods In this study, Carya illinoensis (Jiande and Changlin series) was taken as research object. The spectral data of 53 plants were collected randomly, covering a wavelength range of 350~2500 nm. Firstly, fractional order derivative (FOD) was used for spectral preprocessing. Secondly, the spectral response relationship between LNC and spectral reflectance combining two-band spectral indices (normalized difference spectral index, NDSI; difference spectral index, DSI). The variable combination population analysis (VCPA) strategy was used to screen modeling variables. The extreme gradient boosting algorithm (XGBoost) estimation models of canopy FOD single-band and FOD combined with two-band spectral indices were constructed respectively. Finally, a suitable estimation model of LNC based on the experimental conditions was obtained.Results The results showed that the correlation between canopy spectrum after FOD treatment and LNC was improved by 0.152, compared with the raw spectrum. FOD combined with two-band spectral indices (NDSI, DSI) was better than single-band in improving the correlation between spectral characteristics and target components, which was increased by 0.25 and 0.277, respectively. The final selected subset of spectral variable combinations included both strong and weak information variables, playing a crucial role in improving the accuracy of the estimation models. The optimal LNC model is the 1.5th-order derivative transformation combined with two-band spectral index (difference spectral index, DSI), with R2 P = 0.75, and RMSEP = 1.32 g/kg.Conclusions This study confirms the feasibility of rapid and non-destructive LNC estimation of Carya illinoensis using hyperspectral technology. On the other hand, FOD combined with two-band spectral indices can significantly improve the response relationship between spectral characteristics and target variables, enrich hyperspectral data processing methods, and open up a new idea for plant nutrient monitoring.关键词：Carya illinoensis;canopy scale;hyperspectral remote sensing;nitrogen;fractional order derivative;spectral index;Variable combination population analysis;machine learning- 45
- |
- 34
- |
- 0
更新时间：2024-09-13 - “在月球多模态影像匹配领域，研究人员采用8种算法进行实验比较，发现HAPCG算法匹配效果最佳，为月球多源数据应用提供参考。”
摘要：Multi-modal image matching methods have been widely applied in the registration of multi-source remote sensing images of the Earth, but there is a lack of comparative research on the application of multi-modal registration of lunar images. To facilitate high-precision alignment between high-resolution lunar optical imagery and SAR (Synthetic Aperture Radar) imagery, this paper conducts an experimental comparison across various lunar regions, including mid-latitude, low-latitude, the Antarctic, and the Arctic, using a suite of eight algorithms: SIFT (Scale-Invariant Feature Transform), the region-based CFOG (Channel Features of Orientated Gradients), HOPC (Histogram of Orientated Phase Congruency), and the structural feature-based RIFT (Radiation Invariant Feature Transform), HAPCG (Histogram of Absolute Phase Consistency Gradients), HOWP (Histogram of the Orientation of the Weighted Phase descriptor), along with the deep learning models SuperGlue and LoFTR (Local Feature Transformer). The performance of these algorithms is evaluated through four metrics: the number of correct matches, root mean square error (RMSE), redundancy rate, and coverage. The findings reveal that the HAPCG algorithm, which integrates anisotropic filtering with a composite feature vector, outperforms the others in terms of matching quality. The LoFTR algorithm, leveraging self-attention and cross-attention mechanisms, demonstrates robust performance, particularly for lunar imagery with sparse textures. The HOWP and SuperGlue algorithms exhibit mid-range performance in terms of matching efficacy. In contrast, the CFOG, HOPC, and RIFT algorithms yield the least satisfactory results, with the SIFT algorithm failing to establish any matches. The distribution of matched points is influenced by factors such as imaging illumination conditions and the extent of the overlapping regions, with matches in mid and low latitude areas proving more successful than those in polar regions. A statistical analysis of the Stokes parameter for the HAPCG matches indicates that the mean values of the scattering characteristic parameters for points in the Mare and upland experimental areas are higher than those in polar regions, aligning with the topographical characteristics. Scatter plots also show a correlation between the Stokes parameter of the HAPCG-matched points and the grayscale values of the optical images, underscoring the algorithm's robustness in matching under conditions of nonlinear radiative variability between optical and SAR imagery. To improve the multi modal matching algorithms in the future, effective feature descriptors can be developed to extract the feature points, combining with the geological knowledge of lunar images. At the same time, robust error removal models can be studied to improve the accuracy of matching feature points between lunar optical image and SAR image. Moreover, we can construct public datasets of lunar optical images and SAR images for deep learning methods to improve the generalization ability of existing machine learning models. Based on the imaging mechanism of lunar optical images and SAR images, we can try to find the deep semantic information of lunar multi modal images, and construct new multi modal image matching network. This study offers insights into the selection of appropriate matching methodologies for lunar optical and SAR imagery, thereby enhancing the utility of lunar multi-source data applications.关键词：Moon;Multi-modal image matching;SAR images;Optical images;Stokes vector- 39
- |
- 51
- |
- 0
更新时间：2024-09-13 - “在月球科学领域，嫦娥二号数据揭示了FFCs的微波热辐射特征，证实了岩浆侵入成因，为月球热演化研究提供新支撑。”
摘要：The formation mechanisms and evolutionary history of lunar floor-fractured craters (FFCs) have been a hot topic of research in lunar science. FFCs are characterized by shallow, often plate-like floors and contain radial, concentric and/or polygonal fractures; additional interior features may include ridges, pits of mare material and dark-haloed pits associated with volcanic activities. Current studies of FFCs are based on visible data, gravity data, radar data, and numerical simulations based on observational data. The penetration depth of visible and infrared radiation in the lunar weathering layer is limited to a few microns. At this limited depth, the lunar weathering layer is easily contaminated by surrounding impact ejecta. The main mechanisms of formation are currently classified into two views: viscous relaxation and magmatic intrusion, and the major difference between these two mechanisms is the presence or absence of dikes in the deeper part of the impact crater. Therefore, based on the Chang’E-2 microwave radiometer (MRM ) data, which has a certain penetration depth and can reflect the thermophysical properties of the material, we selected eight representative FFCs with a diameter greater than 80 km and with center coordinates within 60 degrees north-south latitude, according to the following criteria: (1) to better display the bright temperature characteristics inside the impact crater, we selected the FFCs whose brightness temperature characteristics are less affected by the material outside the crater, i.e. there is no significant amount of basaltic material outside the crater; (2) to study the thermophysical properties of the crater, we selected FFCs whose surfaces are less affected by impact events; (3) we selected impact craters with larger diameters to represent the similar behaviors of FFCs. Meanwhile, based on the 24h brightness temperature (TB) mapping, normalized brightness temperature (nTB) mapping and brightness temperature difference (dTB) mapping methods and combined with the exposure of surface basalt, we systematically study the microwave thermal radiation characteristics of lunar FFCs. The main findings are as follows: (1) The dTB behaviors show that there are regions with high dTB values in all four channels and that the surface fractures in these regions are well-developed. (2) In the FFCs with basalt exposed on the surface, there are microwave thermal emission anomalies in the basalt exposed areas, which indicate that the dike, forming the surface volcanic features; (3) In the FFCs with no basalt exposed on the surface, there are microwave thermal emission anomalies on the bottom of the craters, which indicate that there exist the dike in the deep part of the craters. These results confirm that the lunar FFCs were caused by magma intrusion from the perspective of microwave thermal emission, and provide important new support for the study of the thermal evolution history of the Moon.关键词：Chang'E-2 Microwave Radiometer Data;Floor-Fractured Crater;Lunar magmatic activities;Moon- 113
- |
- 26
- |
- 0
更新时间：2024-09-13 - “广州地表形变监测研究取得进展，利用InSAR技术揭示了地铁沿线、垃圾填埋场和农田区的形变特征，为城市安全提供数据支持。”
摘要：As one of the core cities of the Pearl River Delta, Guangzhou plays an important role in economic development and transportation. However, with the advancement of large-scale engineering construction and the increase in human activities, geologic hazards have become more prominent. Therefore, high-precision deformation monitoring and cause analysis are essential to safeguard the city's socio-economic development and public safety. Moreover, attribution analyses of surface deformation along metro lines are executed based on a buffer zone, but there are few studies on the sensitivity of buffer size selection to the influence factors of subsidence. Clear attribution analysis will provide guidance for the prevention and control of geological disasters along the subway.In this study, we collected 85 scenes of Sentinel-1A data covering Guangzhou from May 2017 to May 2020 and utilized IPTA time-series InSAR technology to obtain the surface deformation time series of Guangzhou. By combining GIS spatial analysis techniques and Pearson correlation statistics, the influencing factors behind the deformation were quantitatively analyzed. Additionally, field survey data were introduced to examine the impact of buffer zone distance selection along subway lines on the correlation between various influencing factors and surface deformation. The results show that surface deformation within Guangzhou exhibits a decentralized distribution, characterized by localized deformation along metro lines and residential areas, large-scale deformation in landfill sites, and widespread deformation in farmland areas. The largest deformation is observed at LiKeng landfill, with a deformation rate of -54.5 mm/yr. Specific to the subsidence along metro lines, obvious deformations (<-20 mm/yr) primarily concentrated on lines 4, 9, 14, 6, and 18, with a pixel percentage of 0.14%，0.08%，0.07%，0.05%, and 0.04% respectively. The largest deformation rate was recorded at KeMuLang station on line 6, reaching -39.5 mm/yr. Moreover, attribution analysis was carried out between surface deformation, operation time, subway distance, road network density, and building load. Settlement along the metro line demonstrates a moderate negative correlation with operation time (r=-0.53), suggesting that as metro lines operate for longer durations, settlement magnitudes decrease. For subway distance, a negative correlation between settlement and distance was observed, the closer to the subway, the greater the chance of settlement. There is a positive correlation between settlement and road network density, and between settlement and building loads, the correlation coefficient of each line is mostly less than 0.2. Additionally, three buffer zones were selected, i.e., 800m, 1000m, and 1500m, to analyze their sensitivity to the abovementioned factors. The results show that only subway distance is sensitive to the buffer zone size, the remaining factors (road network density and building load) are not sensitive to buffer change. Moreover, 58 field samples were collected to select the appropriate buffer zone for the attribution analysis of deformation along Guangzhou’s metro lines, the result is 1000 meters. The influence factors studied in this paper are relatively limited, attention should be focused on building construction, groundwater level, excavation depth, geological conditions, and other factors, and the quantitative analysis based on machine learning should be studied in future work.关键词：Guangzhou Metro;InSAR;deformation monitoring;Attribution analysis- 42
- |
- 44
- |
- 0
更新时间：2024-09-13 - “在海洋测绘领域，研究者提出了一种结合地理位置特征的水深反演方法，通过BP神经网络模型显著提升了浅海测深的精度和可靠性，为海上航运和资源保护提供了高效数据支持。”
摘要：Bathymetric maps with high spatial resolution can display topographic details and provide data support for maritime navigation, coastline management, and marine resource utilization and development.This study conducted experiments Weizhou Island, China, and Molokai Island, USA sea areas. With the support of Sentinel-2and Landsat-9 images, a water depth inversion method was proposed, which incorporates geographic location features as modeling elements, and an optimal water depth inversion model based on a BP neural network was constructed. Finally, different remote sensing data were used to conduct accuracy tests of the inversion method proposed in this paper in various sea areas.The conclusion demonstrated that using all bands of the image for modeling, the inversion of the water depth map is smoother, and better able to invert regional bathymetry trends, with fewer outliers and more accurate inversion results. After incorporating geographic location features, the addition of vegetation index features did not yield better results. Instead, it slightly decreased the modeling accuracy of the model. This indicates that blindly adding modeling elements does not necessarily improve modeling accuracy. Analyzing the autocorrelation between each element and making comprehensive decisions on modeling factors is important. In summary, the water depth inversion model constructed in this paper has high accuracy, strong reliability, and good portability, and can be effectively used for shallow sea depth measurement.The results indicate that During the model selection process, it was found that machine learning models demonstrated higher modeling accuracy than all empirical models.The BP neural network model exhibits the highest modeling accuracy in machine learning models. In addition, the machine learning model is more stable, the inversion of the water depth map can better invert the actual water depth change in the experimental area, and the inversion image is smoother. The introduction of geographic location features can significantly improve the accuracy of water depth inversion. Experimental results have shown that the inversion accuracy in the Weizhou Island was improved from an R2 value of 0.7666 to 0.9952 and the RMSE was reduced from 2.5016m to 0.3578m. As a validation experiment, the R2 value in the Molokai Island area was 0.9939, and the RMSE decreased from 3.0165m to 1.0189m. At the same time, the introduction of geographic location features can also eliminate the influence of some clouds and fog on remote sensing images, and obtain more accurate water depth inversion results.关键词：optical remote sensing;Offshore waters;Geographic location characterization;BP neural network model;Weizhou Island;Molokai Island;accuracy validation- 19
- |
- 24
- |
- 0
更新时间：2024-09-13 - “在可持续能源转型领域，专家构建了多源遥感数据与人工智能算法评估框架，为屋顶光伏潜力评估提供新工具。”
摘要：Objective Rooftop solar photovoltaic (PV) systems are becoming increasingly critical in the global shift towards sustainable energy. Despite their importance, the fragmented and small-scale spatial distribution of rooftop PV systems poses significant challenges for accurate and detailed regional potential assessments. This study aims to tackle these challenges by developing a comprehensive assessment framework that integrates multi-source remote sensing data and advanced artificial intelligence algorithms. The objective is to provide a robust methodology for evaluating the potential of rooftop PV systems on a large scale.Method The assessment framework developed in this study leverages a combination of geostationary meteorological satellite imagery and deep learning inversion models to estimate hourly surface solar radiation. To extract building outlines accurately, high-resolution remote sensing images are processed using advanced image segmentation models. Furthermore, the framework integrates a geometric optical model to simulate the PV generation process. This holistic approach enables the precise revelation of spatial and temporal variations in solar energy resources. It also facilitates the investigation of the total available rooftop resources and the determination of PV power generation potential at meter-level resolution and hourly scales.Result The framework's effectiveness was validated through a case study conducted in Jiangsu Province, China. The results demonstrated the scalability and applicability of the framework across different geographic locations and multiple temporal scales. The estimation results revealed that the rooftop resources in Jiangsu Province could support a PV installed capacity of 236.25 GW, with an annual power generation potential of 303.81 TWh. This substantial output could meet 41.1% of the province's total electricity consumption. The case study highlights the framework's ability to provide detailed and accurate assessments of rooftop PV potential on a large scale.Conclusion This study illustrates the feasibility and effectiveness of integrating multi-source remote sensing observations for spatiotemporal assessment of rooftop PV potential. The developed framework offers robust tools and technical support for advancing the sustainable energy transition. By providing insights into the spatial and temporal variability of solar resources, this framework paves the way for optimized utilization of rooftop PV systems. This research contributes to the broader effort of achieving sustainable energy goals by enabling more precise and large-scale assessments of rooftop PV potential.关键词：Renewable energy;rooftop photovoltaics;remote sensing image segmentation;surface solar radiation inversion;carbon reduction- 536
- |
- 80
- |
- 0
更新时间：2024-08-23 - “森林冠层结构复杂性研究取得新进展，激光雷达技术为全面表征冠层结构提供新机遇，推动生态遥感研究拓展应用。”
摘要：The forest canopy structure plays a crucial role in regulating the exchange of substances and energy between plants and the atmosphere, thereby influencing regional microclimate and ecosystem functionality. Accurate characterization of vegetation canopy structure is of significant importance for forest ecosystem research, such carbon storage estimation, carbon cycle simulation etc. Canopy structural complexity, also known as canopy structural biodiversity, which describes the spatial distribution of branches and leaves within the canopy, has emerged as a key attribute in forest ecosystems and has found wide application in related research. For example, carbon cycle, mechanisms of community composition, sustainable forest management, wildlife conservation, forest disturbance monitoring and restoration, forest microclimate research and so on.Traditional ground-based survey methods have limitations as they only provide partial information through statistical values, which primarily involve plot-based surveys using tools such as diameter tapes, clinometers, and angle gauges to obtain individual tree information such as tree position, diameter at breast height, tree height, and crown width. The heterogeneity of these measured tree attributes and their distribution, such as diameter at breast height and tree height, or combinations of tree height, diameter at breast height, and tree density, are used to quantify canopy structure complexity, including the standard deviation, coefficient of variation, and Gini coefficient of survey attributes. However, these indices may not fully represent canopy structural complexity.The rapid development of lidar technology has enabled the rapid acquisition of three-dimensional structural information for entire forests, offering new opportunities for comprehensive and accurate characterization of canopy structure complexity. In addition to the indicators used in traditional ground-based survey methods, existing quantitative indices for canopy structure complexity based on lidar data can generally be divided into three categories: horizontal distribution indices, vertical distribution indices, and integrated distribution indices. Horizontal distribution indices primarily quantify the horizontal spatial distribution of canopy elements, without considering their vertical distribution, such as canopy cover, canopy closure, and leaf area index. Vertical distribution indices mainly describe the heterogeneity of canopy element distribution in the vertical direction while neglecting their horizontal distribution including canopy effective layers and leaf height diversity and so on. Integrated distribution indices consider both the horizontal and vertical distribution heterogeneity of canopy structure, thereby overcoming the limitations of solely considering a single direction in horizontal or vertical distribution indices, for example canopy fractal dimension, canopy roughness, and canopy entropy.Finally, we summarize the current applications of canopy structure complexity in regulating forest ecosystem functions, including light resource utilization, precipitation interception, microclimate modulation, productivity, and ecosystem stability. Additionally, there are key issues and directions that require emphasis in forest ecosystem research related to canopy structure complexity. These include investigating the cross-platform generality of lidar-based indicators, addressing scale issues, and establishing long-term monitoring methods. While the concept of forest canopy structure complexity is relatively new and has limited application in China, we anticipate that advancements in characterization methods and a deeper understanding of its implications will be facilitated by the increasing availability of long-term, multi-source remote sensing data and the utilization of various deep learning methods.关键词：ecological remote sensing;forest canopy structure;forest ecosystem;light regulation;precipitation interception;productivity;microclimate;forest stability- 531
- |
- 57
- |
- 0
更新时间：2024-08-22 - “在点云轮廓特征点提取领域，本研究通过构建邻域特征投影平面和多参数提取模型，提出了一种直接利用点云邻域特征的提取方法。实验结果表明，该方法在准确率、召回率及F1分数上均优于现有技术，为点云轮廓特征点提取提供了高效解决方案。”
摘要：The feature points of point cloud profiles are the key to determine the geometry of objects, and play an important role in target detection and location. The objective of this study is to extract the point cloud contour feature points directly by using the point cloud neighborhood features.First, the Cholesky decomposition was used to determine the main and secondary eigenvectors, and the neighborhood projection plane based on the main and secondary eigenvectors as normal vectors was constructed respectively. Secondly, the optimal number of neighboring points is determined by constructing the entropy model of neighborhood dimensional feature information, and the angular distribution characteristics of the vector composed of target points and neighborhood points on the projection plane are analyzed. Based on the characteristics of the azimuth Angle, a fine extraction method of boundary points based on neighborhood feature distribution is proposed. Finally, a two-dimensional view formation method of neighborhood points on the projection plane based on quaternion method is proposed, and a multi-parameters extraction model based on the distance from point to line and the deviation of points on both sides of the line is established. Experimental results show that the proposed method is superior to ordered point Hough transform, patch segmentation and binary image methods. In terms of noise immunity, the proposed method can extract contour feature points under different noises, and its robustness is better than that of binary image method, region clustering curvature method and Regional growth method. In addition, the accuracy rate, recall rate and F1 score of this method are all higher than 90%. The F1 score of the proposed method is 4.2% higher than that of the region clustering curvature method and 32.4% higher than that of the Hough transform method. The conclusion that the method in this paper is not only suitable for regular planar building shapes, but also suitable for extracting contour feature points of irregular curved building shapes.关键词：point cloud;Outline points;Feature points;Boundary points;Fold points;Neighborhood feature;Projection plane;feature extraction- 476
- |
- 41
- |
- 0
更新时间：2024-08-21 - “本研究深入分析了中国三大粮食作物的蒸散发耗水特征，揭示了其时空变化规律，为制定精准灌溉方案和水资源保护利用提供了科学依据。”
摘要：Water is a key factor influencing crop growth. Crops consume water primarily through transpiration, which is a crucial process of plant growth. With increasing drought risk, accurately estimating crop water consumption is of utmost importance, especially in areas where water resources are scarce and agriculture dominates the economy. Understanding the water requirements of crops at different growth stages and spatial-temporal scales is crucial for developing efficient irrigation strategies and improving water resource utilization. Evapotranspiration is the primary way of crop water consumption. This study employed remote sensing-derived evapotranspiration and crop phenology data, with high spatial and temporal resolution and high accuracy, to analyze the evapotranspiration water consumption characteristics of three staple crops (maize, rice, and wheat) in nine agricultural regions of China. The analysis focused on different time scales, including the yearly scale, growth seasons, phenological dates, and changes from 2001 to 2019. The findings revealed that the yearly evapotranspiration water consumption in the planting regions of the three crops gradually decreased from south to north. Irrigation significantly increased surface evapotranspiration, and evapotranspiration exceeded precipitation in arid and semi-arid areas, such as the Huang-Huai-Hai Plain and the Loess Plateau. From 2001 to 2019, yearly evapotranspiration water consumption in most crop planting areas experienced an increasing trend, with a significantly higher increasing rate in the north compared to the south. The spatial and temporal variability of evapotranspiration water consumption in winter was more pronounced than in summer. Maize and rice exhibited higher evapotranspiration water consumption during the growing season than wheat. Single rice exhibited higher evapotranspiration compared to early and late rice, while spring wheat had higher evapotranspiration than winter wheat. The increase rate in evapotranspiration water consumption during the growing seasons of maize was more pronounced than that of wheat and rice. The proportion of evapotranspiration water consumption during a single growing season to the whole year was greater in the north than in the south. Daily evapotranspiration water consumption among the three staple crops showed significant differences at key phonological dates. Maize consumed more water during the heading stage than at the V3 and maturity stages. There is a similar daily water consumption at the heading and maturity stages for wheat, higher than that at the green-up or emergence stages. Daily evapotranspiration water consumption of early rice increased gradually at the transplanting, heading, and maturity stages, but the opposite trend for late rice. Single-rice had the highest daily evapotranspiration at the heading stage, followed by the transplanting stage. Despite the inherent uncertainties associated with remote sensing-based evapotranspiration and phenology, the advantages of remote sensing in monitoring evapotranspiration water consumption over large areas are noteworthy. This information on crop water consumption can provide a scientific foundation for devising more precise irrigation strategies and planting systems, ultimately contributing to food security and effective water resource management.关键词：remote sensing;evapotranspiration water consumption;three staple crops;key phonological dates;Spatial and temporal variability- 506
- |
- 41
- |
- 0
更新时间：2024-08-21 - “云南省红河州冰雹灾害时空分布特征研究取得新进展，采用多源遥感卫星影像时空融合方法，提高了冰雹灾害监测精度与风险评估精度，为减少冰雹灾害损失提供重要参考。”
摘要：Hail has occurred frequently and caused significant losses to local agricultural production in Honghe Prefecture, Yunnan Province, Since 1961. The hail disasters distribution data at the county scale or weather station scale, which obtained by using statistical analysis method, cannot meet the needs of agricultural hail prevention. Some hail disaster remote sensing monitoring methods, which limited by single remote sensing data sources and the characteristics of designing for global scale, lacks applicability in mountainous areas. In order to grasp the spatial and temporal distribution characteristics of hail, and build hail remote sensing monitoring model at parcel level. The paper used hail record data from hail suppression operation stations since 2009 to 2022, and research on multi-source data fusion approach based on Ross Li and STARFM, propose a multi-level grid normalized vegetation index standardization model and hail remote sensing monitoring recognition index RNDVI_M. Use Kneed method to extract trend turning points of RNDVI_M as the thresholds to extract hail disasters area. Then applied the phenomenon universality verification method to verify the effectiveness of the RNDVI_M threshold and evaluate the accuracy of hail monitoring Based on hail survey data from 2009 to 2022, and found the maximum relative error is 9.08%, the average error is 5.62%, and the standard deviation is 1.66%. Use spatial overlay analysis and spatial correlation analysis method to quantitative analysis hail frequency in different disaster-prone environments such as landform types, terrain undulations, slopes, and terrain types at the level of cultivated land plots. Propose hail disaster risk assessment model to calculate the spatial distribution characteristics of hail risk caused by natural conditions such as climate, meteorology, terrain, and topography. The advantages are (1) comprehensively utilized multi-source remote sensing data to reconstruct RNDVI_M time series data to address the limitations of poor satellite data quality in mountainous areas with frequent clouds and rain. (2) Use parameter adaptation for multi-level grid models to improve model adaptability in three-dimensional climate conditions of mountainous areas and improved the accuracy of hail monitoring and risk assessment from county scale to cultivated land plots scale. (3) Spatial correlation quantitative analysis was conducted between the spatial distribution of hail disasters and terrain such as altitude, slope, undulation, river valleys, valleys, and ridges at the scale of cultivated land plots, found that hail disasters in mountainous areas is significantly correlated with altitude, and has moderate correlation with slope and undulation. Hailstones usually move along mountain ranges and valleys, making farmland along these valleys more susceptible to hail disasters. (4) Constructed the hail susceptibility assessment model at cultivated land plots scale, and the research results contribute to the rational adjustment of crop planting structure, the planning and layout of artificial hail control operation points, and the reduction of hail disaster losses.关键词：Hail disaster;Hail Remote Sensing Identification Index (RNDVI_M);Hail Disasters Remote Sensing Monitoring;Temporal and spatial distribution of hail disasters;Honghe- 498
- |
- 29
- |
- 0
更新时间：2024-08-21 - “最新研究揭示了寒温带针叶林蓄积量估测的新方法，通过分析多波段极化SAR数据，发现P波段对蓄积量敏感性最高，为森林资源监测提供了新的技术手段。”
摘要：Objective Research investigating the estimation ability of forest stock volume combining multi-band polarimetric SAR (PolSAR) has hardly been explored, particularly the complementarity between long wavelengths such as P-band and other shorter wavelengths. This study takes cold temperate coniferous forests in Inner Mongolia as the research object. Having available a multi-band stack of airborne P-, L-, S-, C-, and X-band PolSAR data acquired by the high-resolution airborne Multidimensional Space Joint-observation SAR (MSJosSAR) system, the aim is to systematically analyze the response and sensitivity of polarimetric characteristics in different bands to forest stock and evaluate the performance of forest stock retrieval using single and multi-band PolSAR data.Method Firstly, geocoding and terrain radiometric correction were performed on multi-band PolSAR data, and then a polarimetric feature set containing backscatter intensity and polarization decomposition components was extracted. Secondly, based on the water cloud model and correlation coefficient, the response law and sensitivity of polarimetric characteristics in different bands to forest stock was analyzed. Finally, machine learning algorithms were used to perform feature selection and modeling, and the ability of each band and jointly with multi-band to estimate forest stock was evaluated.Result The response of backscatter intensity in different bands to forest stock shows a similar upward trend, but the saturation point varies depending on wavelength and polarimetric channel. Among them, the saturation point for the P-band is higher than 160 m3/ha, whereas it does not exceed 110 m3/ha for the other bands. In addition, the correlation between forest stock and the P-band, L/S-band, and C/X-band decreases in order, with values above 0.6, between 0.3-0.4, and below 0.3, respectively. When estimating forest stock based on a single band, the accuracy of the P-band was 73.79%, and the accuracy of other bands did not exceed 60%. When using multi-band joint estimation, the estimation accuracy of L- or S-band and P-band joint estimation is about 2% higher than using P-band alone. The contribution of adding the C- or X-band to the accuracy improvement was minimal. The best estimation performance was achieved through the combination of all bands with an accuracy of 77.25%.Conclusion Taking into account various indicators such as signal dynamic range, saturation point, and correlation, the P-band exhibits the highest sensitivity to forest stock, followed by the L/S-band, and the C/X-band is the least sensitive. Therefore, when estimating forest stock using PolSAR data, the P-band should be the first choice. Additionally, when using multi-band joint estimation, the combination of P- and L- or S-band should be preferred. In recent years, long wavelength SAR satellites are being vigorously developed from both China and overseas, such as: China’s LT-1 satellite is already in orbit, ESA BIOMASS and NASA-ISRO NISAR missions are about to be launched, and China’s civil P-band SAR satellite has also entered the preliminary research stage. The above long wavelength SAR satellites will greatly enhance the estimation ability of regional forest stock in our country and provide strong support for the refined and scientific management of forest resources.关键词：multi-band SAR;polarimetric SAR;saturation point;forest stock;water cloud model- 468
- |
- 23
- |
- 0
更新时间：2024-08-21 -
SWSACNet： A change detection network for collapsed buildings extraction using multi-source images AI导读
“在遥感影像变化检测领域，研究者提出了一种新的网络模型SWSACNet，通过改进FTN模型，有效提高了倒塌建筑物的识别精度，为地震后建筑物损害评估提供了新方法。”
摘要：Objective Change detection networks based on deep learning are widely used in water monitor, urban change, etc.. However, collapsed buildings, as one of the change objectives, are rarely targeted for change detection networks. This study proposes an end-to-end collapsed building extraction model based on a change detection network including the sliding-window feature enhancement and convolution attention mix mechanism, which called SWSACNet (Sliding-Window-Shift Attention Convolution mix Network).MethodSWSACNet is an improvement of Fully Transformer Network (FTN). FTN is a network completely composed of Swin Transformer. Besides, it has a unique frame, which involves four parts: SFE (Siamese Feature Extraction), DFE (Deep Feature Enhancement), PCP (Progressive Change Prediction), DS (Deep Supervision). By encoding and decoding the feature of change objects deeply, FTN is able to learn what the collapsed buildings has changed in two temporal images and suppress irrelevant information. ACmix, a blend of convolution and attention mechanism, has been proved better performance than Swin Transformer in mainstream datasets. However, due to the different sensors, platforms, etc., the spatial heterogeneity of target features in different source remote sensing images will affect the accuracy of change detection. Concerning this problem, we designed a similarity sliding window to match the feature maps of two temporal images. Hence, we replace Swin Transformer with ACmix to extract and restore earthquake-damaged features efficiently in the phase of SFE and PCP, and using similarity sliding window to reduce misidentifications of collapsed buildings in different source image pairs before the phase of DFE. Result Taking the earthquake with 7.8 magnitude on February 6th, 2023, in Turkey as an example, establish a building seismic damage change detection dataset which consists of pre-earthquake Gaofen-2, Google images and post-earthquake Beijing-3 images, and collapsed buildings were extracted based on the SWSACNet, FTN, STANet based on the Siamese self-attention mechanism, DASNet based on a dual-attention fully-convolutional neural network, and the conventional fully-convolutional early fusion FC-EF network. The experimental results show that SWSACNet achieves the highest accuracy with F1 score of 80.8% and mIoU of 67.8%. The ablation experiments of the improved model indicates that SWSACNet obtains highest precision among three structure combinations. Beyond that, by smoothing the BJ-3 image that has higher spatial resolution to make the gradient change rate of image pairs closer, we acquire a new dataset and use it to retrain the five models. We found that the precision of five retrained models increases 1% at least, which also illustrates that appropriately narrowing the gap of gradient change rate of image pairs is an effective preprocessing for models to recognize the collapsed buildings. Finally, applying SWSACNet to three different data combinations covering Fevaipasa, Nurdagi and Islahiye area, the results show that it achieves 60.84% of average F1 score. Conclusion The application of SWSACNet indicates that the model needs richer pre- and post-earthquake training dataset and structural improvement to enhance its generalization.关键词：multi-source images;deep learning;change detection;collapsed building extraction- 613
- |
- 127
- |
- 0
更新时间：2024-07-16 - “在气象资料同化领域取得新突破，研究人员基于神经网络技术，成功构建了风云四号A星可见光反射率资料同化的快速观测算子NNFO，计算效率显著提升，为提高气象预报精度提供有力支持。”
摘要：Satellite all-sky visible reflectance contain critical information on cloud and precipitation. Therefore, data assimilation (DA) of these data has great potential to improve the forecasting skills of numerical weather prediction (NWP) models. Conventional forward operators for the DA of visible reflectance are based on numerical methods to simulate the radiative transferring processes and suffer from high computational burden. Therefore, conventional forward operators cannot meet the needs for operational DA.Objective The study is designed to construct a fast and accurate forward operator for the DA of visible reflectance data provided by the Advanced Geostationary Radiation Imager (AGRI) onboard the Fengyun-4A satellite. The forward operator should be comparable to conventional forward operators in accuracy and overwhelms the latter in computational efficiency.Method A feed-forward neural network was utilized to construct the forward operator. The input parameters of the neural network-based forward operator (NNFO) include cloud water path converted into logarithmic space, the mixing ratio of ice cloud water path and the total cloud water path, the effective radius of cloud liquid droplets, underlying surface albedo, solar zenith angle, satellite zenith angle, and relative azimuth angle between the sun and the satellite. The output of NNFO is the top-of-atmosphere reflectance. A series of sensitivity studies were performed to find the optimal (or at least, the sub-optimal) neural network settings, which include 5 hidden layers, 57 nodes in each hidden layer, the “swish” activation function for the hidden layers, and 512 batch size. In addition, the neural network is trained with an adaptive learning rate depending on the training epoch and the loss for the validation dataset, which was defined by the root mean square error (RMSE).ResultNNFO is compared with RTTOV-DOM, a typical forward operator based on discrete ordinate method to simulate the radiative transfer processes. The results indicate that NNFO is 15 or 6 times faster than RTTOV-DOM for serial or parallel modes. The mean difference, RMSE, and mean absolute error of the difference of reflectance simulated by RTTOV-DOM and NNFO (RTTOV-DOM minus NNFO) is 0.001, 0.048, 0.029, implying that the simulation accuracy between the two forward operators are comparable to each other. In addition, NNFO is validated by FY-4A/AGRI one-week reflectance observations. The results revealed that the probability density function of the simulation errors conforms to a Gaussian function, with the mean bias and standard deviation of -0.016 and 0.052, respectively.ConclusionNNFO is comparable to traditional forward operators in accuracy with a distinguished advantage in computational efficiency. Nevertheless, it is noteworthy that the current version of NNFO only supports the Ensemble Kalman Filter methods (including its variants). When it comes to the four-dimensional variational methods, NNFO should be developed by including its adjoint. In addition, NNFO could be further improved by including the aerosol effects. Improving NNFO in the aforementioned aspects and extending NNFO to DA applications are ongoing.关键词：Fengyun-4A satellite;Advanced Geostationary Radiation Imager (AGRI);visible reflectance;RTTOV;data assimilation;forward operator;neural network;adaptive learning rate- 531
- |
- 98
- |
- 0
更新时间：2024-07-16
- Postal code：100190
- Tel：010-58887052 Email：nrsb@aircas.ac.cn
- Technical support is provided by Beijing Founder electronics co., LTD 京ICP备20021838号-9
京公网安备11010802024621 - It is recommended to read the content of this site in Chrome&IE9+. Please switch to extreme mode in browser 360.
- Cookies We use cookies to help provide and enhance our service and tailor content. By continuing, you agree to the use of cookies.
0