原油作为一种重要的战略物资,在我国经济和军事等多个领域均起到重要作用。本文提出一种基于深度学习的目标检测模型TCS-YOLO(Transformer-CBAM-SIoU YOLO),该模型在YOLOv5的基础上进行优化,同时基于吉林一号光学遥感卫星影像数据集进...原油作为一种重要的战略物资,在我国经济和军事等多个领域均起到重要作用。本文提出一种基于深度学习的目标检测模型TCS-YOLO(Transformer-CBAM-SIoU YOLO),该模型在YOLOv5的基础上进行优化,同时基于吉林一号光学遥感卫星影像数据集进行实验,对全球范围内的储油罐进行识别与分类。优化内容包括:添加基于Transformer架构的C3TR层对网络进行优化;使用CBAM(Convolutional Block Attention Module)在网络层中添加注意力机制;使用SIoU(Scale-Sensitive Intersection over Union) loss代替CIoU(Complete Intersection over Union) loss作为定位损失函数。实验结果表明:与YOLOv5相比,TCS-YOLO的模型复杂度(Giga Floating Point of Operations,GFLOPs)平均减少3.13%,模型参数量(Parameters)平均减少0.88%,推理速度(Inference Speed)平均降低0.2 ms,mAP0.5(mean Average Precision)平均提升0.2%,mAP0.5∶0.95平均提升1.26%。与此同时,将TCS-YOLO模型与通用目标识别模型YOLOv3,YOLOv4,YOLOv5和Swin Transformer进行对比实验,TCS-YOLO均体现出了更高效的特点。TCS-YOLO模型对全球储油罐的目标识别具有通用可行性,可为遥感数据在能源期货领域提供技术参考。展开更多
The origin of fast radio bursts(FRBs),the brightest cosmic explosion in radio bands,remains unknown.We introduce here a novel method for a comprehensive analysis of active FRBs’behaviors in the timeenergy domain.Usin...The origin of fast radio bursts(FRBs),the brightest cosmic explosion in radio bands,remains unknown.We introduce here a novel method for a comprehensive analysis of active FRBs’behaviors in the timeenergy domain.Using"Pincus Index"and"Maximum Lyapunov Exponent",we were able to quantify the randomness and chaoticity,respectively,of the bursting events and put FRBs in the context of common transient physical phenomena,such as pulsar,earthquakes,and solar flares.In the bivariate timeenergy domain,repeated FRB bursts’behaviors deviate significantly(more random,less chaotic)from pulsars,earthquakes,and solar flares.The waiting times between FRB bursts and the corresponding energy changes exhibit no correlation and remain unpredictable,suggesting that the emission of FRBs does not exhibit the time and energy clustering observed in seismic events.The pronounced stochasticity may arise from a singular source with high entropy or the combination of diverse emission mechanisms/sites.Consequently,our methodology serves as a pragmatic tool for illustrating the congruities and distinctions among diverse physical processes.展开更多
In a multi-bubble system, the bubble behavior is modulated by the primary acoustic field and the secondary acoustic field. To explore the translational motion of bubbles in cavitation liquids containing high-concentra...In a multi-bubble system, the bubble behavior is modulated by the primary acoustic field and the secondary acoustic field. To explore the translational motion of bubbles in cavitation liquids containing high-concentration cavitation nuclei,evolutions of bubbles are recorded by a high-speed camera, and translational trajectories of several representative bubbles are traced. It is found that translational motion of bubbles is always accompanied by the fragmentation and coalescence of bubbles, and for bubbles smaller than 10 μm, the possibility of bubble coalescence is enhanced when the spacing of bubbles is less than 30 μm. The measured signals and their spectra show the presence of strong negative pressure, broadband noise,and various harmonics, which implies that multiple interactions of bubbles appear in the region of high-intensity cavitation.Due to the strong coupling effect, the interaction between bubbles is random. A simplified triple-bubble model is developed to explore the interaction patterns of bubbles affected by the surrounding bubbles. Patterns of bubble interaction, such as attraction, repulsion, stable spacing, and rebound of bubbles, can be predicted by the theoretical analysis, and the obtained results are in good agreement with experimental observations. Mass exchange between the liquid and bubbles as well as absorption in the cavitation nuclei also plays an important role in multi-bubble cavitation, which may account for the weakening of the radial oscillations of bubbles.展开更多
文摘原油作为一种重要的战略物资,在我国经济和军事等多个领域均起到重要作用。本文提出一种基于深度学习的目标检测模型TCS-YOLO(Transformer-CBAM-SIoU YOLO),该模型在YOLOv5的基础上进行优化,同时基于吉林一号光学遥感卫星影像数据集进行实验,对全球范围内的储油罐进行识别与分类。优化内容包括:添加基于Transformer架构的C3TR层对网络进行优化;使用CBAM(Convolutional Block Attention Module)在网络层中添加注意力机制;使用SIoU(Scale-Sensitive Intersection over Union) loss代替CIoU(Complete Intersection over Union) loss作为定位损失函数。实验结果表明:与YOLOv5相比,TCS-YOLO的模型复杂度(Giga Floating Point of Operations,GFLOPs)平均减少3.13%,模型参数量(Parameters)平均减少0.88%,推理速度(Inference Speed)平均降低0.2 ms,mAP0.5(mean Average Precision)平均提升0.2%,mAP0.5∶0.95平均提升1.26%。与此同时,将TCS-YOLO模型与通用目标识别模型YOLOv3,YOLOv4,YOLOv5和Swin Transformer进行对比实验,TCS-YOLO均体现出了更高效的特点。TCS-YOLO模型对全球储油罐的目标识别具有通用可行性,可为遥感数据在能源期货领域提供技术参考。
基金supported by the Open Project Program of the Key Laboratory of FAST,Chinese Academy of Sciencessupported by the National Natural Science Foundation of China(NSFC)(11988101,11725313,11690024,and12203045)+11 种基金the Key Research Project of Zhejiang Lab(2021PE0AC03)supported by NSFC(fU2031117)the Youth Innovation Promotion Association CAS(ID.2021055)the Cultivation Project for FAST Scientific Payoff and Research Achievement of CAMS-CASsupported by NSFC(12203069)the National SKA Program of China(2022SKA0130100)the Office Leading Group for Cyberspace Affairs,CAS(CAS-WX2023PY-0102)funded by the Australian Governmentsupported by the National Science Foundation of Xinjiang Uygur Autonomous Region(2022D01D85)the Major Science and Technology Program of Xinjiang Uygur Autonomous Region(2022A03013-2)the Tianchi Talent projectthe CAS Project for Young Scientists in Basic Research(YSBR-063)。
文摘The origin of fast radio bursts(FRBs),the brightest cosmic explosion in radio bands,remains unknown.We introduce here a novel method for a comprehensive analysis of active FRBs’behaviors in the timeenergy domain.Using"Pincus Index"and"Maximum Lyapunov Exponent",we were able to quantify the randomness and chaoticity,respectively,of the bursting events and put FRBs in the context of common transient physical phenomena,such as pulsar,earthquakes,and solar flares.In the bivariate timeenergy domain,repeated FRB bursts’behaviors deviate significantly(more random,less chaotic)from pulsars,earthquakes,and solar flares.The waiting times between FRB bursts and the corresponding energy changes exhibit no correlation and remain unpredictable,suggesting that the emission of FRBs does not exhibit the time and energy clustering observed in seismic events.The pronounced stochasticity may arise from a singular source with high entropy or the combination of diverse emission mechanisms/sites.Consequently,our methodology serves as a pragmatic tool for illustrating the congruities and distinctions among diverse physical processes.
基金Project supported by the National Natural Science Foundation of China(Grant Nos.11974232 and 12374441)the Fund from the Yulin Science and Technology Bureau,China(Grant No.CXY-2022-178).
文摘In a multi-bubble system, the bubble behavior is modulated by the primary acoustic field and the secondary acoustic field. To explore the translational motion of bubbles in cavitation liquids containing high-concentration cavitation nuclei,evolutions of bubbles are recorded by a high-speed camera, and translational trajectories of several representative bubbles are traced. It is found that translational motion of bubbles is always accompanied by the fragmentation and coalescence of bubbles, and for bubbles smaller than 10 μm, the possibility of bubble coalescence is enhanced when the spacing of bubbles is less than 30 μm. The measured signals and their spectra show the presence of strong negative pressure, broadband noise,and various harmonics, which implies that multiple interactions of bubbles appear in the region of high-intensity cavitation.Due to the strong coupling effect, the interaction between bubbles is random. A simplified triple-bubble model is developed to explore the interaction patterns of bubbles affected by the surrounding bubbles. Patterns of bubble interaction, such as attraction, repulsion, stable spacing, and rebound of bubbles, can be predicted by the theoretical analysis, and the obtained results are in good agreement with experimental observations. Mass exchange between the liquid and bubbles as well as absorption in the cavitation nuclei also plays an important role in multi-bubble cavitation, which may account for the weakening of the radial oscillations of bubbles.
基金National Natural Science Foundation of China(Grant No.82003953,82360777)China Postdoctoral Science Foundation(Grant No.2019M662278)+3 种基金National Natural Science Foundation of Jiangxi Province(Grant No.20202BAB216039,20232ACB216015,20212BAB216009 and 20232BAB206166)2020-2022 Young Talents Support Project of Chinese Society of Chinese Medicine(Grant No.2020-QNRC2-07)Jiangxi University of Chinese Medicine Science and Technology Innovation Team Development Program(Grant No.CXTD-22004)Doctoral Research start-up Fund Project of Jiangxi University of Chinese Medicine(Grant No.2021BSZR015,2022BSZR003)。
