Multimodality image registration and fusion are essential steps in building 3-D models from remotesensing data. We present in this paper a neural network technique for the registration and fusion of multimodali-ty rem...Multimodality image registration and fusion are essential steps in building 3-D models from remotesensing data. We present in this paper a neural network technique for the registration and fusion of multimodali-ty remote sensing data for the reconstruction of 3-D models of terrain regions. A FeedForward neural network isused to fuse the intensity data sets with the spatial data set after learning its geometry. Results on real data arepresented. Human performance evaluation is assessed on several perceptual tests in order to evaluate the fusionresults.展开更多
A 360°video stream provide users a choice of viewing one's own point of interest inside the immersive contents.Performing head or hand manipulations to view the interesting scene in a 360°video is very t...A 360°video stream provide users a choice of viewing one's own point of interest inside the immersive contents.Performing head or hand manipulations to view the interesting scene in a 360°video is very tedious and the user may view the interested frame during his head/hand movement or even lose it.While automatically extracting user's point of interest(UPI)in a 360°video is very challenging because of subjectivity and difference of comforts.To handle these challenges and provide user's the best and visually pleasant view,we propose an automatic approach by utilizing two CNN models:object detector and aesthetic score of the scene.The proposed framework is three folded:pre-processing,Deepdive architecture,and view selection pipeline.In first fold,an input 360°video-frame is divided into three sub frames,each one with 120°view.In second fold,each sub-frame is passed through CNN models to extract visual features in the sub-frames and calculate aesthetic score.Finally,decision pipeline selects the sub frame with salient object based on the detected object and calculated aesthetic score.As compared to other state-of-the-art techniques which are domain specific approaches i.e.,support sports 360°video,our syste m support most of the 360°videos genre.Performance evaluation of proposed framework on our own collected data from various websites indicate performance for different categories of 360°videos.展开更多
The Amazon basin has experienced an extreme drought that started in the austral summer of 2022-23 and extends into 2024. This drought started earlier than other previous droughts. Although some rain fell during the au...The Amazon basin has experienced an extreme drought that started in the austral summer of 2022-23 and extends into 2024. This drought started earlier than other previous droughts. Although some rain fell during the austral summer, totals remained below average. Higher temperatures during austral winter and spring 2023, which affected most of Central South America, then aggravated drought conditions. This coincided with an intense El Niño and abnormally warm tropical North Atlantic Ocean temperatures since mid-2023. Decreased rainfall across the Amazon basin, negative anomalies in evapotranspiration (derived from latent heat) and soil moisture indicators, as well as increased temperatures during the dry-to-wet transition season, September-October-November (SON) 2023, combined to delay the onset of the wet season in the hydrological year 2023-24 by nearly two months and caused it to be uncharacteristically weak. SON 2023 registered a precipitation deficit of the order of 50 to 100 mm/month, and temperatures +3˚C higher than usual in Amazonia, leading to reduced evapotranspiration and soil moisture indicators. These processes, in turn, determined an exceptionally late onset and a lengthening of the dry season, affecting the 2023-2024 hydrological year. These changes were aggravated by a heat wave from June to December 2023. Drought-heat compound events and their consequences are the most critical natural threats to society. River levels reached record lows, or dried up completely, affecting Amazonian ecosystems. Increased risk of wildfires is another concern exacerbated by these conditions.展开更多
Currently three up-to-date Terrestrial Reference Frames(TRF) are available, the ITRF2014 from IGN, the DTRF2014 from DGFI-TUM, and JTRF2014 from JPL. All use the identical input data of space-geodetic station position...Currently three up-to-date Terrestrial Reference Frames(TRF) are available, the ITRF2014 from IGN, the DTRF2014 from DGFI-TUM, and JTRF2014 from JPL. All use the identical input data of space-geodetic station positions and Earth orientation parameters, but the concept of combining these data is fundamentally different. The IGN approach is based on the combination of technique solutions, while the DGFI is combining the normal equation systems. Both yield in reference epoch coordinates and velocities for a global set of stations. JPL uses a Kalman filter approach, realizing a TRF through weekly time series of geocentric coordinates. As the determination of the CRF is not independent of the TRF and vice versa, the choice of the TRF might impact on the CRF, Within this work we assess this effect.We find that the estimated Earth orientation parameter(EOP) from DTRF2014 agree best with those from ITRF2014, the EOP resulting from JTRF2014 show besides clear yearly signals also some artifacts linked to certain stations. The estimated source position time series however, agree with each other better than ±1 μas. When fixing EOP and station positions we can see the maximal effect of the TRF on the CRF. Here large systematics in position as well as proper motion arise. In case of ITRF2008 they can be linked to the missing data after 2008. By allowing the EOP and stations to participate in the adjustment,the agreement increases, however, systematics remain.展开更多
This paper presents the knowledge-based image database system (ISDBS) developed inour laboratory and proposes principles of the system design of a knowledge-based data model,a new data structure and the knowledge repr...This paper presents the knowledge-based image database system (ISDBS) developed inour laboratory and proposes principles of the system design of a knowledge-based data model,a new data structure and the knowledge representation language IDL applicable to the comput-er-aided regional planning. ISDBS was applied successfully to the land planning of AnsaiCounty on the loess plateau and the forest resource management of Songjianghe territory inNortheast China.展开更多
The effectiveness of machine learning algorithms and the limited reference data introduce uncertainness for sugarcane classification.To address these problems,our study classified sugarcane plantations at the field sc...The effectiveness of machine learning algorithms and the limited reference data introduce uncertainness for sugarcane classification.To address these problems,our study classified sugarcane plantations at the field scale using multi-temporal and multi-sensor data together with a large number of ground truth datasets(>13,000 points)and compared the efficacy of ensemble and kernel classifier methods over 3 years(2021,2022,and 2023)across Northeast Thailand.In the first step,land cover was generated from a random forest classifier,demonstrating excellent results for all years with an OA higher than 95%.In the second step,the discretization of sugarcane from non-sugarcane classes in the agricultural category was conducted using four efficient machine learning algorithms(decision tree(DT),random forest(RF),support vector machine(SVM),and one-class SVM).The RF classifier gave the optimal results with over 90%accuracy.Our results aligned with provincial statistics from the Office of the Cane and Sugar Board,thereby highlighting the efficacy and reliability of the RF method in mapping sugarcane in small fields and cloudy regions.A temporal evolution analysis of sugarcane cultivation spanning the preceding 3 years revealed a significant increase in the productive area.Our findings provide crucial information for sustainable management practices.展开更多
Remote sensing techniques are a useful tool for continuous observation of the Earth at global scale.However,products derived from remote sensing data require a rigorous validation using in situ data.Moderate Resolutio...Remote sensing techniques are a useful tool for continuous observation of the Earth at global scale.However,products derived from remote sensing data require a rigorous validation using in situ data.Moderate Resolution Imaging Spectroradiometer(MODIS)is not really a sounding instrument,but it does have 16 infrared bands(bands 20-36 covering the spectral range from 3μm to 14μm)that allow the retrieval of temperature and moisture profiles as well as total column integrated magnitudes.In this paper we show the results obtained in the evaluation of MOD07 daytime and nighttime products over the Iberian Peninsula during the decade from 2000 to 2010 using nine radiosonde stations.Although MODIS limitations in comparison with other sounding instruments,the validation provided satisfactory results,with bias(MOD07 minus radiosonde)<0.3 cm and a standard deviation of 0.5 cm for the total column water vapor,and bias around 1 K on average with standard deviations between 2 K and 3 K for air temperature at different pressure levels.On average,bias was positive and below 2 K with standard deviations around 5 K for the dew point temperature case.Large errors were found in this case for pressure levels higher than 50 hPa.展开更多
Remote sensing and climate digital products have become increasingly available in recent years.Access to these products has favored a variety of Digital Earth studies,such as the analysis of the impact of global warmi...Remote sensing and climate digital products have become increasingly available in recent years.Access to these products has favored a variety of Digital Earth studies,such as the analysis of the impact of global warming over different biomes.The study of the Amazon forest response to drought has recently received particular attention from the scientific community due to the occurrence of extreme droughts and anomalous warming over the last decade.This paper focuses on the differences observed between surface thermal anomalies obtained from remote sensing moderate resolution imaging spectroradiometer(MODIS)and climatic(ERA-Interim)monthly products over the Amazon forest.With a few exceptions,results show that the spatial pattern of standardized anomalies is similar for both products.In terms of absolute anomalies,the differences between the two products show a bias near to zero with a standard deviation of around 0.2 K,although the differences can be up to 1 K over particular regions and months.Despite this general agreement,the proper filtering of MODIS daily values in order to construct a new monthly product showed a dramatic reduction in the number of reliable pixels during the rainy season,while for the dry season this reduction is only seen in Northern Amazonia.展开更多
文摘Multimodality image registration and fusion are essential steps in building 3-D models from remotesensing data. We present in this paper a neural network technique for the registration and fusion of multimodali-ty remote sensing data for the reconstruction of 3-D models of terrain regions. A FeedForward neural network isused to fuse the intensity data sets with the spatial data set after learning its geometry. Results on real data arepresented. Human performance evaluation is assessed on several perceptual tests in order to evaluate the fusionresults.
文摘A 360°video stream provide users a choice of viewing one's own point of interest inside the immersive contents.Performing head or hand manipulations to view the interesting scene in a 360°video is very tedious and the user may view the interested frame during his head/hand movement or even lose it.While automatically extracting user's point of interest(UPI)in a 360°video is very challenging because of subjectivity and difference of comforts.To handle these challenges and provide user's the best and visually pleasant view,we propose an automatic approach by utilizing two CNN models:object detector and aesthetic score of the scene.The proposed framework is three folded:pre-processing,Deepdive architecture,and view selection pipeline.In first fold,an input 360°video-frame is divided into three sub frames,each one with 120°view.In second fold,each sub-frame is passed through CNN models to extract visual features in the sub-frames and calculate aesthetic score.Finally,decision pipeline selects the sub frame with salient object based on the detected object and calculated aesthetic score.As compared to other state-of-the-art techniques which are domain specific approaches i.e.,support sports 360°video,our syste m support most of the 360°videos genre.Performance evaluation of proposed framework on our own collected data from various websites indicate performance for different categories of 360°videos.
文摘The Amazon basin has experienced an extreme drought that started in the austral summer of 2022-23 and extends into 2024. This drought started earlier than other previous droughts. Although some rain fell during the austral summer, totals remained below average. Higher temperatures during austral winter and spring 2023, which affected most of Central South America, then aggravated drought conditions. This coincided with an intense El Niño and abnormally warm tropical North Atlantic Ocean temperatures since mid-2023. Decreased rainfall across the Amazon basin, negative anomalies in evapotranspiration (derived from latent heat) and soil moisture indicators, as well as increased temperatures during the dry-to-wet transition season, September-October-November (SON) 2023, combined to delay the onset of the wet season in the hydrological year 2023-24 by nearly two months and caused it to be uncharacteristically weak. SON 2023 registered a precipitation deficit of the order of 50 to 100 mm/month, and temperatures +3˚C higher than usual in Amazonia, leading to reduced evapotranspiration and soil moisture indicators. These processes, in turn, determined an exceptionally late onset and a lengthening of the dry season, affecting the 2023-2024 hydrological year. These changes were aggravated by a heat wave from June to December 2023. Drought-heat compound events and their consequences are the most critical natural threats to society. River levels reached record lows, or dried up completely, affecting Amazonian ecosystems. Increased risk of wildfires is another concern exacerbated by these conditions.
基金supported by the Deutsche Forschungsgemeinschaft(DFG), Project Nr.:HE 5937/2-1 and NO318/ 13-1supported by the European Research Council(ERC) under the ERC-2017-STG SENTIFLEX project(Grant Agreement 755617)
文摘Currently three up-to-date Terrestrial Reference Frames(TRF) are available, the ITRF2014 from IGN, the DTRF2014 from DGFI-TUM, and JTRF2014 from JPL. All use the identical input data of space-geodetic station positions and Earth orientation parameters, but the concept of combining these data is fundamentally different. The IGN approach is based on the combination of technique solutions, while the DGFI is combining the normal equation systems. Both yield in reference epoch coordinates and velocities for a global set of stations. JPL uses a Kalman filter approach, realizing a TRF through weekly time series of geocentric coordinates. As the determination of the CRF is not independent of the TRF and vice versa, the choice of the TRF might impact on the CRF, Within this work we assess this effect.We find that the estimated Earth orientation parameter(EOP) from DTRF2014 agree best with those from ITRF2014, the EOP resulting from JTRF2014 show besides clear yearly signals also some artifacts linked to certain stations. The estimated source position time series however, agree with each other better than ±1 μas. When fixing EOP and station positions we can see the maximal effect of the TRF on the CRF. Here large systematics in position as well as proper motion arise. In case of ITRF2008 they can be linked to the missing data after 2008. By allowing the EOP and stations to participate in the adjustment,the agreement increases, however, systematics remain.
基金Project supported by the National Natural Science Foundation of China.
文摘This paper presents the knowledge-based image database system (ISDBS) developed inour laboratory and proposes principles of the system design of a knowledge-based data model,a new data structure and the knowledge representation language IDL applicable to the comput-er-aided regional planning. ISDBS was applied successfully to the land planning of AnsaiCounty on the loess plateau and the forest resource management of Songjianghe territory inNortheast China.
基金financially supported by Mahasarakham University.
文摘The effectiveness of machine learning algorithms and the limited reference data introduce uncertainness for sugarcane classification.To address these problems,our study classified sugarcane plantations at the field scale using multi-temporal and multi-sensor data together with a large number of ground truth datasets(>13,000 points)and compared the efficacy of ensemble and kernel classifier methods over 3 years(2021,2022,and 2023)across Northeast Thailand.In the first step,land cover was generated from a random forest classifier,demonstrating excellent results for all years with an OA higher than 95%.In the second step,the discretization of sugarcane from non-sugarcane classes in the agricultural category was conducted using four efficient machine learning algorithms(decision tree(DT),random forest(RF),support vector machine(SVM),and one-class SVM).The RF classifier gave the optimal results with over 90%accuracy.Our results aligned with provincial statistics from the Office of the Cane and Sugar Board,thereby highlighting the efficacy and reliability of the RF method in mapping sugarcane in small fields and cloudy regions.A temporal evolution analysis of sugarcane cultivation spanning the preceding 3 years revealed a significant increase in the productive area.Our findings provide crucial information for sustainable management practices.
基金We acknowledge funding from European Union[CEOP-AEGIS,project FP7-ENV-2007-1 Proposal No.212921]the Ministerio de Economía y Competitividad[EODIX,project AYA2008-0595-C04-01+1 种基金CEOS-Spain,project AYA2011-29334-C02-01]the Universitat de València[grant number PRECOM13-115366].
文摘Remote sensing techniques are a useful tool for continuous observation of the Earth at global scale.However,products derived from remote sensing data require a rigorous validation using in situ data.Moderate Resolution Imaging Spectroradiometer(MODIS)is not really a sounding instrument,but it does have 16 infrared bands(bands 20-36 covering the spectral range from 3μm to 14μm)that allow the retrieval of temperature and moisture profiles as well as total column integrated magnitudes.In this paper we show the results obtained in the evaluation of MOD07 daytime and nighttime products over the Iberian Peninsula during the decade from 2000 to 2010 using nine radiosonde stations.Although MODIS limitations in comparison with other sounding instruments,the validation provided satisfactory results,with bias(MOD07 minus radiosonde)<0.3 cm and a standard deviation of 0.5 cm for the total column water vapor,and bias around 1 K on average with standard deviations between 2 K and 3 K for air temperature at different pressure levels.On average,bias was positive and below 2 K with standard deviations around 5 K for the dew point temperature case.Large errors were found in this case for pressure levels higher than 50 hPa.
基金funded by the University of Valencia[UV-INV-PRECOMP13-115366]Ministerio de Ciencia e Innovación[CEOS-Spain,AYA2011-29334-C02-01],CONICYT[Fondecyt Iniciacion–1130359]the University of Chile[Ayuda U Viajes,VID-Uchile 2014–2015].
文摘Remote sensing and climate digital products have become increasingly available in recent years.Access to these products has favored a variety of Digital Earth studies,such as the analysis of the impact of global warming over different biomes.The study of the Amazon forest response to drought has recently received particular attention from the scientific community due to the occurrence of extreme droughts and anomalous warming over the last decade.This paper focuses on the differences observed between surface thermal anomalies obtained from remote sensing moderate resolution imaging spectroradiometer(MODIS)and climatic(ERA-Interim)monthly products over the Amazon forest.With a few exceptions,results show that the spatial pattern of standardized anomalies is similar for both products.In terms of absolute anomalies,the differences between the two products show a bias near to zero with a standard deviation of around 0.2 K,although the differences can be up to 1 K over particular regions and months.Despite this general agreement,the proper filtering of MODIS daily values in order to construct a new monthly product showed a dramatic reduction in the number of reliable pixels during the rainy season,while for the dry season this reduction is only seen in Northern Amazonia.
