检索结果-南通市图书馆

CVTD: A Robust Car-Mounted Video text Detector

维普期刊数据库评论

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

Computers, Materials & Continua 2024年第2期78卷 1821-1842页

作者： Di zhou Jianxun Zhang chao Li Yifan Guo Bowen Li Department of Computer Science and Engineering Chongqing University of TechnologyChongqingChina College of Information and Engineering Jingdezhen Ceramic UniversityJingdezhenChina

text perception is crucial for understanding the semantics of outdoor scenes,making it a key requirement for building intelligent systems for driver assistance or autonomous *** information in car-mounted videos can assist drivers in making ***,Car-mounted video text images pose challenges such as complex backgrounds,small fonts,and the need for real-time *** proposed a robust Car-mounted Video text Detector(CVTD).It is a lightweight text detection model based on ResNet18 for feature extraction,capable of detecting text in arbitrary *** model efficiently extracted global text positions through the Coordinate Attention Threshold Activation(CATA)and enhanced the representation capability through stacking two Feature Pyramid Enhancement Fusion Modules(FPEFM),strengthening feature representation,and integrating text local features and global position information,reinforcing the representation capability of the CVTD *** enhanced feature maps,when acted upon by text Activation Maps(TAM),effectively distinguished text foreground from non-text ***,we collected and annotated a dataset containing 2200 images of Car-mounted Video text(CVT)under various road conditions for training and evaluating our model’s *** further tested our model on four other challenging public natural scene text detection benchmark datasets,demonstrating its strong generalization ability and real-time detection *** model holds potential for practical applications in real-world scenarios.

关键词： Deep learning text detection Car-mounted video text detector intelligent driving assistance arbitrary shape text detector

Low Resource Chinese Geological text Named Entity Recognition Based on Prompt Learning

维普期刊数据库评论

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

Journal of Earth Science 2024年第3期35卷 1035-1043页

作者： Hang He chao Ma Shan Ye Wenqiang Tang Yuxuan zhou Zhen Yu Jiaxin Yi Li Hou Mingcai Hou State Key Laboratory of Oil and Gas Reservoir Geology and Exploitation Institute of Sedimentary GeologyCollege of Computer Science and Cyber SecurityChengdu University of TechnologyChengdu 610059China Key Laboratory of Deep-Time Geography and Environment Reconstruction and Applications of Ministry of Natural Resources Chengdu University of TechnologyChengdu 610059China School of Information Engineering China University of Geosciences(Beijing)Beijing 100083China

Geological reports are a significant accomplishment for geologists involved in geological investigations and scientific research as they contain rich data and textual *** the rapid development of science and technology,a large number of textual reports have accumulated in the field of ***,many non-hot topics and non-English speaking regions are neglected in mainstream geoscience databases for geological information mining,making it more challenging for some researchers to extract necessary information from these *** Language Processing(NLP)has obvious advantages in processing large amounts of textual *** objective of this paper is to identify geological named entities from Chinese geological texts using NLP *** propose the RoBERTa-Prompt-Tuning-NER method,which leverages the concept of Prompt Learning and requires only a small amount of annotated data to train superior models for recognizing geological named entities in low-resource dataset *** RoBERTa layer captures context-based information and longer-distance dependencies through dynamic word ***,we conducted experiments on the constructed Geological Named Entity Recognition(GNER)*** experimental results show that the proposed model achieves the highest F1 score of 80.64%among the four baseline algorithms,demonstrating the reliability and robustness of using the model for Named Entity Recognition of geological texts.

关键词： Prompt Learning Named Entity Recognition(NER) low resource geological text text information mining big data geology.

Prospect of ultrahigh-resolution fast neutron absorption spectroscopy based on a laser plasma electron accelerator

在线全文

学校读者我要写书评

暂无评论

High Power Laser Science and Engineering 2024年第1期12卷 108-115页

作者： Wen-Zhao Wang Jie Feng Xiao-Peng Zhang Yao-Jun Li Wei-Jun zhou Wen-chao Yan Guo-Qiang Zhang Chang-Bo Fu Li-Ming Chen Key Laboratory of Laser Plasma(MoE) School of Physics and AstronomyShanghai Jiao Tong UniversityShanghaiChina IFSA Collaborative Innovation Center Shanghai Jiao Tong UniversityShanghaiChina Institute of High Energy Physics Chinese Academy of SciencesBeijingChina Shanghai Advanced Research Institute Chinese Academy of SciencesShanghaiChina Key Laboratory of Nuclear Physics and Ion-beam Application(MoE) Institute of Modern PhysicsFudan UniversityShanghaiChina

Fast neutron absorption spectroscopy is widely used in the study of nuclear structure and element analysis. However,due to the traditional neutron source pulse duration being of the order of nanoseconds, it is difficult to obtain a high-resolution absorption spectrum. Thus, we present a method of ultrahigh energy-resolution absorption spectroscopy via a high repetition rate, picosecond duration pulsed neutron source driven by a terawatt laser. The technology of single neutron count is used, which results in easily distinguishing the width of approximately 20 keV at 2 MeV and an asymmetric shape of the neutron absorption peak. The absorption spectroscopy based on a laser neutron source has one order of magnitude higher energy-resolution power than the state-of-the-art traditional neutron sources, which could be of benefit for precisely measuring nuclear structure data.

关键词： fast neutron absorption spectroscopy laser plasma accelerator photo-nuclear neutrons single neutron count

YOUTH BLOOMS ON THE PLATEAU

在线全文

学校读者我要写书评

暂无评论

China's Tibet 2021年第6期32卷 53-58页

作者： zhou Qingqing(text/photo) Dekyi Drolma(text/photo) Huang Wenjuan(translated) 不详

When looking at the remarkable attainments regarding the absolute eradication of poverty throughout China,it would be worth to mention the contributions of the hundreds of thousands of hardworking grassroots cadres.A great number of youths who served in the naturally abundant but economically relatively disadvantaged Tibet Autonomous Region are of particular note,and many of them fall within the post-90 age *** for their boundless aspiration and energy,these young men and women chose Tibet as their point of deployment through whatever organization or enterprise they were part of.

关键词： hundreds disadvantage looking

State-of-the-art advancements of transition metal oxides as photoelectrode materials for solar water splitting

维普期刊数据库评论

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

Rare Metals 2022年第7期41卷 2370-2386页

作者： Gai-Li Ke Bi Jia Hui-chao He Yong zhou Ming zhou State Key Laboratory of Environmental-Friendly Energy Materials School of Materials Science and EngineeringSouthwest University of Science and TechnologyMianyang 621010China Institute of Environmental Energy Materials and Intelligent Devices School of Metallurgy and Materials EngineeringChongqing University of Science and TechnologyChongqing 401331China Ecomaterials and Renewable Energy Research Center School of PhysicsNanjing UniversityNanjing 211102China Department of Chemistry State University of New YorkBinghamton NY 13902USA

photoelectrochemical(PEC)water splitting can convert renewable solar energy into clean hydrogen *** are the core components of water-splitting *** the past 40 years,a series of binary and ternary transition metal oxides have been investigated as photo-electrode materials for solar water splitting,and numerous studies have been carried out to modify their water-split-ting *** satisfactory transition metal oxide photoelectrode materials have not been found,it is necessary to summarize the recent advancements in tran-sition metal oxide photoelectrode materials to guide future *** this review,the background and principle of PEC water splitting are *** semiconductor properties and modification progress of typical binary and ternary metal oxide photoanodes and photocathodes for solar water splitting are *** on the newly developed strategies in recent years,a brief outlook is presented for efficient PEC water splitting using transition metal oxide photoelectrodes.

关键词： Transition metal oxides photoanodes photocathodes Water splitting Review

在线全文

学校读者我要写书评

暂无评论

基于文献计量分析土壤无机碳研究现状与趋势

土壤通报 2024年第4期55卷 1174-1184页

作者：祝慧陶静静王明达周建斌范利超西北农林科技大学资源环境学院陕西杨凌712100 农业农村部西北植物营养与农业环境重点实验室陕西杨凌712100

土壤无机碳在固碳、维持土壤质量、促进有机碳稳定性与肥力和生态系统服务功能等方面有重要作用。以Web of Science核心合集数据库和CNKI数据库中472篇土壤无机碳研究的文献为数据源,通过VOSviewer、CiteSpace和HistCite分别从发文量、... 详细信息

土壤无机碳在固碳、维持土壤质量、促进有机碳稳定性与肥力和生态系统服务功能等方面有重要作用。以Web of Science核心合集数据库和CNKI数据库中472篇土壤无机碳研究的文献为数据源,通过VOSviewer、CiteSpace和HistCite分别从发文量、被引频次、合作强度及关键词等角度对发文国家、机构、作者、研究热点及其变化趋势进行计量分析与可视化展示。结果表明:2003~2023 a期间土壤无机碳研究的发文数量不断增长,主要从土壤学与环境学向其他学科延伸;科研机构以中国科学院为首,而发文作者形成了以Kuzyakov Y为核心的作者群,总体而言我国的发文量和国际影响力最大。当前的研究热点集中在土壤无机碳储量及其影响因素,土壤无机碳在碳循环中的作用及其固碳潜力,土壤无机碳的生态系统服务功能等方面。然而,当前对碳酸盐再沉淀过程中有机碳向无机碳的转移及固碳作用研究不足,尤其是植物光合同化碳向土壤次生碳酸盐累积转化及对氮肥的响应机制尚不清楚。今后应通过实地研究收集无机碳数据并结合机器学习建模等方法,提高估算全球碳库及碳周转速率的准确性;深化土壤有机碳-二氧化碳-土壤无机碳的转化机理,进一步明晰无机碳对稳定土壤碳库和维持土壤健康等生态服务方面的作用。

关键词：土壤无机碳次生碳酸盐文献计量研究进展未来趋势

在线全文

学校读者我要写书评

暂无评论

基于射线追踪的微波成像诊断极向位置修正

核聚变与等离子体物理 2024年第2期44卷 170-176页

作者：高远杨州军张智超罗蔷金子阳高丽潘垣华中科技大学电气与电子工程学院强电磁工程与新技术国家重点实验室武汉430074

在微波成像诊断准光学设计中对等离子体介质对成像位置的影响进行了研究。采用射线追踪方法,在J-text实验参数条件下,基于冷等离子体近似的色散关系,求解了电子回旋辐射成像(ECEI)诊断的极向成像位置。对比分析了射线追踪相较于传统准... 详细信息

在微波成像诊断准光学设计中对等离子体介质对成像位置的影响进行了研究。采用射线追踪方法,在J-text实验参数条件下,基于冷等离子体近似的色散关系,求解了电子回旋辐射成像(ECEI)诊断的极向成像位置。对比分析了射线追踪相较于传统准光学成像的成像位置的修正效果,发现聚焦位置位于高场侧和高密度等离子体情况下,实际观测位置与准光学设计位置有较大偏差。该修正提高了诊断数据测量的可靠性,优化了微波诊断信号分析。

关键词：电子回旋辐射成像准光学设计射线追踪

Fish Feeding Behavior Recognition Using Adaptive DMCA-UMT Algorithm

在线全文

学校读者我要写书评

暂无评论

Journal of Beijing Institute of Technology 2023年第3期32卷 285-297页

作者： Caiwei Yang Xinting Yang Kaijie Zhu chao zhou National Engineering Research Center for Information Technology in Agriculture Beijing 100097China Information Technology Research Center Beijing Academy of Agriculture and Forestry SciencesBeijing 100097China National Engineering Laboratory for Agri-product Quality Traceability Beijing 100097China College of Computer and Information Engineering Tianjin Agricultural UniversityTianjin 300384China

Realtime analyzing the feeding behavior of fish is the premise and key to accurate guidance on *** identification of fish behavior using a single information is susceptible to various *** overcome the problems,this paper proposes an adaptive deep modular co-attention unified multi-modal transformers(DMCA-UMT).By fusing the video,audio and water quality parameters,the whole process of fish feeding behavior could be ***,for the input video,audio and water quality parameter information,features are extracted to obtain feature vectors of different ***,deep modular co-attention(DMCA)is introduced on the basis of the original cross-modal encoder,and the adaptive learnable weights are *** feature vector of video and audio joint representation is obtained by automatic learning based on fusion ***,the information of visual-audio modality fusion and text features are used to generate clip-level moment *** query decoder decodes the input features and uses the prediction head to obtain the final joint moment retrieval,which is the start-end time of feeding the *** results show that the mAP Avg of the proposed algorithm reaches 75.3%,which is37.8%higher than that of unified multi-modal transformers(UMT)algorithm.

关键词： aquaculture multi-modal fusion deep modular co-attention(DMCA) unified multimodal transformers(UMT) video moment retrieval

Lignin-based electrospun nanofiber membrane decorated with photo-Fenton Ag@MIF-100(Fe) heterojunctions for complex wastewater remediation

在线全文

学校读者我要写书评

暂无评论

Frontiers of Chemical Science and Engineering 2023年第7期17卷 930-941页

作者： Guodong Tian chao Duan Bingxu zhou chaochao Tian Qiang Wang Jiachuan Chen State Key Laboratory of Biobased Material and Green Papermaking Qilu University of TechnologyShandong Academy of SciencesJinan 250353China College of Bioresources Chemical and Materials Engineering Shaanxi University of Science and TechnologyXi’an 710021China

Membrane technology for wastewater remediation has aroused wide interest owing to its unique properties and potential ***,it remains challenging to explore green,efficient and robust membrane material and technique for complex wastewater ***,we proposed using a simple electrospinning and in situ seeding method to fabricate a lignin-based electrospun nanofiber membrane(LENM)decorated with photo-Fenton Ag@MIL-100(Fe)heterojunctions for efficient separation of oil/water emulsions and degradation of organic *** to the embedded lignin in LENM,an ultrahigh MIL-100(Fe)loading(53 wt%)with good wettability and high porosity was *** a result,the hybrid Ag@MIL-100(Fe)/LENM exhibited excellent oil/water emulsions separation efficiency(more than 97%)without a compromise of water ***,the hybrid membrane showed an excellent dye removal with degradation of 99%methylene blue within 30 min under illumination,which is attributed to a synergy of dye adsorption/enrichment and photo-Fenton catalytic degradation from Ag@MIL-100(Fe).Therefore,the lignin-based photo-Fenton hybrid membrane can lay the foundation for the preparation and application of green,sustainable and versatile membrane materials and technologies for efficient complex wastewater remediation.

关键词： lignin electrospinning heterojunctions photo-Fenton catalysis wastewater remediation