Traffic scene captioning technology automatically generates one or more sentences to describe the content of traffic scenes by analyzing the content of the input traffic scene images,ensuring road safety while providi...Traffic scene captioning technology automatically generates one or more sentences to describe the content of traffic scenes by analyzing the content of the input traffic scene images,ensuring road safety while providing an important decision-making function for sustainable transportation.In order to provide a comprehensive and reasonable description of complex traffic scenes,a traffic scene semantic captioningmodel withmulti-stage feature enhancement is proposed in this paper.In general,the model follows an encoder-decoder structure.First,multilevel granularity visual features are used for feature enhancement during the encoding process,which enables the model to learn more detailed content in the traffic scene image.Second,the scene knowledge graph is applied to the decoding process,and the semantic features provided by the scene knowledge graph are used to enhance the features learned by the decoder again,so that themodel can learn the attributes of objects in the traffic scene and the relationships between objects to generate more reasonable captions.This paper reports extensive experiments on the challenging MS-COCO dataset,evaluated by five standard automatic evaluation metrics,and the results show that the proposed model has improved significantly in all metrics compared with the state-of-the-art methods,especially achieving a score of 129.0 on the CIDEr-D evaluation metric,which also indicates that the proposed model can effectively provide a more reasonable and comprehensive description of the traffic scene.展开更多
CE-1 Lunar Microwave Sounder (CELMS) is the first passive microwave radiometer in the world to sound the surface of the Moon in the lunar orbit at altitude of 200 km. The scientific objective of CELMS is to obtain glo...CE-1 Lunar Microwave Sounder (CELMS) is the first passive microwave radiometer in the world to sound the surface of the Moon in the lunar orbit at altitude of 200 km. The scientific objective of CELMS is to obtain global brightness temperature (TB) of the Moon, to retrieve information on lunar regolith, and to evaluate the distribution of helium-3 on the Moon implanted by solar wind. Before launch of CELMS, a series of experiments were carried out in laboratories to test the performances of the systems, and to calibrate the responses between the input of TB and the output of voltage from the receivers. However, the thermal condition exposed to CELMS is more complicated in lunar orbit than on the Earth, which makes the temperatures of different parts of CELMS wave vary greatly, and the cosmic background is not very clean due to the pointing of cold space antenna to the direction of the satellite running, which brings uncertainties into data-processing of CELMS when the temperature of cold space is used as a calibrator. Furthermore, the lack of knowledge on the lunar ingredients and compositions, distributions of physical temperatures, and properties on lunar microwave radiation leads to difficulties in validating the measurements and retrievals of CELMS. By analyzing the results of ground experiments and the measurements of CELMS in-orbit, along with our knowledge of the properties of lunar surface, here we give algorithms on calibration and antenna pattern correction (APC) of CELMS. We also describe in detail the principle of microwave transfer among the elements of CELMS, and discuss the method on testing calibration parameters of the system. In addition, the theory and model on correction antenna pattern of CELMS are developed by comparing antenna temperatures by CELMS with those simulated by microwave radiative transfer models. The global distribution of TB is given and the features of TB are analyzed. Our results show rich information included in TB on the properties of lunar regolith, especially the thickness and dielectric constant, which are nearly directly reflected by the differences of TB at day and those at night.展开更多
Investigation on Lunar polar area is almost every lunar mission’s primary objective in recent years. The rationale behind it is that illumination and ice resources in this area can be potentially very helpful for con...Investigation on Lunar polar area is almost every lunar mission’s primary objective in recent years. The rationale behind it is that illumination and ice resources in this area can be potentially very helpful for constructing lunar human base. In this paper, we analyze microwave radiometric characteristics of the Moon by using the newly acquired Chang’E-1 Lunar Microwave Sounder (CELMS) data. Microwave brightness temperature at Lunar South Pole (LSP) is distributed regularly with a style of "ring-in-ring", decreasing from equator to pole. Regolith temperature gradient is bigger at lunar equator than at polar area. Brightness temperature diurnal difference decreases with observation frequency. Microwave brightness temperature distribution maps at LSP and Lunar North Pole (LNP) have been made based on the analysis. It is found that microwave brightness temperature becomes to synchronize with elevation beyond -85° latitude. This phenomenon is related to lightening condition and indicates temperature distribution at LSP. The brightness temperature anomaly cold points are potentially cold trap areas for water or ice while hot points imply plenty of illumination resources there.展开更多
Chang'e-5(CE-5)mission is expected to land and sample in the Rümker region,north of the Oceanus Procellarum.To select optimal sampling points,the microwave radiation features in this mare unit are analyzed,an...Chang'e-5(CE-5)mission is expected to land and sample in the Rümker region,north of the Oceanus Procellarum.To select optimal sampling points,the microwave radiation features in this mare unit are analyzed,and the dielectric constant and thickness of lunar regolith are retrieved based on Chang'e lunar microwave sounder(CELMS)data.According to the microwave brightness temperature(TB)maps,TB varying with frequency at noon and midnight are different,in which noon TB increases with frequency,but midnight TB does not follow this rule.Moreover,there are differences among TB distributions at the three higher frequencies,especially in the southern part.In addition,a great TB difference between noon and midnight can be seen in Rümker E that shows higher daytime temperature but lower nighttime temperature.The inversion results reveal that the dielectric constant is lower in the west and north,higher in the middle and east,and the largest value occurs in Rümker E.As for the regolith thickness,most areas are uniform and no more than 4 m,with the thickest and thinnest regolith appearing in the northwest and Rümker E,no less than 6 m and about 2.5 m,respectively.Considering the safety and operability of landing and sampling,the northwest with thick regolith is recommended,while the Rümker E should be eluded.展开更多
基金funded by(i)Natural Science Foundation China(NSFC)under Grant Nos.61402397,61263043,61562093 and 61663046(ii)Open Foundation of Key Laboratory in Software Engineering of Yunnan Province:No.2020SE304.(iii)Practical Innovation Project of Yunnan University,Project Nos.2021z34,2021y128 and 2021y129.
文摘Traffic scene captioning technology automatically generates one or more sentences to describe the content of traffic scenes by analyzing the content of the input traffic scene images,ensuring road safety while providing an important decision-making function for sustainable transportation.In order to provide a comprehensive and reasonable description of complex traffic scenes,a traffic scene semantic captioningmodel withmulti-stage feature enhancement is proposed in this paper.In general,the model follows an encoder-decoder structure.First,multilevel granularity visual features are used for feature enhancement during the encoding process,which enables the model to learn more detailed content in the traffic scene image.Second,the scene knowledge graph is applied to the decoding process,and the semantic features provided by the scene knowledge graph are used to enhance the features learned by the decoder again,so that themodel can learn the attributes of objects in the traffic scene and the relationships between objects to generate more reasonable captions.This paper reports extensive experiments on the challenging MS-COCO dataset,evaluated by five standard automatic evaluation metrics,and the results show that the proposed model has improved significantly in all metrics compared with the state-of-the-art methods,especially achieving a score of 129.0 on the CIDEr-D evaluation metric,which also indicates that the proposed model can effectively provide a more reasonable and comprehensive description of the traffic scene.
文摘CE-1 Lunar Microwave Sounder (CELMS) is the first passive microwave radiometer in the world to sound the surface of the Moon in the lunar orbit at altitude of 200 km. The scientific objective of CELMS is to obtain global brightness temperature (TB) of the Moon, to retrieve information on lunar regolith, and to evaluate the distribution of helium-3 on the Moon implanted by solar wind. Before launch of CELMS, a series of experiments were carried out in laboratories to test the performances of the systems, and to calibrate the responses between the input of TB and the output of voltage from the receivers. However, the thermal condition exposed to CELMS is more complicated in lunar orbit than on the Earth, which makes the temperatures of different parts of CELMS wave vary greatly, and the cosmic background is not very clean due to the pointing of cold space antenna to the direction of the satellite running, which brings uncertainties into data-processing of CELMS when the temperature of cold space is used as a calibrator. Furthermore, the lack of knowledge on the lunar ingredients and compositions, distributions of physical temperatures, and properties on lunar microwave radiation leads to difficulties in validating the measurements and retrievals of CELMS. By analyzing the results of ground experiments and the measurements of CELMS in-orbit, along with our knowledge of the properties of lunar surface, here we give algorithms on calibration and antenna pattern correction (APC) of CELMS. We also describe in detail the principle of microwave transfer among the elements of CELMS, and discuss the method on testing calibration parameters of the system. In addition, the theory and model on correction antenna pattern of CELMS are developed by comparing antenna temperatures by CELMS with those simulated by microwave radiative transfer models. The global distribution of TB is given and the features of TB are analyzed. Our results show rich information included in TB on the properties of lunar regolith, especially the thickness and dielectric constant, which are nearly directly reflected by the differences of TB at day and those at night.
基金supported by National Natural Science Foundation of China (Grant No. 40601066)Chinese "Chang’E-1" Project Microwave Radiometer Item
文摘Investigation on Lunar polar area is almost every lunar mission’s primary objective in recent years. The rationale behind it is that illumination and ice resources in this area can be potentially very helpful for constructing lunar human base. In this paper, we analyze microwave radiometric characteristics of the Moon by using the newly acquired Chang’E-1 Lunar Microwave Sounder (CELMS) data. Microwave brightness temperature at Lunar South Pole (LSP) is distributed regularly with a style of "ring-in-ring", decreasing from equator to pole. Regolith temperature gradient is bigger at lunar equator than at polar area. Brightness temperature diurnal difference decreases with observation frequency. Microwave brightness temperature distribution maps at LSP and Lunar North Pole (LNP) have been made based on the analysis. It is found that microwave brightness temperature becomes to synchronize with elevation beyond -85° latitude. This phenomenon is related to lightening condition and indicates temperature distribution at LSP. The brightness temperature anomaly cold points are potentially cold trap areas for water or ice while hot points imply plenty of illumination resources there.
基金supported by the National Natural Science Foundation of China(Grant No.41771405)。
文摘Chang'e-5(CE-5)mission is expected to land and sample in the Rümker region,north of the Oceanus Procellarum.To select optimal sampling points,the microwave radiation features in this mare unit are analyzed,and the dielectric constant and thickness of lunar regolith are retrieved based on Chang'e lunar microwave sounder(CELMS)data.According to the microwave brightness temperature(TB)maps,TB varying with frequency at noon and midnight are different,in which noon TB increases with frequency,but midnight TB does not follow this rule.Moreover,there are differences among TB distributions at the three higher frequencies,especially in the southern part.In addition,a great TB difference between noon and midnight can be seen in Rümker E that shows higher daytime temperature but lower nighttime temperature.The inversion results reveal that the dielectric constant is lower in the west and north,higher in the middle and east,and the largest value occurs in Rümker E.As for the regolith thickness,most areas are uniform and no more than 4 m,with the thickest and thinnest regolith appearing in the northwest and Rümker E,no less than 6 m and about 2.5 m,respectively.Considering the safety and operability of landing and sampling,the northwest with thick regolith is recommended,while the Rümker E should be eluded.