To address low detection accuracy in near-coastal vessel target detection under complex conditions,a novel near-coastal vessel detection model based on an improved YOLOv7 architecture is proposed in this paper.The att...To address low detection accuracy in near-coastal vessel target detection under complex conditions,a novel near-coastal vessel detection model based on an improved YOLOv7 architecture is proposed in this paper.The attention mechanism Coordinate Attention is used to improve channel attention weight and enhance a network’s ability to extract small target features.In the enhanced feature extraction network,the lightweight convolution algorithm Grouped Spatial Convolution is used to replace MPConv to reduce model calculation costs.EIoU Loss is used to replace the regression frame loss function in YOLOv7 to reduce the probability of missed and false detection.The performance of the improved model was verified using an enhanced dataset obtained through rainy and foggy weather simulation.Experiments were conducted on the datasets before and after the enhancement.The improved model achieved a mean average precision(mAP)of 97.45%on the original dataset,and the number of parameters was reduced by 2%.On the enhanced dataset,the mAP of the improved model reached 88.08%.Compared with seven target detection models,such as Faster R-CNN,YOLOv3,YOLOv4,YOLOv5,YOLOv7,YOLOv8-n,and YOLOv8-s,the improved model can effectively reduce the missed and false detection rates and improve target detection accuracy.The improved model not only accurately detects vessels in complex weather environments but also outperforms other methods on original and enhanced SeaShip datasets.This finding shows that the improved model can achieve near-coastal vessel target detection in multiple environments,laying the foundation for vessel path planning and automatic obstacle avoidance.展开更多
Three-dimensional reconstruction of wood carving based on CT tomography data is important.In this paper,we propose a novel 3D variational framework for this task,which includes two procedures.First,a fitting approach ...Three-dimensional reconstruction of wood carving based on CT tomography data is important.In this paper,we propose a novel 3D variational framework for this task,which includes two procedures.First,a fitting approach is applied to a sequence of wood carving images acquired by CT scanner.The regions of interest(ROIs)can be obtained for the second segmentation after fitting.Second,we utilise a 3D TV(total variation)L1 variational model to directly segment the 3D volume of the ROIs.In addition,the TV-L1 model can smooth the volume and gives a clear 3D volume rendering result.By introducing a dual auxiliary variable,the fast primal-dual method is developed to improve the computational efficiency of the 3D TV-L1 model.Extensive experimental results are conducted to demonstrate the performance of the proposed framework.展开更多
Grasslands,one of the major terrestrial ecosystems,are essential for the maintenance of ecological and production functions;however,they are undergoing extensive degradation.The development and cutting-edge exploratio...Grasslands,one of the major terrestrial ecosystems,are essential for the maintenance of ecological and production functions;however,they are undergoing extensive degradation.The development and cutting-edge explorations in grassland science are critical to addressing challenges such as climate change and the increasing influence of human activities.To identify research trends in grassland science,latent Dirichlet allocation(LDA)topic modelling was used to conduct an automated content analysis on 123,829 papers available on Web of Science Core Collection from 1900 to 2020.Results from this analysis showed that grassland research has become increasing multidisciplinary,accompanied by a pronounced reduction in the relative frequency of traditional production-oriented research and an increase in the themes focusing on ecological functions and modern technologies.Changes in research activities have been uneven globally,with a significant increase in the number of publications in China and Brazil,which probably reflects an increased support from various governmental agencies in these countries.Additionally,in 2019,China surpassed the United States in terms of the total number of publications.Further,this study identified important topics and emerging challenges in grassland research,such as biodiversity conservation,climate changes,and genetic considerations.Comprehensive improvement of education,research,global cooperation,and funding strategies will be necessary to promote grassland science research on frontier themes and to effectively address the social and environmental challenges in the new era.展开更多
Objective:Intracerebral delivery of agents in liquid form is usually achieved through commercially available and durable metal needles.However,their size and texture may contribute to mechanical brain damage.Glass pip...Objective:Intracerebral delivery of agents in liquid form is usually achieved through commercially available and durable metal needles.However,their size and texture may contribute to mechanical brain damage.Glass pipettes with a thin tip may significantly reduce injection-associated brain damage but require access to prohibitively expensive programmable pipette pullers.This study is to remove the economic barrier to the application of minimally invasive delivery of therapeutics to the brain,such as chemical compounds,viral vectors,and cells.Methods:We took advantage of the rapid development of free educational online resources and emerging low-cost 3D printers by designing an affordable pipette puller(APP)to remove the cost obstacle.Results:We showed that our APP could produce glass pipettes with a sharp tip opening down to 20μm or less,which is sufficiently thin for the delivery of therapeutics into the brain.A pipeline from pipette pulling to brain injection using low-cost and open-source equipment was established to facilitate the application of the APP.Conclusion:In the spirit of frugal science,our device may democratize glass pipette-puling and substantially promote the application of minimally invasive and precisely controlled delivery of therapeutics to the brain for finding more effective therapies of brain diseases.展开更多
Grasslands are one of the major biomes on Earth,covering approximately 25%of the terrestrial planet.Human history is deeply intertwined with grassland biomes,where we,as a natural species,first walked upright 2 millio...Grasslands are one of the major biomes on Earth,covering approximately 25%of the terrestrial planet.Human history is deeply intertwined with grassland biomes,where we,as a natural species,first walked upright 2 million years ago.Today,grassland ecosystems continue to play an important role in people's livelihoods by producing meat and dairy products,providing habitats for biodiversity,and delivering essential ecosystem services such as climate regulation and cultural heritage.展开更多
文摘To address low detection accuracy in near-coastal vessel target detection under complex conditions,a novel near-coastal vessel detection model based on an improved YOLOv7 architecture is proposed in this paper.The attention mechanism Coordinate Attention is used to improve channel attention weight and enhance a network’s ability to extract small target features.In the enhanced feature extraction network,the lightweight convolution algorithm Grouped Spatial Convolution is used to replace MPConv to reduce model calculation costs.EIoU Loss is used to replace the regression frame loss function in YOLOv7 to reduce the probability of missed and false detection.The performance of the improved model was verified using an enhanced dataset obtained through rainy and foggy weather simulation.Experiments were conducted on the datasets before and after the enhancement.The improved model achieved a mean average precision(mAP)of 97.45%on the original dataset,and the number of parameters was reduced by 2%.On the enhanced dataset,the mAP of the improved model reached 88.08%.Compared with seven target detection models,such as Faster R-CNN,YOLOv3,YOLOv4,YOLOv5,YOLOv7,YOLOv8-n,and YOLOv8-s,the improved model can effectively reduce the missed and false detection rates and improve target detection accuracy.The improved model not only accurately detects vessels in complex weather environments but also outperforms other methods on original and enhanced SeaShip datasets.This finding shows that the improved model can achieve near-coastal vessel target detection in multiple environments,laying the foundation for vessel path planning and automatic obstacle avoidance.
文摘Three-dimensional reconstruction of wood carving based on CT tomography data is important.In this paper,we propose a novel 3D variational framework for this task,which includes two procedures.First,a fitting approach is applied to a sequence of wood carving images acquired by CT scanner.The regions of interest(ROIs)can be obtained for the second segmentation after fitting.Second,we utilise a 3D TV(total variation)L1 variational model to directly segment the 3D volume of the ROIs.In addition,the TV-L1 model can smooth the volume and gives a clear 3D volume rendering result.By introducing a dual auxiliary variable,the fast primal-dual method is developed to improve the computational efficiency of the 3D TV-L1 model.Extensive experimental results are conducted to demonstrate the performance of the proposed framework.
文摘Grasslands,one of the major terrestrial ecosystems,are essential for the maintenance of ecological and production functions;however,they are undergoing extensive degradation.The development and cutting-edge explorations in grassland science are critical to addressing challenges such as climate change and the increasing influence of human activities.To identify research trends in grassland science,latent Dirichlet allocation(LDA)topic modelling was used to conduct an automated content analysis on 123,829 papers available on Web of Science Core Collection from 1900 to 2020.Results from this analysis showed that grassland research has become increasing multidisciplinary,accompanied by a pronounced reduction in the relative frequency of traditional production-oriented research and an increase in the themes focusing on ecological functions and modern technologies.Changes in research activities have been uneven globally,with a significant increase in the number of publications in China and Brazil,which probably reflects an increased support from various governmental agencies in these countries.Additionally,in 2019,China surpassed the United States in terms of the total number of publications.Further,this study identified important topics and emerging challenges in grassland research,such as biodiversity conservation,climate changes,and genetic considerations.Comprehensive improvement of education,research,global cooperation,and funding strategies will be necessary to promote grassland science research on frontier themes and to effectively address the social and environmental challenges in the new era.
基金UMGCCC American Cancer Society Institutional Research,Grant/Award Numbers:IRG-18-160-16,NIH1R21AG077631-01,R03NS123733Maryland Stem Cell Research Fund,Grant/Award Numbers:2022-MSCRFL-5893,R03NS128459。
文摘Objective:Intracerebral delivery of agents in liquid form is usually achieved through commercially available and durable metal needles.However,their size and texture may contribute to mechanical brain damage.Glass pipettes with a thin tip may significantly reduce injection-associated brain damage but require access to prohibitively expensive programmable pipette pullers.This study is to remove the economic barrier to the application of minimally invasive delivery of therapeutics to the brain,such as chemical compounds,viral vectors,and cells.Methods:We took advantage of the rapid development of free educational online resources and emerging low-cost 3D printers by designing an affordable pipette puller(APP)to remove the cost obstacle.Results:We showed that our APP could produce glass pipettes with a sharp tip opening down to 20μm or less,which is sufficiently thin for the delivery of therapeutics into the brain.A pipeline from pipette pulling to brain injection using low-cost and open-source equipment was established to facilitate the application of the APP.Conclusion:In the spirit of frugal science,our device may democratize glass pipette-puling and substantially promote the application of minimally invasive and precisely controlled delivery of therapeutics to the brain for finding more effective therapies of brain diseases.
文摘Grasslands are one of the major biomes on Earth,covering approximately 25%of the terrestrial planet.Human history is deeply intertwined with grassland biomes,where we,as a natural species,first walked upright 2 million years ago.Today,grassland ecosystems continue to play an important role in people's livelihoods by producing meat and dairy products,providing habitats for biodiversity,and delivering essential ecosystem services such as climate regulation and cultural heritage.
