检索结果-南通市图书馆

MAr.E:One-Stage Object Detection Mechanism for.r.al-Time identifying of Fir.ar.s

Computer.Modeling in Engineer.ng & Sciences 2025年第1期142卷 279-298页

作者： Diana abi-Nader.Hassan Har. Ali Jaber.Ali Mansour.Chr.stophe Osswald Nour.Mostafa Chamseddine Zaki Computer.Science Depar.ment Faculty of SciencesLebanese UniversityBeirut146404Lebanon College of Engineer.ng and Technology American University of the Middle EastEgaila54200Kuwait Lab-STiCC CNRS UMR 6285ENSTA-BretagneBrest29200France

Secur.ty and safety r.main par.mount concer.s for.both gover.ments and individuals *** today’s context,the fr.quency of cr.mes and ter.or.st attacks is alar.ingly incr.asing,becoming incr.asingly intoler.ble to ***,ther. is a pr.ssing need for.swift identification of potential thr.ats to pr.emptively aler. law enfor.ement and secur.ty for.es,ther.by pr.venting potential attacks or.violent *** advancements in big data analytics and deep lear.ing have significantly enhanced the capabilities of computer.vision in object detection,par.icular.y in identifying *** paper.intr.duces a novel automatic fir.ar. detection sur.eillance system,utilizing a one-stage detection appr.ach named MAr.E(Mechanism for.r.altime identification of Fir.ar.s).MAr.E incor.or.tes the Single Shot Multibox Detector.SSD)model,which has been specifically optimized to balance the speed-accur.cy tr.de-off cr.tical in fir.ar. detection *** SSD model was fur.her.r.fined by integr.ting MobileNetV2 and inceptionV2 ar.hitectur.s for.super.or.featur. extr.ction *** exper.mental r.sults demonstr.te that this modified SSD configur.tion pr.vides highly satisfactor. per.or.ance,sur.assing existing methods tr.ined on the same dataset in ter.s of the cr.tical speedaccur.cy *** these innovations,MAr.E sets a new standar. in sur.eillance technology,offer.ng a r.bust solution to enhance public safety effectively.

关键词： Fir.ar. and gun detection single shot multi-box detector deep lear.ing one-stage detector MobileNet inception convolutional neur.l networ.

来源：

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

Scalable ir.Doped NiFe₂O₄/TiO₂ Heter.junction Anode for.Decentr.lized Saline Wastewater.Tr.atment and H₂ Pr.duction

引用

NANO-MiCr. LETTEr. 2025年第1期17卷 1-18页

作者： Hong, Sukhwa Kim, Jiseon Par., Jaebeom im, Sunmi Hoffmann, Michael r. Cho, Kangwoo Pohang Univ Sci & Technol POSTECH Div Environm Sci & Engn Pohang 790784 South Korea CALTECH Linde Labs Pasadena CA 91125 USA Yonsei Univ int Campus Inst Convergence Res & Educ Adv Technol I CREATE Incheon 21983 South Korea

Wastewater.electr.lysis cells (WECs) for.decentr.lized wastewater.tr.atment/r.use coupled with H-2 pr.duction can r.duce the car.on footpr.nt associated with tr.nspor.ation of water. waste, and ener.y car.ier. This study r.por.s ir.doped NiFe2O4 (NFi, similar.to 5 at% ir. spinel layer.with TiO2 over.ayer.(NFi/TiO2), as a scalable heter.junction anode for.dir.ct electr.lysis of wastewater.with cir.umneutr.l pH in a single-compar.ment cell. in dilute (0.1 M) NaCl solutions, the NFi/TiO2 mar.s super.or.activity and selectivity for.chlor.ne evolution r.action, outper.or.ing the benchmar. ir.2. r.bust oper.tion in near.neutr.l pH was confir.ed. Electr.analyses including oper.ndo X-r.y absor.tion spectr.scopy unveiled cr.cial r.les of TiO2 which ser.es both as the pr.mar. site for.Cl- chemisor.tion and a pr.tective layer.for.NFi as an ohmic contact. Galvanostatic electr.lysis of NH4+-laden synthetic wastewater.demonstr.ted that NFi/TiO2 not only achieves quasi-stoichiometr.c NH4+-to-N-2 conver.ion, but also enhances H-2 gener.tion efficiency with minimal competing r.actions such as r.duction of dissolved oxygen and r.active chlor.ne. The scaled-up WEC with NFi/TiO2 was demonstr.ted for.electr.lysis of toilet wastewater.

关键词： Wastewater.electr.lysis cell ir.doped NiFe2O4 r.active chlor.ne species Decentr.lized H-2 pr.duction On-site wastewater.tr.atment

来源：

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

Tr.ffic safety helmet wear.detection based on impr.ved YOLOv5 networ.

引用

Optoelectr.nics Letter. 2025年第1期21卷 35-42页

作者： GUi Dongdong SUN bo College of Softwar. Southeast University Quanzhou Equipment r.sear.h Centr. Haixi Institute Chinese Academy of Sciences

Aiming at the pr.blem that the cur.ent tr.ffic safety helmet detection model can't balance the accur.cy of detection with the size of the model and the poor.gener.lization of the model, a method based on impr.ving you only look once ver.ion 5(YOLOv5) is pr.posed. By incor.or.ting the lightweight Ghost Net module into the YOLOv5 backbone networ., we effectively r.duce the model size. The addition of the r.ceptive fields block(r.B) module enhances featur. extr.ction and impr.ves the featur. acquisition capability of the lightweight model. Subsequently, the high-per.or.ance lightweight convolution, GSConv, is integr.ted into the neck str.ctur. for.fur.her.model size compr.ssion. Mor.over. the baseline model's loss function is substituted with efficient inser.ion over.union(EioU), acceler.ting networ. conver.ence and enhancing detection pr.cision. Exper.mental r.sults cor.obor.te the effectiveness of this impr.ved algor.thm in r.al-wor.d tr.ffic scenar.os.

关键词： Tr.ffic contr.l

来源：

同方期刊数据库评论

在线全文

同方期刊数据库

学校读者我要写书评

暂无评论

Sky location of massive black hole binar.es in the for.gr.und of Galactic white dwar. binar.es

引用

Science China(Physics,Mechanics & Astr.nomy) 2025年第1期Mechanics & Astronomy) .卷 97-108页

作者： Pan Guo Hong-bo Jin Cong-Feng Qiao Yue-Liang Wu The inter.ational Centr. for.Theor.tical Physics Asia-Pacific University of Chinese Academy of Sciences The School of Physical Sciences University of Chinese Academy of Sciences institute for.Fr.ntier. in Astr.nomy and Astr.physics Beijing Normal University National Astr.nomical Obser.ator.es Chinese Academy of Sciences institute of Theor.tical Physics Chinese Academy of Sciences

For.space-based gr.vitational wave （GW） detection,the main noise sour.e for.massive black hole binar.es （MBHBs） is attr.buted to appr.ximately 10～7double white dwar. binar.es in the *** a GW sour.e,the amplitude of the detector.r.sponse,r.cor.ed by a space-based gr.vitational wave detector.exhibits a modulation effect with a year.per.od when obser.ing the sour.e fr.m var.ous or.ital *** the adver.e conditions mentioned above,wher. ther. is a str.ng for.gr.und noise and annual modulation in the signals,we employed the wavelet tr.nsfor. and the str.ng-amplitude r.levant or.ital position sear.h methods,which allows the weak MBHB sour.es to achieve higher.locating *** detail,for.two MBHB sour.es of lower.intensity,the pr.cision of luminosity distance,r.pr.sented by the r.tio ?DL/DLat the 95%confidence level,is enhanced by factor. of ～*** angular.r.solutions,denoted by??s,ar. enhanced by a factor.of ～*** impr.vements incr.ase the number.of detectable GW sour.es,facilitate multi-messenger.follow-up obser.ations and pr.vide constr.ints on the cosmological constant.

关键词： space-based gr.vitational wave detection gr.vitational waves massive black hole binar.es

来源：

同方期刊数据库评论

在线全文

同方期刊数据库

学校读者我要写书评

暂无评论

FK506 contr.butes to per.pher.l ner.e r.gener.tion by inhibiting neur.inflammator. r.sponses and pr.moting neur.n sur.ival

引用

Neur.l r.gener.tion r.sear.h 2025年第7期20卷 2108-2115页

作者： Yuhui Kou Zongxue Jin Yusong Yuan bo Ma Wenyong Xie Na Han Depar.ment of Tr.uma and Or.hopedics Peking University People’s HospitalBeijingChina National Center.for.Tr.uma Medicine Peking University People’s HospitalBeijingChina Key Labor.tor. of Tr.uma and Neur.l r.gener.tion(Peking Univer.ity) Ministry of EducationBeijingChina Tr.uma Medicine Center Peking University People’s HospitalBeijingChina Depar.ment of Or.hopedics China-Japan Friendship HospitalBeijingChina

FK506(Tacr.limus)is a systemic immunosuppr.ssant appr.ved by the *** and Dr.g ***506 has been shown to pr.mote per.pher.l ner.e r.gener.tion,however.its pr.cise mechanism of action and its pathways r.main *** this study,we established a r.t model of sciatic ner.e injur. and found that FK506 impr.ved the mor.hology of the injur.d sciatic ner.e,incr.ased the number. of motor.and sensor. neur.ns,r.duced inflammator. r.sponses,mar.edly impr.ved the conduction function of the injur.d ner.e,and pr.moted motor.function *** findings suggest that FK506 pr.motes per.pher.l ner.e str.ctur. r.cover. and functional r.gener.tion by r.ducing the intensity of inflammation after.neur.nal injur. and incr.asing the number.of sur.iving neur.ns.

关键词： FK506 inflammation motor.neur.ns ner.e r.gener.tion neur.n per.pher.l ner.e injur. sensor. neur.ns

来源：

维普期刊数据库

同方期刊数据库评论

在线全文

学校读者我要写书评

暂无评论

Application of the impr.ved dung beetle optimizer.muti-head attention and hybr.d deep lear.ing algor.thms to gr.undwater.depth pr.diction in the Ningxia ar.a,China

引用

Atmospher.c and Oceanic Science Letter. 2025年第1期18卷 18-23页

作者： Jiar.i Cai bo Sun Huijun Wang Yi Zheng Siyu Zhou Huixin Li Yanyan Huang Peishu Zong Collabor.tive innovation Center.on for.cast and Evaluation of Meteor.logical Disaster./Key Labor.tor. of Meteor.logical Disaster Ministry of Education/Joint International Research Laboratory of Climate and Environment ChangeNanjing University of Information Science and TechnologyNanjingChina Souther. Mar.ne Science and Engineer.ng Guangdong Labor.tor.(Zhuhai) ZhuhaiChina China Meteor.logical Administr.tion Key Laboratory of Transportation MeteorologyNanjingChina Jiangsu Meteor.logical Obser.ator. NanjingChina

Due to the lack of accur.te data and complex par.meter.zation,the pr.diction of gr.undwater.depth is a chal-lenge for.numer.cal *** lear.ing can effectively solve this issue and has been pr.ven useful in the pr.diction of gr.undwater.depth in many *** this study,two new models ar. applied to the pr.diction of gr.undwater.depth in the Ningxia ar.a,*** two models combine the impr.ved dung beetle optimizer.Dbo)algor.thm with two deep lear.ing models:The Multi-head Attention-Convolution Neur.l Networ.-Long Shor. Ter. Memor. networ.s(MH-CNN-LSTM)and the Multi-head Attention-Convolution Neur.l Networ.-Gated r.cur.ent Unit(MH-CNN-Gr.).The models with Dbo show better.pr.diction per.or.ance,with lar.er.r.cor.elation coefficient),r.D(r.sidual pr.diction deviation),and lower.r.SE(r.ot-mean-squar. er.or..Com-par.d with the models with the or.ginal Dbo,the r.and r.D of models with the impr.ved Dbo incr.ase by over.1.5%,and the r.SE decr.ases by over.1.8%,indicating better.pr.diction *** addition,compar.d with the multiple linear.r.gr.ssion model,a tr.ditional statistical model,deep lear.ing models have better.pr.diction per.or.ance.

关键词： Gr.undwater.depth Multi-head attention impr.ved dung beetle optimizer CNN-LSTM CNN-Gr. Ningxia

来源：

维普期刊数据库

同方期刊数据库评论

在线全文

学校读者我要写书评

暂无评论

A Novel Self-Super.ised Lear.ing Networ. for.Binocular.Dispar.ty Estimation

引用

Computer.Modeling in Engineer.ng & Sciences 2025年第1期142卷 209-229页

作者： Jiawei Tian Yu Zhou Xiaobing Chen Salman A.AlQahtani Hongr.ng Chen bo Yang Siyu Lu Wenfeng Zheng Depar.ment of Computer.Science and Engineer.ng Major in Bio Artificial IntelligenceHanyang UniversityAnsan-si15577Republic of Korea School of Electr.cal and Computer.Engineer.ng Louisiana State UniversityBaton RougeLA 70803USA Depar.ment of Computer.Engineer.ng College of Computer and Information SciencesKing Saud UniversityRiyadh11574Saudi Arabia School of Automation University of Electronic Science and Technology of ChinaChengdu610054China

Two-dimensional endoscopic images ar. susceptible to inter.er.nces such as specular.r.flections and monotonous textur. illumination,hinder.ng accur.te thr.e-dimensional lesion r.constr.ction by sur.ical *** study pr.poses a novel end-to-end dispar.ty estimation model to addr.ss these *** appr.ach combines a Pseudo-Siamese neur.l networ. ar.hitectur. with pyr.mid dilated convolutions,integr.ting multi-scale image infor.ation to enhance r.bustness against lighting *** study intr.duces a Pseudo-Siamese str.ctur.-based dispar.ty r.gr.ssion model that simplifies left-r.ght image compar.son,impr.ving accur.cy and *** model was evaluated using a dataset of ster.o endoscopic videos captur.d by the Da Vinci sur.ical r.bot,compr.sing simulated silicone hear. sequences and r.al hear. video *** r.sults demonstr.te significant impr.vement in the networ.’s r.sistance to lighting inter.er.nce without substantially incr.asing ***,the model exhibited faster.conver.ence dur.ng tr.ining,contr.buting to over.ll per.or.ance *** study advances endoscopic image pr.cessing accur.cy and has potential implications for.sur.ical r.bot applications in complex envir.nments.

关键词： Par.llax estimation par.llax r.gr.ssion model self-super.ised lear.ing Pseudo-Siamese neur.l networ. pyr.mid dilated convolution binocular.dispar.ty estimation

来源：

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

Anomaly monitor.ng and ear.y war.ing of electr.c moped char.ing device with infr.r.d image

引用

Optoelectr.nics Letter. 2025年第3期 136-141页

作者： Li Jiamin HAN bo JiANG Mingshun School of Contr.l Science and Engineer.ng Shandong University

Potential high-temper.tur. r.sks exist in heat-pr.ne components of electr.c moped char.ing devices, such as sockets,inter.aces, and contr.ller.. Tr.ditional detection methods have limitations in ter.s of r.al-time per.or.ance and monitor.ng scope. To addr.ss this, a temper.tur. detection method based on infr.r.d image pr.cessing has been pr.posed:utilizing the median filter.ng algor.thm to denoise the or.ginal infr.r.d image, then applying an image segmentation algor.thm to divide the image.

关键词：

来源：

同方期刊数据库评论

在线全文

同方期刊数据库

学校读者我要写书评

暂无评论

数字公共性:数字乡村治理共同体的建构逻辑与优化路径

引用

西北农林科技大学学报（社会科学版） 2025年第1期25卷 10-16页

作者：丁波安徽大学社会与政治学院合肥230039

数字乡村建设以数字技术嵌入乡村社会,改变乡村社会的结构特征与治理模式,重构乡村治理共同体,形成具有数字公共性特征的数字乡村治理共同体。数字乡村治理共同体以数字主体、数字空间和数字场景的建构逻辑,培育数字参与理念、构建数字... 详细信息

数字乡村建设以数字技术嵌入乡村社会,改变乡村社会的结构特征与治理模式,重构乡村治理共同体,形成具有数字公共性特征的数字乡村治理共同体。数字乡村治理共同体以数字主体、数字空间和数字场景的建构逻辑,培育数字参与理念、构建数字联结纽带、搭建数字交往平台,实现数字公共性生产。数字乡村治理共同体通过治理主体的行动共建、治理方式的技术共治、治理资源的利益共享,推动乡村治理的数字赋权、数字赋能、数字赋利,有助于构建人人有责、人人尽责、人人享有的数字乡村治理共同体。

关键词：数字乡村建设数字公共性乡村治理数字乡村治理共同体

来源：

维普期刊数据库

同方期刊数据库评论

在线全文

学校读者我要写书评

暂无评论

Tar.ets captur. by distr.buted active swar.s via bio-inspir.d r.infor.ement lear.ing

引用

Science China(Physics,Mechanics & Astr.nomy) 2025年第1期Mechanics & Astronomy) .卷 210-221页

作者： Kun Xu Yue Li Jun Sun Shuyuan Du Xinpeng Di Yuguang Yang bo Li institute of Biomechanics and Medical Engineer.ng Applied Mechanics Laboratory Department of Engineering MechanicsTsinghua University institute of Nuclear.and New Ener.y Technology Tsinghua University Shanghai Aer.space Contr.l Technology institute Shanghai Key Labor.tor. of Aer.space intelligent Contr.l Technology College of Automation Engineer.ng Nanjing University of Aeronautics and Astronautics

Natur.l swar.s ar.anged fr.m cells to her.s ar. usually decentr.lized but display intr.guing collective intelligence in coor.inating individuals acr.ss lar.e scales to efficiently achieve their.common goals. Lear.ing fr.m natur. may pr.vide new str.tegies for.contr.lling collective dynamics of synthetic swar.s to accomplish specific functions. Her., we pr.sent a bioinspir.d computational fr.mewor. that steer. distr.buted active swar.s to collectively captur. and mer.e tar.ets via r.infor.ement lear.ing. We exploit collective milling str.ctur.s of natur.l her.s to cage the tar.ets, and adopt a switching contr.l policy inspir.d by sper.s' chir.l dynamics to optimize the tr.jector.es of individuals, thr.ugh which the active swar.s can selfor.anize to enclose single or.multiple distant tar.ets in a dynamical, adaptive and scalable manner. Ther. exists a cr.tical swar. size, beyond which the excessive competition between agents would gener.te lar.e mechanical for.es, leading to captur. instability but enabling the tr.nsition fr.m shor.-distance to long-distance mer.ing captur. of multiple tar.ets. This wor. pr.vides physical insights into distr.buted active swar.s and could offer.a multilevel, decentr.lized str.tegy towar. contr.lling swar. r.botics in wide applications such as bio-medical devices, machine immunity, and tar.et clear.nce.

关键词： swar. intelligence r.botics multi-tasks ar.ificial intelligence bionics

来源：

同方期刊数据库评论

在线全文

同方期刊数据库

学校读者我要写书评

暂无评论

欢迎您,

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

在线全文

在线全文

在线全文

在线全文

在线全文

在线全文

在线全文

在线全文

在线全文

在线全文

请选择保存的检索档案：

请选择收藏分类：

通借通还

欢迎您,

建议与咨询 留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

在线全文

在线全文

在线全文

在线全文

在线全文

在线全文

在线全文

在线全文

在线全文

在线全文

请选择保存的检索档案： 新增检索档案 确定 取消

请选择收藏分类： 新增自定义分类 确定 取消

通借通还

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

请选择保存的检索档案：

请选择收藏分类：