This paper addresses the accuracy and timeliness limitations of traditional comprehensive prediction methods by proposing an approach of decision-level fusion of multisource data.A risk prediction indicator system was...This paper addresses the accuracy and timeliness limitations of traditional comprehensive prediction methods by proposing an approach of decision-level fusion of multisource data.A risk prediction indicator system was established for water and mud inrush in tunnels by analyzing advanced prediction data for specifi c tunnel segments.Additionally,the indicator weights were determined using the analytic hierarchy process combined with the Huber weighting method.Subsequently,a multisource data decision-layer fusion algorithm was utilized to generate fused imaging results for tunnel water and mud inrush risk predictions.Meanwhile,risk analysis was performed for different tunnel sections to achieve spatial and temporal complementarity within the indicator system and optimize redundant information.Finally,model feasibility was validated using the CZ Project Sejila Mountain Tunnel segment as a case study,yielding favorable risk prediction results and enabling effi cient information fusion and support for construction decision-making.展开更多
This research investigates the application of multisource data fusion using a Multi-Layer Perceptron (MLP) for Human Activity Recognition (HAR). The study integrates four distinct open-source datasets—WISDM, DaLiAc, ...This research investigates the application of multisource data fusion using a Multi-Layer Perceptron (MLP) for Human Activity Recognition (HAR). The study integrates four distinct open-source datasets—WISDM, DaLiAc, MotionSense, and PAMAP2—to develop a generalized MLP model for classifying six human activities. Performance analysis of the fused model for each dataset reveals accuracy rates of 95.83 for WISDM, 97 for DaLiAc, 94.65 for MotionSense, and 98.54 for PAMAP2. A comparative evaluation was conducted between the fused MLP model and the individual dataset models, with the latter tested on separate validation sets. The results indicate that the MLP model, trained on the fused dataset, exhibits superior performance relative to the models trained on individual datasets. This finding suggests that multisource data fusion significantly enhances the generalization and accuracy of HAR systems. The improved performance underscores the potential of integrating diverse data sources to create more robust and comprehensive models for activity recognition.展开更多
Urban geography has always been concerned about the influence of human settlements on urban vitality,but few studies reveal the influence of human settlements on urban vitality at a micro-scale.This paper analyzes the...Urban geography has always been concerned about the influence of human settlements on urban vitality,but few studies reveal the influence of human settlements on urban vitality at a micro-scale.This paper analyzes the spatial distribution characteristics of human settlements’quality and urban vitality at the micro-scale using Geodetectors and geographic weighted regression to analyze the relationship between human settlements and urban vitality.The results are shown as follows:there is still a significant development space for human settlements quality in Shahekou District,with obvious spatial dependence characteristics and significant gaps between various systems;the urban vitality of Shahekou District has obvious timeliness,and the urban vitality undergoes significant changes over time,which is related to the human settlements quality.The spatial distribution presents a single core spatial distribution structure with strong relative stability.The spatial distribution of cold and hot spots shows a pattern of“high in the north and low in the south,high in the east and low in the west”,with an increasing trend from southwest to northeast;the reachability of public transport has a significant impact on urban vitality.Its synergy with other variables is the leading force forming the spatial distribution of urban vitality.The environmental system,support system and social system are the significant factors affecting the urban vitality of Shahekou District.展开更多
Urban areas are particularly vulnerable to surface water flooding in a changing environment.A large number of urban surface water flood models have been developed to derive flood inundations and support risk managemen...Urban areas are particularly vulnerable to surface water flooding in a changing environment.A large number of urban surface water flood models have been developed to derive flood inundations and support risk management.However,unlike fluvial and coastal flooding,urban pluvial flooding is often associated with shallow water and thus the model is difficult to validate with traditional monitoring data.In this study,we first developed a full two-dimensional(2D)hydrodynamic model for simulating surface water floods.We further evaluated the model performance with multisource data from flood incidents,including official reports and social media data.The model was tested in the cities of Baoji and Linyi,China,where two surface water flood events recently occurred and caused considerable losses and casualties.In total,350 localized flooding incidents were obtained for the two cities(220 in Baoji and 130 in Linyi)and 313 reports were retained after data cleaning(202 in Baoji and 111 in Linyi).Over 90%of the reported flood incidents fall in urban areas where water depths are predicted to be higher than 0.15 m.The results demonstrate that the model is able to derive the broad patterns of flood inundation at the city scale.The approach tested here could be applied to other flood-prone cities and future research could include water depth information for more robust model validation.展开更多
Effective fault diagnosis has a crucial impact on the safety and cost of complex manufacturing systems.However,the complex structure of the collected multisource data and scarcity of fault samples make it difficult to...Effective fault diagnosis has a crucial impact on the safety and cost of complex manufacturing systems.However,the complex structure of the collected multisource data and scarcity of fault samples make it difficult to accurately identify multiple fault conditions.To address this challenge,this paper proposes a novel deep-learning model for multisource data augmentation and small sample fault diagnosis.The raw multisource data are first converted into two-dimensional images using the Gramian Angular Field,and a generator is built to transform random noise into images through transposed convolution operations.Then,two discriminators are constructed to evaluate the authenticity of input images and the fault diagnosis ability.The Vision Transformer network is built to diagnose faults and obtain the classification error for the discriminator.Furthermore,a global optimization strategy is designed to upgrade parameters in the model.The discriminators and generator compete with each other until Nash equilibrium is achieved.A real-world multistep forging machine is adopted to compare and validate the performance of different methods.The experimental results indicate that the proposed method has multisource data augmentation and minority sample fault diagnosis capabilities.Compared with other state-of-the-art models,the proposed approach has better fault diagnosis accuracy in various scenarios.展开更多
Poverty has always been a global concern that has restricted human development.The first goal(SDG 1)of the United Nations Sustainable Development Goals(SDGs)is to eliminate all forms of poverty all over the world.The ...Poverty has always been a global concern that has restricted human development.The first goal(SDG 1)of the United Nations Sustainable Development Goals(SDGs)is to eliminate all forms of poverty all over the world.The establishment of a scientific and effective localized SDG 1 evaluation and monitoring method is the key to achieving SDG 1.This paper proposes SDG 1 China district and county-level localization evaluation method based on multi-source remote sensing data for the United Nations Sustainable Development Framework.The temporal and spatial distribution characteristics of China’s poverty areas and their SDG 1 evaluation values in 2012,2014,2016,and 2018 have been analyzed.Based on the SDGs global indicator framework,this paper first constructed SDG 1 China’s district and county localization indicator system and then extracted multidimensional feature factors from nighttime light images,land cover data,and digital elevation model data.Secondly,we establish SDG 1 China’s localized partial least squares estimation model and SDG 1 China’s localized machine learning estimation model.Finally,we analyze and verify the spatiotemporal distribution characteristics of China’s poverty areas and counties and their SDG 1 evaluation values.The results show that SDG 1 China’s district and county localization indicator system proposed in this study and SDG 1 China’s localized partial least squares estimation model can better reflect the poverty level of China’s districts and counties.The estimated model R^(2) is 0.65,which can identify 72.77%of China’s national poverty counties.From 2012 to 2018,the spatial distribution pattern of SDG evaluation values in China’s districts and counties is that the SDG evaluation values gradually increase from western China to eastern China.In addition,the average SDG 1 evaluation value of China’s districts and counties increased by 23%from 2012 to 2018.This paper is oriented to the United Nations SDGs framework,explores the SDG 1 localized evaluation method of China’s districts and counties based on multisource remote sensing data,and provides a scientific and rapid regional poverty monitoring and evaluation program for the implementation of the 2030 agenda poverty alleviation goals.展开更多
Development of drought monitoring techniques is important for understanding and mitigating droughts and for rational agricultural management. This study used data from multiple sources, including MOD13 A3, TRMM 3 B43,...Development of drought monitoring techniques is important for understanding and mitigating droughts and for rational agricultural management. This study used data from multiple sources, including MOD13 A3, TRMM 3 B43, and SRTMDEM, for Yunnan Province, China from 2009 to 2018 to calculate the tropical rainfall condition index(TRCI), vegetation condition index(VCI), temperature condition index(TCI), and elevation factors. Principal component analysis(PCA) and analytic hierarchy process(AHP) were used to construct comprehensive drought monitoring models for Yunnan Province. The reliability of the models was verified, following which the drought situation in Yunnan Province for the past ten years was analysed. The results showed that:(1) The comprehensive drought index(CDI) had a high correlation with the standardized precipitation index, standardized precipitation evapotranspiration index, temperature vegetation dryness index, and CLDAS(China Meteorological Administration land data assimilation system), indicating that the CDI was a strong indicator of drought through meteorological, remote sensing and soil moisture monitoring.(2) The droughts from 2009 to 2018 showed generally consistent spatiotemporal changes. Droughts occurred in most parts of the province, with an average drought frequency of 29% and four droughtprone centres.(3) Monthly drought coverage during 2009 to 2014 exceeded that over 2015 to 2018. January had the largest average drought coverage over the study period(61.92%). Droughts at most stations during the remaining months except for October exhibited a weakening trend(slope > 0). The CDI provides a novel approach for drought monitoring in areas with complex terrain such as Yunnan Province.展开更多
The karst mountainous area is an ecologically fragile region with prominent humanland contradictions.The resource-environment carrying capacity(RECC)of this region needs to be further clarified.The development of remo...The karst mountainous area is an ecologically fragile region with prominent humanland contradictions.The resource-environment carrying capacity(RECC)of this region needs to be further clarified.The development of remote sensing(RS)and geographic information system(GIS)provides data sources and processing platform for RECC monitoring.This study analyzed and established the evaluation index system of RECC by considering particularity in the karst mountainous area of Southwest China;processed multisource RS data(Sentinel-2,Aster-DEM and Landsat-8)to extract the spatial distributions of nine key indexes by GIS techniques(information classification,overlay analysis and raster calculation);proposed the methods of index integration and fuzzy comprehensive evaluation of the RECC by GIS;and took a typical area,Guangnan County in Yunnan Province of China,as an experimental area to explore the effectiveness of the indexes and methods.The results showed that:(1)The important indexes affecting the RECC of karst mountainous area are water resources,tourism resources,position resources,geographical environment and soil erosion environment.(2)Data on cultivated land,construction land,minerals,transportation,water conservancy,ecosystem services,topography,soil erosion and rocky desertification can be obtained from RS data.GIS techniques integrate the information into the RECC results.The data extraction and processing methods are feasible on evaluating RECC.(3)The RECC of Guangnan County was in the mid-carrying level in 2018.The midcarrying and low-carrying levels were the main types,accounting for more than 80.00%of the total study area.The areas with high carrying capacity were mainly distributed in the northern regions of the northwest-southeast line of the county,and other areas have a low carrying capacity comparatively.The coordination between regional resource-environment status and socioeconomic development is the key to improve RECC.This study explores the evaluation index system of RECC in karst mountainous area and the application of multisource RS data and GIS techniques in the comprehensive evaluation.The methods can be applied in related fields to provide suggestions for data/information extraction and integration,and sustainable development.展开更多
To meet the increasing demand of national spatial database infrastructure construction and application, a concept model of China's coastal zone scientific data platform is established based on the information feat...To meet the increasing demand of national spatial database infrastructure construction and application, a concept model of China's coastal zone scientific data platform is established based on the information feature analysis of a compound dataset, consisting of remote sensing data and conventional data. Based on this concept model, the detailed logical database structure and the storage strategy of remote sensing data and their metadata using ArcSDE are designed. The complicated technology of multisources data combination in this research is crucial to the future coastal zone and offshore database construction and practical running, which will provide intelligent information analysis and technological service for coastal zone and offshore investigation, research, development and management.展开更多
Poverty threatens human development especially for developing countries,so ending poverty has become one of the most important United Nations Sustainable Development Goals(SDGs).This study aims to explore China’s pro...Poverty threatens human development especially for developing countries,so ending poverty has become one of the most important United Nations Sustainable Development Goals(SDGs).This study aims to explore China’s progress in poverty reduction from 2016 to 2019 through time-series multi-source geospatial data and a deep learning model.The poverty reduction efficiency(PRE)is measured by the difference in the out-of-poverty rates(which measures the probability of being not poor)of 2016 and 2019.The study shows that the probability of poverty in all regions of China has shown an overall decreasing trend(PRE=0.264),which indicates that the progress in poverty reduction during this period is significant.The Hu Huanyong Line(Hu Line)shows an uneven geographical pattern of out-of-poverty rate between Southeast and Northwest China.From 2016 to 2019,the centroid of China’s out-of-poverty rate moved 105.786 km to the northeast while the standard deviation ellipse of the out-of-poverty rate moved 3 degrees away from the Hu Line,indicating that the regions with high out-of-poverty rates are more concentrated on the east side of the Hu Line from 2016 to 2019.The results imply that the government’s future poverty reduction policies should pay attention to the infrastructure construction in poor areas and appropriately increase the population density in poor areas.This study fills the gap in the research on poverty reduction under multiple scales and provides useful implications for the government’s poverty reduction policy.展开更多
We developed a forest type classification technology for the Daxing＇an Mountains of northeast China using multisource remote sensing data.A SPOT-5 image and two temporal images of RADARSAT-2 full-polarization SAR wer...We developed a forest type classification technology for the Daxing＇an Mountains of northeast China using multisource remote sensing data.A SPOT-5 image and two temporal images of RADARSAT-2 full-polarization SAR were used to identify forest types in the Pangu Forest Farm of the Daxing＇an Mountains.Forest types were identified using random forest（RF） classification with the following data combination types： SPOT-5 alone,SPOT-5 and SAR images in August or November,and SPOT-5 and two temporal SAR images.We identified many forest types using a combination of multitemporal SAR and SPOT-5 images,including Betula platyphylla,Larix gmelinii,Pinus sylvestris and Picea koraiensis forests.The accuracy of classification exceeded 88% and improved by 12% when compared to the classification results obtained using SPOT data alone.RF classification using a combination of multisource remote sensing data improved classification accuracy compared to that achieved using single-source remote sensing data.展开更多
With the improvement of multisource information sensing and data acquisition capabilities inside tunnels,the availability of multimodal data in tunnel engineering has significantly increased.However,due to structural ...With the improvement of multisource information sensing and data acquisition capabilities inside tunnels,the availability of multimodal data in tunnel engineering has significantly increased.However,due to structural differences in multimodal data,traditional intelligent advanced geological prediction models have limited capacity for data fusion.Furthermore,the lack of pre-trained models makes it difficult for neural networks trained from scratch to deeply explore the features of multimodal data.To address these challenges,we utilize the fusion capability of knowledge graph for multimodal data and the pre-trained knowledge of large language models(LLMs)to establish an intelligent advanced geological prediction model(GeoPredict-LLM).First,we develop an advanced geological prediction ontology model,forming a knowledge graph database.Using knowledge graph embeddings,multisource and multimodal data are transformed into low-dimensional vectors with a unified structure.Secondly,pre-trained LLMs,through reprogramming,reconstruct these low-dimensional vectors,imparting linguistic characteristics to the data.This transformation effectively reframes the complex task of advanced geological prediction as a"language-based"problem,enabling the model to approach the task from a linguistic perspective.Moreover,we propose the prompt-as-prefix method,which enables output generation,while freezing the core of the LLM,thereby significantly reduces the number of training parameters.Finally,evaluations show that compared to neural network models without pre-trained models,GeoPredict-LLM significantly improves prediction accuracy.It is worth noting that as long as a knowledge graph database can be established,GeoPredict-LLM can be adapted to multimodal data mining tasks with minimal modifications.展开更多
Based on station precipitation observations,radar quantitative precipitation estimates(QPE), and radar fusion data during Typhoon Fitow(2013), the influence of multisource precipitation data on multiscale urban typhoo...Based on station precipitation observations,radar quantitative precipitation estimates(QPE), and radar fusion data during Typhoon Fitow(2013), the influence of multisource precipitation data on multiscale urban typhoon pluvial flood modeling is studied. Using Shanghai, China,as the study area, a simplified 2D hydrodynamic model is applied to simulations. Combined with actual flood incidents reported by the public and soil moisture data, we perform multiscale verifications and determine the applicability of three precipitation datasets in the modeling. The results are as follows:(1) At the city scale, although QPE have higher spatial resolution, these estimates are lower than station observations. Radar fusion data have both high accuracy and high spatial resolution. For flood depths above 5 cm, the radar fusion precipitation scenario can improve the matching probability by 6%.(2) At the neighborhood scale, the radar fusion precipitation scenario can effectively mitigate the problems of an uneven spatial distribution of stations and a weak QPE to accurately capture pluvial details.(3)One fixed-point assessment shows that different precipitation data have little influence on the temporal characteristics of the modeling result-all three types of data can accurately reflect flood occurrence times. This work can provide a scientific basis for constructing effective urban pluvial flood monitoring systems.展开更多
To analyze the effects of gas cannons on clouds and precipitation,multisource observational data,including those from National Centers for Environmental Prediction(NCEP)reanalysis,Hangzhou and Huzhou new-generation we...To analyze the effects of gas cannons on clouds and precipitation,multisource observational data,including those from National Centers for Environmental Prediction(NCEP)reanalysis,Hangzhou and Huzhou new-generation weather radars,laser disdrometer,ground-based automatic weather station,wind profiler radar,and Lin'an C-band dualpolarization radar,were adopted in this study.Based on the variational dual-Doppler wind retrieval method and the polarimetric variables obtained by the dual-polarization radar,we analyzed the microphysical processes and the variations in the macro-and microphysical quantities in clouds from the perspective of the synoptic background before precipitation enhancement,the polarization echo characteristics before,during and after enhancement,and the evolution of the fine three-dimensional kinematic structure and the microphysical structure.The results show that the precipitation enhancement operation promoted the development of radar echoes and prolonged their duration,and both the horizontal and vertical wind speeds increased.The dual-polarization radar echo showed that the diameter of the precipitation particles increased,and the concentration of raindrops increased after precipitation enhancement.The raindrops were lifted to a height corresponding to 0 to-20℃due to vertical updrafts.Based on the disdrometer data during precipitation enhancement,the concentration of small raindrops(lgN_(w))showed a significant increase,and the mass-weighted diameter D_(m)value decreased,indicating that the precipitation enhancement operation played a certain“lubricating”effect.After the precipitation enhancement,the concentration of raindrops did not change much compared with that during the enhancement process,while the Dm increased,corresponding to an increase in rain intensity.The results suggest the positive effect of gas cannons on precipitation enhancement.展开更多
Thunderstorm gusts are a common and hazardous type of severe convective weather,characterized by a small spatial scale,short duration,and significant destructive power.They often lead to severe disasters,highlighting ...Thunderstorm gusts are a common and hazardous type of severe convective weather,characterized by a small spatial scale,short duration,and significant destructive power.They often lead to severe disasters,highlighting the critical importance of their accurate forecasting.Previous studies have explored the environmental factors and spatiotemporal distribution characteristics of thunderstorm gusts,highlighting the need for improved forecasting methods.In recent years,artificial intelligence techniques have shown promise in enhancing the accuracy of thunderstorm gust forecasting,with various machine learning algorithms and models having been developed.This paper proposes a multiscale feature fusion module called Thunderstorm Gusts Block(TG-Block)and a deep learning model named Thunderstorm Gusts net(TG-net)based on the Attention U-net and TG-TransUnet models,and employs interpretable methods such as Integrated Gradient,Deep Learning Importance Features,and Shapley Additive exPlanations to validate the model’s practical relevance and reliability.The analysis of feature importance underscores the model’s ability to capture key thermodynamic and multiscale weather characteristic information for thunderstorm gust nowcasting.It is,however,worth emphasizing that these conclusions are only based on a limited number of thunderstorm gust examples,and the evaluation results may be affected by specific weather types and sample sizes.Nonetheless,TG-net has been put into real-time operation at the Institute of Urban Meteorology,and we will continue to rigorously validate its performance and make any necessary optimizations and enhancements based on feedback to ensure the robustness and stability of the model.展开更多
Despite its essential importance to various spatial agriculture and environmental applications,the information on actual cropland area and its geographical distribution remain highly uncertain over Africa among remote...Despite its essential importance to various spatial agriculture and environmental applications,the information on actual cropland area and its geographical distribution remain highly uncertain over Africa among remote-sensing products.Each of the African regions has its unique physical and environmental limiting factors to accurate cropland mapping,which leads to high spatial discre-pancies among remote sensing cropland products.Since no dataset could cope with all limitations,multiple datasets initially derived from various remote sensing sensors and classification techniques must be integrated into a more accurate cropland product than individual layers.Here,in the current study,four cropland products,produced initially from multiple sensors(e.g.Landsat-8 OLI,Sentinel-2 MSI,and PROBA-V)to cover the period(2015-2017),were integrated based on their cropland mapping accuracy to build a more accurate cropland layer.The four cropland layers’accuracy was assessed at Agro-ecological zones units via an inten-sive reference dataset(17,592 samples).The most accurate crop-land layer was then identified for each zone to construct the final cropland mask at 30 m resolution for the nominal year of 2016 over Africa.As a result,the new layer was produced in higher cropland mapping accuracy(overall accuracy=91.64%and cropland’s F-score=0.75).The layer mapped the African cropland area as 282 Mha(9.38%of the Continent area).Compared to earlier crop-land synergy layers,the constructed cropland mask showed a considerable improvement in its spatial resolution(30 m instead of 250 m),mapping quality,and closeness to official statistics(R^(2)=0.853 and RMSE=2.85 Mha).The final layer can be down-loaded as described under the“Data Availability Statement”section.展开更多
The construction of human settlements plays a vital role in achieving sustainable development goals.With the growing population,both urban and rural planning—as well as the development of human settlements—on the Qi...The construction of human settlements plays a vital role in achieving sustainable development goals.With the growing population,both urban and rural planning—as well as the development of human settlements—on the Qinghai-Tibet Plateau in China have received increasing attention.This study proposed a comprehensive framework to assess the quality of human settlements and identify their limiting factors of Nyingchi City,southeastern Qinghai-Tibet Plateau of China.The framework integrated a region-specific evaluation system tailored to the plateau’s unique characteristics and utilized multi-source data collected through field-based questionnaires and remote sensing.Analytical methods employed include the Criteria Importance Through Intercriteria Correlation(CRITIC)weighting method,difference analysis for evaluating spatial and categorical variations,and a novel approach introduced in this study for identifying limiting factors.In August 2021,a questionnaire-based survey was conducted in the southeastern Qinghai-Tibet Plateau,yielding 823 valid responses.The average evaluation score for human settlement quality(HSQ)was(6.96±0.94),indicating that settlement conditions were approaching the satisfaction threshold.Notably,the score for the health dimension was(6.28±1.41),reflecting relatively underdeveloped health services in the region.From the perspective of spatial patterns,despite its favorable natural conditions,the HSQ in Medog County was relatively backward,reflecting uncoordinated development that warrants attention.Significant differences were observed in the perceptions of human settlements among different groups in terms of ethnicity,income,and educational background.Low ratings for income and employment conditions represented a major limitation to the local HSQ.The development of human settlements in the southeastern Qinghai-Tibet Plateau still needs further promotion due to population disparities and the challenges of creating a healthy environment in high-altitude regions.The results are valuable for improving human settlements in plateau regions,which can help make targeted policy recommendations.展开更多
基金supported by the National Natural Science Foundation of China (grant numbers 42293351, and U2468221)。
文摘This paper addresses the accuracy and timeliness limitations of traditional comprehensive prediction methods by proposing an approach of decision-level fusion of multisource data.A risk prediction indicator system was established for water and mud inrush in tunnels by analyzing advanced prediction data for specifi c tunnel segments.Additionally,the indicator weights were determined using the analytic hierarchy process combined with the Huber weighting method.Subsequently,a multisource data decision-layer fusion algorithm was utilized to generate fused imaging results for tunnel water and mud inrush risk predictions.Meanwhile,risk analysis was performed for different tunnel sections to achieve spatial and temporal complementarity within the indicator system and optimize redundant information.Finally,model feasibility was validated using the CZ Project Sejila Mountain Tunnel segment as a case study,yielding favorable risk prediction results and enabling effi cient information fusion and support for construction decision-making.
基金supported by the Royal Golden Jubilee(RGJ)Ph.D.Programme(Grant No.PHD/0079/2561)through the National Research Council of Thailand(NRCT)and Thailand Research Fund(TRF).
文摘This research investigates the application of multisource data fusion using a Multi-Layer Perceptron (MLP) for Human Activity Recognition (HAR). The study integrates four distinct open-source datasets—WISDM, DaLiAc, MotionSense, and PAMAP2—to develop a generalized MLP model for classifying six human activities. Performance analysis of the fused model for each dataset reveals accuracy rates of 95.83 for WISDM, 97 for DaLiAc, 94.65 for MotionSense, and 98.54 for PAMAP2. A comparative evaluation was conducted between the fused MLP model and the individual dataset models, with the latter tested on separate validation sets. The results indicate that the MLP model, trained on the fused dataset, exhibits superior performance relative to the models trained on individual datasets. This finding suggests that multisource data fusion significantly enhances the generalization and accuracy of HAR systems. The improved performance underscores the potential of integrating diverse data sources to create more robust and comprehensive models for activity recognition.
文摘Urban geography has always been concerned about the influence of human settlements on urban vitality,but few studies reveal the influence of human settlements on urban vitality at a micro-scale.This paper analyzes the spatial distribution characteristics of human settlements’quality and urban vitality at the micro-scale using Geodetectors and geographic weighted regression to analyze the relationship between human settlements and urban vitality.The results are shown as follows:there is still a significant development space for human settlements quality in Shahekou District,with obvious spatial dependence characteristics and significant gaps between various systems;the urban vitality of Shahekou District has obvious timeliness,and the urban vitality undergoes significant changes over time,which is related to the human settlements quality.The spatial distribution presents a single core spatial distribution structure with strong relative stability.The spatial distribution of cold and hot spots shows a pattern of“high in the north and low in the south,high in the east and low in the west”,with an increasing trend from southwest to northeast;the reachability of public transport has a significant impact on urban vitality.Its synergy with other variables is the leading force forming the spatial distribution of urban vitality.The environmental system,support system and social system are the significant factors affecting the urban vitality of Shahekou District.
基金supported by the National Natural Science Foundation of China(Grant Nos:42371076,42271089,42461160294)the Key Project of Shanghai Municipal Education Commission(Grant No:2024AI01006)Shanghai Pilot Program for Basic Research(Grant No:TQ20240209)。
文摘Urban areas are particularly vulnerable to surface water flooding in a changing environment.A large number of urban surface water flood models have been developed to derive flood inundations and support risk management.However,unlike fluvial and coastal flooding,urban pluvial flooding is often associated with shallow water and thus the model is difficult to validate with traditional monitoring data.In this study,we first developed a full two-dimensional(2D)hydrodynamic model for simulating surface water floods.We further evaluated the model performance with multisource data from flood incidents,including official reports and social media data.The model was tested in the cities of Baoji and Linyi,China,where two surface water flood events recently occurred and caused considerable losses and casualties.In total,350 localized flooding incidents were obtained for the two cities(220 in Baoji and 130 in Linyi)and 313 reports were retained after data cleaning(202 in Baoji and 111 in Linyi).Over 90%of the reported flood incidents fall in urban areas where water depths are predicted to be higher than 0.15 m.The results demonstrate that the model is able to derive the broad patterns of flood inundation at the city scale.The approach tested here could be applied to other flood-prone cities and future research could include water depth information for more robust model validation.
基金supported by“the Fundamental Research Funds for the Central Universities,”Grant/Award Number 30923011008.
文摘Effective fault diagnosis has a crucial impact on the safety and cost of complex manufacturing systems.However,the complex structure of the collected multisource data and scarcity of fault samples make it difficult to accurately identify multiple fault conditions.To address this challenge,this paper proposes a novel deep-learning model for multisource data augmentation and small sample fault diagnosis.The raw multisource data are first converted into two-dimensional images using the Gramian Angular Field,and a generator is built to transform random noise into images through transposed convolution operations.Then,two discriminators are constructed to evaluate the authenticity of input images and the fault diagnosis ability.The Vision Transformer network is built to diagnose faults and obtain the classification error for the discriminator.Furthermore,a global optimization strategy is designed to upgrade parameters in the model.The discriminators and generator compete with each other until Nash equilibrium is achieved.A real-world multistep forging machine is adopted to compare and validate the performance of different methods.The experimental results indicate that the proposed method has multisource data augmentation and minority sample fault diagnosis capabilities.Compared with other state-of-the-art models,the proposed approach has better fault diagnosis accuracy in various scenarios.
基金supported by the National Natural Science Foundation of China[grant numbers 41971423 and 31972951]the Natural Science Foundation of Hunan Province[grant numbers 2020JJ3020 and 2020JJ5164]+1 种基金the Science and Technology Planning Project of Hunan Province[grant numbers 2019RS2043 and 2019GK2132]the Postgraduate Scientific Research Innovation Project of Hunan Province[grant number CX20210991].
文摘Poverty has always been a global concern that has restricted human development.The first goal(SDG 1)of the United Nations Sustainable Development Goals(SDGs)is to eliminate all forms of poverty all over the world.The establishment of a scientific and effective localized SDG 1 evaluation and monitoring method is the key to achieving SDG 1.This paper proposes SDG 1 China district and county-level localization evaluation method based on multi-source remote sensing data for the United Nations Sustainable Development Framework.The temporal and spatial distribution characteristics of China’s poverty areas and their SDG 1 evaluation values in 2012,2014,2016,and 2018 have been analyzed.Based on the SDGs global indicator framework,this paper first constructed SDG 1 China’s district and county localization indicator system and then extracted multidimensional feature factors from nighttime light images,land cover data,and digital elevation model data.Secondly,we establish SDG 1 China’s localized partial least squares estimation model and SDG 1 China’s localized machine learning estimation model.Finally,we analyze and verify the spatiotemporal distribution characteristics of China’s poverty areas and counties and their SDG 1 evaluation values.The results show that SDG 1 China’s district and county localization indicator system proposed in this study and SDG 1 China’s localized partial least squares estimation model can better reflect the poverty level of China’s districts and counties.The estimated model R^(2) is 0.65,which can identify 72.77%of China’s national poverty counties.From 2012 to 2018,the spatial distribution pattern of SDG evaluation values in China’s districts and counties is that the SDG evaluation values gradually increase from western China to eastern China.In addition,the average SDG 1 evaluation value of China’s districts and counties increased by 23%from 2012 to 2018.This paper is oriented to the United Nations SDGs framework,explores the SDG 1 localized evaluation method of China’s districts and counties based on multisource remote sensing data,and provides a scientific and rapid regional poverty monitoring and evaluation program for the implementation of the 2030 agenda poverty alleviation goals.
基金This research was funded by the Multigovernment International Science and Technology Innovation Cooperation Key Project of the National Key Research and Development Program of China(Grant No.2018YFE0184300)Erasmus+Capacity Building in Higher Education of the Education,Audiovisual and Culture Executive Agency(EACEA)(Grant No.586037-EPP-1-2017-1-HU-EPPKA2CBHE-JP)+3 种基金the National Natural Science Foundation of China(Grant No.41561048)the Technical Methods and Empirical Study on Ecological Assets Measurement in County Level of Yunnan Province(Grant No.ZDZZD201506)the Young and Middleaged Academic and Technical Leaders Reserve Talents Training Program of Yunnan Province(Grant No.2008PY056)the Program for Innovative Research Team(in Science and Technology)at the University of Yunnan Province,IRTSTYN。
文摘Development of drought monitoring techniques is important for understanding and mitigating droughts and for rational agricultural management. This study used data from multiple sources, including MOD13 A3, TRMM 3 B43, and SRTMDEM, for Yunnan Province, China from 2009 to 2018 to calculate the tropical rainfall condition index(TRCI), vegetation condition index(VCI), temperature condition index(TCI), and elevation factors. Principal component analysis(PCA) and analytic hierarchy process(AHP) were used to construct comprehensive drought monitoring models for Yunnan Province. The reliability of the models was verified, following which the drought situation in Yunnan Province for the past ten years was analysed. The results showed that:(1) The comprehensive drought index(CDI) had a high correlation with the standardized precipitation index, standardized precipitation evapotranspiration index, temperature vegetation dryness index, and CLDAS(China Meteorological Administration land data assimilation system), indicating that the CDI was a strong indicator of drought through meteorological, remote sensing and soil moisture monitoring.(2) The droughts from 2009 to 2018 showed generally consistent spatiotemporal changes. Droughts occurred in most parts of the province, with an average drought frequency of 29% and four droughtprone centres.(3) Monthly drought coverage during 2009 to 2014 exceeded that over 2015 to 2018. January had the largest average drought coverage over the study period(61.92%). Droughts at most stations during the remaining months except for October exhibited a weakening trend(slope > 0). The CDI provides a novel approach for drought monitoring in areas with complex terrain such as Yunnan Province.
基金the support given by the government and official in Guangnan Countyfunded by[National Natural Science Foundation of China]grant number[41361020,40961031]+3 种基金[Joint Fund of Yunnan Provincial Science and Technology Department and Yunnan University]grant number[2018FY001(-017)][Project of Innovative Talents Cultivation for Graduate Students of Yunnan University]grant number[C176230200][Project of Internationalization and Cultural Inheritance and Innovation of Yunnan University]grant number[C176250202][Science Research Fund of Yunnan Provincial Education Department in 2020:Postgraduate]grant number[2020Y0030]。
文摘The karst mountainous area is an ecologically fragile region with prominent humanland contradictions.The resource-environment carrying capacity(RECC)of this region needs to be further clarified.The development of remote sensing(RS)and geographic information system(GIS)provides data sources and processing platform for RECC monitoring.This study analyzed and established the evaluation index system of RECC by considering particularity in the karst mountainous area of Southwest China;processed multisource RS data(Sentinel-2,Aster-DEM and Landsat-8)to extract the spatial distributions of nine key indexes by GIS techniques(information classification,overlay analysis and raster calculation);proposed the methods of index integration and fuzzy comprehensive evaluation of the RECC by GIS;and took a typical area,Guangnan County in Yunnan Province of China,as an experimental area to explore the effectiveness of the indexes and methods.The results showed that:(1)The important indexes affecting the RECC of karst mountainous area are water resources,tourism resources,position resources,geographical environment and soil erosion environment.(2)Data on cultivated land,construction land,minerals,transportation,water conservancy,ecosystem services,topography,soil erosion and rocky desertification can be obtained from RS data.GIS techniques integrate the information into the RECC results.The data extraction and processing methods are feasible on evaluating RECC.(3)The RECC of Guangnan County was in the mid-carrying level in 2018.The midcarrying and low-carrying levels were the main types,accounting for more than 80.00%of the total study area.The areas with high carrying capacity were mainly distributed in the northern regions of the northwest-southeast line of the county,and other areas have a low carrying capacity comparatively.The coordination between regional resource-environment status and socioeconomic development is the key to improve RECC.This study explores the evaluation index system of RECC in karst mountainous area and the application of multisource RS data and GIS techniques in the comprehensive evaluation.The methods can be applied in related fields to provide suggestions for data/information extraction and integration,and sustainable development.
基金the“863”Marine Monitor of Hitech Research and Development Program of China under contract No.,5 2003AA604040 a, 2002AA639640.
文摘To meet the increasing demand of national spatial database infrastructure construction and application, a concept model of China's coastal zone scientific data platform is established based on the information feature analysis of a compound dataset, consisting of remote sensing data and conventional data. Based on this concept model, the detailed logical database structure and the storage strategy of remote sensing data and their metadata using ArcSDE are designed. The complicated technology of multisources data combination in this research is crucial to the future coastal zone and offshore database construction and practical running, which will provide intelligent information analysis and technological service for coastal zone and offshore investigation, research, development and management.
基金supported by the National Key Research and Development Program of China[grant number 2019YFB2102903]the National Natural Science Foundation of China[grant number 41801306]+1 种基金the“CUG Scholar”Scientific Research Funds at China University of Geosciences(Wuhan)[grant number 2022034]a grant from State Key Laboratory of Resources and Environmental Information System.
文摘Poverty threatens human development especially for developing countries,so ending poverty has become one of the most important United Nations Sustainable Development Goals(SDGs).This study aims to explore China’s progress in poverty reduction from 2016 to 2019 through time-series multi-source geospatial data and a deep learning model.The poverty reduction efficiency(PRE)is measured by the difference in the out-of-poverty rates(which measures the probability of being not poor)of 2016 and 2019.The study shows that the probability of poverty in all regions of China has shown an overall decreasing trend(PRE=0.264),which indicates that the progress in poverty reduction during this period is significant.The Hu Huanyong Line(Hu Line)shows an uneven geographical pattern of out-of-poverty rate between Southeast and Northwest China.From 2016 to 2019,the centroid of China’s out-of-poverty rate moved 105.786 km to the northeast while the standard deviation ellipse of the out-of-poverty rate moved 3 degrees away from the Hu Line,indicating that the regions with high out-of-poverty rates are more concentrated on the east side of the Hu Line from 2016 to 2019.The results imply that the government’s future poverty reduction policies should pay attention to the infrastructure construction in poor areas and appropriately increase the population density in poor areas.This study fills the gap in the research on poverty reduction under multiple scales and provides useful implications for the government’s poverty reduction policy.
基金supported by the National Natural Science Foundation of China(Nos.31500518,31500519,and 31470640)
文摘We developed a forest type classification technology for the Daxing＇an Mountains of northeast China using multisource remote sensing data.A SPOT-5 image and two temporal images of RADARSAT-2 full-polarization SAR were used to identify forest types in the Pangu Forest Farm of the Daxing＇an Mountains.Forest types were identified using random forest（RF） classification with the following data combination types： SPOT-5 alone,SPOT-5 and SAR images in August or November,and SPOT-5 and two temporal SAR images.We identified many forest types using a combination of multitemporal SAR and SPOT-5 images,including Betula platyphylla,Larix gmelinii,Pinus sylvestris and Picea koraiensis forests.The accuracy of classification exceeded 88% and improved by 12% when compared to the classification results obtained using SPOT data alone.RF classification using a combination of multisource remote sensing data improved classification accuracy compared to that achieved using single-source remote sensing data.
基金the National Natural Science Foundation of China(Grant Nos.52279103 and 52379103)。
文摘With the improvement of multisource information sensing and data acquisition capabilities inside tunnels,the availability of multimodal data in tunnel engineering has significantly increased.However,due to structural differences in multimodal data,traditional intelligent advanced geological prediction models have limited capacity for data fusion.Furthermore,the lack of pre-trained models makes it difficult for neural networks trained from scratch to deeply explore the features of multimodal data.To address these challenges,we utilize the fusion capability of knowledge graph for multimodal data and the pre-trained knowledge of large language models(LLMs)to establish an intelligent advanced geological prediction model(GeoPredict-LLM).First,we develop an advanced geological prediction ontology model,forming a knowledge graph database.Using knowledge graph embeddings,multisource and multimodal data are transformed into low-dimensional vectors with a unified structure.Secondly,pre-trained LLMs,through reprogramming,reconstruct these low-dimensional vectors,imparting linguistic characteristics to the data.This transformation effectively reframes the complex task of advanced geological prediction as a"language-based"problem,enabling the model to approach the task from a linguistic perspective.Moreover,we propose the prompt-as-prefix method,which enables output generation,while freezing the core of the LLM,thereby significantly reduces the number of training parameters.Finally,evaluations show that compared to neural network models without pre-trained models,GeoPredict-LLM significantly improves prediction accuracy.It is worth noting that as long as a knowledge graph database can be established,GeoPredict-LLM can be adapted to multimodal data mining tasks with minimal modifications.
基金This study was sponsored by the National Natural Science Foundation of China(Grant Nos.41871164,41806046)the Shanghai Sailing Program(Grant No.21YF1456900)+1 种基金the Shanghai Philosophy and Social Science Planning Program(Grant No.2021XRM005)the Fundamental Research Funds for the Central Universities(Grant No.2022ECNU-XWK-XK001).
文摘Based on station precipitation observations,radar quantitative precipitation estimates(QPE), and radar fusion data during Typhoon Fitow(2013), the influence of multisource precipitation data on multiscale urban typhoon pluvial flood modeling is studied. Using Shanghai, China,as the study area, a simplified 2D hydrodynamic model is applied to simulations. Combined with actual flood incidents reported by the public and soil moisture data, we perform multiscale verifications and determine the applicability of three precipitation datasets in the modeling. The results are as follows:(1) At the city scale, although QPE have higher spatial resolution, these estimates are lower than station observations. Radar fusion data have both high accuracy and high spatial resolution. For flood depths above 5 cm, the radar fusion precipitation scenario can improve the matching probability by 6%.(2) At the neighborhood scale, the radar fusion precipitation scenario can effectively mitigate the problems of an uneven spatial distribution of stations and a weak QPE to accurately capture pluvial details.(3)One fixed-point assessment shows that different precipitation data have little influence on the temporal characteristics of the modeling result-all three types of data can accurately reflect flood occurrence times. This work can provide a scientific basis for constructing effective urban pluvial flood monitoring systems.
基金National Natural Science Foundation of China(41675029)Postgraduate Research&Practice Innovation Program of Jiangsu Province(KYCX18_0998)+1 种基金Science and Technology Program of Huzhou(2021GZ14,2020GZ31)Science and Technology(Key)Program of Zhejiang Meteorological Service(2021ZD27)。
文摘To analyze the effects of gas cannons on clouds and precipitation,multisource observational data,including those from National Centers for Environmental Prediction(NCEP)reanalysis,Hangzhou and Huzhou new-generation weather radars,laser disdrometer,ground-based automatic weather station,wind profiler radar,and Lin'an C-band dualpolarization radar,were adopted in this study.Based on the variational dual-Doppler wind retrieval method and the polarimetric variables obtained by the dual-polarization radar,we analyzed the microphysical processes and the variations in the macro-and microphysical quantities in clouds from the perspective of the synoptic background before precipitation enhancement,the polarization echo characteristics before,during and after enhancement,and the evolution of the fine three-dimensional kinematic structure and the microphysical structure.The results show that the precipitation enhancement operation promoted the development of radar echoes and prolonged their duration,and both the horizontal and vertical wind speeds increased.The dual-polarization radar echo showed that the diameter of the precipitation particles increased,and the concentration of raindrops increased after precipitation enhancement.The raindrops were lifted to a height corresponding to 0 to-20℃due to vertical updrafts.Based on the disdrometer data during precipitation enhancement,the concentration of small raindrops(lgN_(w))showed a significant increase,and the mass-weighted diameter D_(m)value decreased,indicating that the precipitation enhancement operation played a certain“lubricating”effect.After the precipitation enhancement,the concentration of raindrops did not change much compared with that during the enhancement process,while the Dm increased,corresponding to an increase in rain intensity.The results suggest the positive effect of gas cannons on precipitation enhancement.
基金Supported by the National Key Research and Development Program of China(2022YFC3004103)Beijing Natural Science Foundation(8222051)+1 种基金China Meteorological Administration Key Innovation Team(CMA2022ZD04 and CMA2022ZD07)Nanjing Joint Institute for Atmospheric Sciences Beijige Open Research Fund(BJG202407).
文摘Thunderstorm gusts are a common and hazardous type of severe convective weather,characterized by a small spatial scale,short duration,and significant destructive power.They often lead to severe disasters,highlighting the critical importance of their accurate forecasting.Previous studies have explored the environmental factors and spatiotemporal distribution characteristics of thunderstorm gusts,highlighting the need for improved forecasting methods.In recent years,artificial intelligence techniques have shown promise in enhancing the accuracy of thunderstorm gust forecasting,with various machine learning algorithms and models having been developed.This paper proposes a multiscale feature fusion module called Thunderstorm Gusts Block(TG-Block)and a deep learning model named Thunderstorm Gusts net(TG-net)based on the Attention U-net and TG-TransUnet models,and employs interpretable methods such as Integrated Gradient,Deep Learning Importance Features,and Shapley Additive exPlanations to validate the model’s practical relevance and reliability.The analysis of feature importance underscores the model’s ability to capture key thermodynamic and multiscale weather characteristic information for thunderstorm gust nowcasting.It is,however,worth emphasizing that these conclusions are only based on a limited number of thunderstorm gust examples,and the evaluation results may be affected by specific weather types and sample sizes.Nonetheless,TG-net has been put into real-time operation at the Institute of Urban Meteorology,and we will continue to rigorously validate its performance and make any necessary optimizations and enhancements based on feedback to ensure the robustness and stability of the model.
基金was funded by the Strategic Priority Research Program of the Chinese Academy of Sciences[XDA19030201]National Natural Science Foundation of China[41861144019 and 41561144013].
文摘Despite its essential importance to various spatial agriculture and environmental applications,the information on actual cropland area and its geographical distribution remain highly uncertain over Africa among remote-sensing products.Each of the African regions has its unique physical and environmental limiting factors to accurate cropland mapping,which leads to high spatial discre-pancies among remote sensing cropland products.Since no dataset could cope with all limitations,multiple datasets initially derived from various remote sensing sensors and classification techniques must be integrated into a more accurate cropland product than individual layers.Here,in the current study,four cropland products,produced initially from multiple sensors(e.g.Landsat-8 OLI,Sentinel-2 MSI,and PROBA-V)to cover the period(2015-2017),were integrated based on their cropland mapping accuracy to build a more accurate cropland layer.The four cropland layers’accuracy was assessed at Agro-ecological zones units via an inten-sive reference dataset(17,592 samples).The most accurate crop-land layer was then identified for each zone to construct the final cropland mask at 30 m resolution for the nominal year of 2016 over Africa.As a result,the new layer was produced in higher cropland mapping accuracy(overall accuracy=91.64%and cropland’s F-score=0.75).The layer mapped the African cropland area as 282 Mha(9.38%of the Continent area).Compared to earlier crop-land synergy layers,the constructed cropland mask showed a considerable improvement in its spatial resolution(30 m instead of 250 m),mapping quality,and closeness to official statistics(R^(2)=0.853 and RMSE=2.85 Mha).The final layer can be down-loaded as described under the“Data Availability Statement”section.
基金Under the auspices of the Second Tibetan Plateau Scientific Expedition and Research Program(STEP,No.2019QZKK0608)。
文摘The construction of human settlements plays a vital role in achieving sustainable development goals.With the growing population,both urban and rural planning—as well as the development of human settlements—on the Qinghai-Tibet Plateau in China have received increasing attention.This study proposed a comprehensive framework to assess the quality of human settlements and identify their limiting factors of Nyingchi City,southeastern Qinghai-Tibet Plateau of China.The framework integrated a region-specific evaluation system tailored to the plateau’s unique characteristics and utilized multi-source data collected through field-based questionnaires and remote sensing.Analytical methods employed include the Criteria Importance Through Intercriteria Correlation(CRITIC)weighting method,difference analysis for evaluating spatial and categorical variations,and a novel approach introduced in this study for identifying limiting factors.In August 2021,a questionnaire-based survey was conducted in the southeastern Qinghai-Tibet Plateau,yielding 823 valid responses.The average evaluation score for human settlement quality(HSQ)was(6.96±0.94),indicating that settlement conditions were approaching the satisfaction threshold.Notably,the score for the health dimension was(6.28±1.41),reflecting relatively underdeveloped health services in the region.From the perspective of spatial patterns,despite its favorable natural conditions,the HSQ in Medog County was relatively backward,reflecting uncoordinated development that warrants attention.Significant differences were observed in the perceptions of human settlements among different groups in terms of ethnicity,income,and educational background.Low ratings for income and employment conditions represented a major limitation to the local HSQ.The development of human settlements in the southeastern Qinghai-Tibet Plateau still needs further promotion due to population disparities and the challenges of creating a healthy environment in high-altitude regions.The results are valuable for improving human settlements in plateau regions,which can help make targeted policy recommendations.
