High Spatial and Spectral Resolution(HSSR)remote-sensing images can provide rich spectral bands and detailed ground information,but there is a relative lack of research on this new type of remote-sensing data.Although...High Spatial and Spectral Resolution(HSSR)remote-sensing images can provide rich spectral bands and detailed ground information,but there is a relative lack of research on this new type of remote-sensing data.Although there are already some HSSR datasets for deep learning model training and testing,the data volume of these datasets is small,resulting in low classification accuracy and weak generalization ability of the trained models.In this paper,an HSSR dataset Luojia-HSSR is constructed based on aerial hyperspectral imagery of southern Shenyang City of Liaoning Province in China.To our knowledge,it is the largest HSSR dataset to date,with 6438 pairs of 256×256 sized samples(including 3480 pairs in the training set,2209 pairs in the test set,and 749 pairs in the validation set),covering area of 161 km2 with spatial resolution 0.75 m,249 Visible and Near-Infrared(VNIR)spectral bands,and corresponding to 23 classes of field-validated ground coverage.It is an ideal experimental data for spatial-spectral feature extraction.Furthermore,a new deep learning model 3D-HRNet for interpreting HSSR images is proposed.The conv-neck in HRNet is modified to better mine the spatial information of the images.Then,a 3D convolution module with attention mechanism is designed to capture the global-local fine spectral information simultaneously.Subsequently,the 3D convolution is inserted into the HRNet to optimize the performance.The experiments show that the 3D-HRNet model has good interpreting ability for the Luojia-HSSR dataset with the Frequency Weighted Intersection over Union(FWIoU)reaching 80.54%,indicating that the Luojia-HSSR dataset constructed in this paper and the proposed 3D-HRnet model have good applicable prospects for processing HSSR remote sensing images.展开更多
In engineering practice, tubular X-joints have been widely used in offshore structures. The fatigue failure of tubular X-joints in offshore engineering is mainly caused by axial tensile stress. In this study, the stre...In engineering practice, tubular X-joints have been widely used in offshore structures. The fatigue failure of tubular X-joints in offshore engineering is mainly caused by axial tensile stress. In this study, the stress concentration factor distribution along the weld toe in the hot spot stress region for tubular X-joints subject to axial loads have been analyzed by use of finite element method. Through numerical analysis, it has been found that the peak stress concentration factor is located at the saddle position. Thereafter, 80 models have been analyzed, and the effect of the geometric parameters of a tubular X-joint on the stress concentration factor has been investigated. Based on the experimental values of the numerical stress concentration factor, a parametric equation to calculate the stress concentration factor of tubular X-joints has been proposed. The accuracy of this equation has been verified against the requirement of the Fatigue Guidance Review Panel, and the proposed equation is found capable of producing reasonably accurate stress concentration factor values for tubular X-joints subject to axial loads.展开更多
The passive optical network(PON)is the most promising candidate for supporting multiple service classes,i.e.,high priority(HP)traffic and best-effort(BE)traffic,in the next-generation broadband access networks.Providi...The passive optical network(PON)is the most promising candidate for supporting multiple service classes,i.e.,high priority(HP)traffic and best-effort(BE)traffic,in the next-generation broadband access networks.Providing an efficient dynamic bandwidth allocation(DBA)algorithm is the most important issue for supporting multiple service classes of traffic simultaneously.This paper presents a new DBA algorithm,which is the modified version of the hybrid slot-size/rate(HSSR)scheme,and is called the modified HSSR(MHSSR)algorithm.We have also modified the control message scheduling algorithm to fit it to the proposed scheme.In the proposed MHSSR scheme,each cycle time,i.e.,the length of a time cycle,is equally divided into two parts.In the 1st part of the cycle time,the bandwidth is dynamically allocated to the HP traffic of all the optical network units(ONUs).In contrast,in the 2nd part of the cycle time,the bandwidth is dynamically allocated to the BE traffic of one or multiple ONUs.Furthermore,to ensure guaranteed service for the HP traffic,the 1st part of the cycle time is extended while the 2nd part of the cycle time is compressed if the demand of the HP traffic is very high.However,the 1st part of the cycle time will never be compressed,even at the lightly loaded condition of the HP traffic.The modified control message-scheduling algorithm effectively coordinates the timing sequence between the two parts of the cycle time and ensures the synchronization between two Gate messages to each ONU.We have evaluated the performance of the proposed schemes through numerical simulations in terms of endto-end packet delay,bandwidth wastage,remaining bytes per time cycle,jitter,and throughput for different offered loads.From the comparative analysis,it can be seen that the proposed scheme provides better performance than the existing HSSR and delay variation-guaranteed schemes.展开更多
基金supported by the Major Program of the National Natural Science Foundation of China[grant number 92038301]The research was also supported by the National Natural Science Foundation of China[grant number 41971295]+1 种基金the Foundation for Innovative Research Groups of the Natural Science Foundation of Hubei Province[grant number 2020CFA003]the Special Fund of Hubei Luojia Laboratory.
文摘High Spatial and Spectral Resolution(HSSR)remote-sensing images can provide rich spectral bands and detailed ground information,but there is a relative lack of research on this new type of remote-sensing data.Although there are already some HSSR datasets for deep learning model training and testing,the data volume of these datasets is small,resulting in low classification accuracy and weak generalization ability of the trained models.In this paper,an HSSR dataset Luojia-HSSR is constructed based on aerial hyperspectral imagery of southern Shenyang City of Liaoning Province in China.To our knowledge,it is the largest HSSR dataset to date,with 6438 pairs of 256×256 sized samples(including 3480 pairs in the training set,2209 pairs in the test set,and 749 pairs in the validation set),covering area of 161 km2 with spatial resolution 0.75 m,249 Visible and Near-Infrared(VNIR)spectral bands,and corresponding to 23 classes of field-validated ground coverage.It is an ideal experimental data for spatial-spectral feature extraction.Furthermore,a new deep learning model 3D-HRNet for interpreting HSSR images is proposed.The conv-neck in HRNet is modified to better mine the spatial information of the images.Then,a 3D convolution module with attention mechanism is designed to capture the global-local fine spectral information simultaneously.Subsequently,the 3D convolution is inserted into the HRNet to optimize the performance.The experiments show that the 3D-HRNet model has good interpreting ability for the Luojia-HSSR dataset with the Frequency Weighted Intersection over Union(FWIoU)reaching 80.54%,indicating that the Luojia-HSSR dataset constructed in this paper and the proposed 3D-HRnet model have good applicable prospects for processing HSSR remote sensing images.
基金The research work was financially supported by the National Natural Scientice Foundation of China(Grant No.10142001)by the Shandong Provincial Natural Scientice Foundation(Grant No.Y2006F46)
文摘In engineering practice, tubular X-joints have been widely used in offshore structures. The fatigue failure of tubular X-joints in offshore engineering is mainly caused by axial tensile stress. In this study, the stress concentration factor distribution along the weld toe in the hot spot stress region for tubular X-joints subject to axial loads have been analyzed by use of finite element method. Through numerical analysis, it has been found that the peak stress concentration factor is located at the saddle position. Thereafter, 80 models have been analyzed, and the effect of the geometric parameters of a tubular X-joint on the stress concentration factor has been investigated. Based on the experimental values of the numerical stress concentration factor, a parametric equation to calculate the stress concentration factor of tubular X-joints has been proposed. The accuracy of this equation has been verified against the requirement of the Fatigue Guidance Review Panel, and the proposed equation is found capable of producing reasonably accurate stress concentration factor values for tubular X-joints subject to axial loads.
文摘The passive optical network(PON)is the most promising candidate for supporting multiple service classes,i.e.,high priority(HP)traffic and best-effort(BE)traffic,in the next-generation broadband access networks.Providing an efficient dynamic bandwidth allocation(DBA)algorithm is the most important issue for supporting multiple service classes of traffic simultaneously.This paper presents a new DBA algorithm,which is the modified version of the hybrid slot-size/rate(HSSR)scheme,and is called the modified HSSR(MHSSR)algorithm.We have also modified the control message scheduling algorithm to fit it to the proposed scheme.In the proposed MHSSR scheme,each cycle time,i.e.,the length of a time cycle,is equally divided into two parts.In the 1st part of the cycle time,the bandwidth is dynamically allocated to the HP traffic of all the optical network units(ONUs).In contrast,in the 2nd part of the cycle time,the bandwidth is dynamically allocated to the BE traffic of one or multiple ONUs.Furthermore,to ensure guaranteed service for the HP traffic,the 1st part of the cycle time is extended while the 2nd part of the cycle time is compressed if the demand of the HP traffic is very high.However,the 1st part of the cycle time will never be compressed,even at the lightly loaded condition of the HP traffic.The modified control message-scheduling algorithm effectively coordinates the timing sequence between the two parts of the cycle time and ensures the synchronization between two Gate messages to each ONU.We have evaluated the performance of the proposed schemes through numerical simulations in terms of endto-end packet delay,bandwidth wastage,remaining bytes per time cycle,jitter,and throughput for different offered loads.From the comparative analysis,it can be seen that the proposed scheme provides better performance than the existing HSSR and delay variation-guaranteed schemes.