期刊文献+
共找到522,673篇文章
< 1 2 250 >
每页显示 20 50 100
Therapeutic potential of traditional Chinese medicine for the treatment of chemotherapy-induced diarrhea:clinical efficacy and underlying mechanisms
1
作者 Yun-Jing He Wei-Jian Chen Ke Nie 《Traditional Medicine Research》 2026年第1期75-99,共25页
Chemotherapy-induced diarrhea(CID)is a major concern for cancer patients and is associated with significant morbidity and mortality.Currently,the clinical management of CID is limited.The utilization of antidiarrheal ... Chemotherapy-induced diarrhea(CID)is a major concern for cancer patients and is associated with significant morbidity and mortality.Currently,the clinical management of CID is limited.The utilization of antidiarrheal medications,such as loperamide and octreotide,is relatively limited because of their unsatisfactory efficacy and adverse effects.In recent years,traditional Chinese medicine(TCM)has attracted great interest because of its beneficial effect in treating CID,which has multitarget and low-toxicity therapeutic characteristics.TCM exhibits remarkable therapeutic potential in the prevention and treatment of CID.It can alleviate and treat CID by regulating chemical drug metabolism,improving the integrity of the intestinal barrier,stimulating proliferation while suppressing the apoptosis of intestinal epithelial cells,ameliorating oxidative stress and inflammation and regulating bile acids and aquaporins.However,large-scale,randomized,double-blind clinical trials of TCM for the treatment of CID are lacking,and most preclinical experiments have not been translated to clinical trials.Accordingly,this review highlights the clinical efficacy and molecular mechanisms of TCM against CID via PubMed,Web of Science and China National Knowledge Infrastructure and proposes that future research on TCM against CID should focus on strengthening the connection from bench to bed,which may help to comprehensively evaluate the therapeutic potential of TCM against CID. 展开更多
关键词 traditional Chinese medicine chemotherapy-induced diarrhea clinical efficacy pharmacological mechanism
暂未订购
Integrating high-resolution mass spectrometry and transcriptomics to explore the therapeutic mechanism of Sanhuang Oil in diabetic foot
2
作者 Ping Sun Yu-Feng Zhang +4 位作者 Shuang Li Wei Zhang Peng-Fei Zhao Chen-Xia Li Chen-Ning Zhang 《Traditional Medicine Research》 2026年第1期19-38,共20页
Background:Diabetic foot,a severe complication of diabetes,is characterized by chronic refractory wounds.Sanhuang Oil,a topical herbal formula,demonstrates significant therapeutic effects including antibacterial,anti-... Background:Diabetic foot,a severe complication of diabetes,is characterized by chronic refractory wounds.Sanhuang Oil,a topical herbal formula,demonstrates significant therapeutic effects including antibacterial,anti-inflammatory,and immunomodulatory activities.However,its active constituents and mechanisms of action against diabetic foot remain to be elucidated.Methods:In this study,the chemical constituents of Sanhuang Oil were identified using UPLC-QE-Orbitrap-MS.Subsequently,the mechanism by which Sanhuang Oil promotes diabetic foot ulcer healing was predicted by integrating network pharmacology and molecular docking.Additionally,diabetic mouse model was established in ICR mice using a combination of a high-fat diet(HFD)and streptozotocin(STZ)chemical induction.A full-thickness skin defect was created on the dorsum of the mice.Wound healing and the healing rate were observed following Sanhuang Oil intervention.The mechanism underlying Sanhuang Oil’s promotion of diabetic ulcer healing was further investigated using transcriptomics and histopathological examination(H&E staining).Results:A total of 97 active ingredients were identified from Sanhuang Oil.Network pharmacology analysis predicted 543 common targets,and Kyoto Encyclopedia of Genes and Genomes(KEGG)pathway enrichment analysis identified 203 relevant pathways.Molecular docking further confirmed high binding affinity(binding energy≤−5.0 kcal/mol)between specific active components in Sanhuang Oil(e.g.,coptisine,phellodendrine,baicalein)and key targets associated with diabetic foot ulcers(e.g.,EGFR,AKT1,STAT3).In vivo experimental results demonstrated that the wound healing rate was significantly higher in Sanhuang Oil-treated groups compared to the model group(P<0.001).HE staining revealed that the high-dose Sanhuang Oil group exhibited more pronounced epithelial tissue coverage over the wound,reduced inflammatory cell infiltration,and increased collagen deposition and fibroblast proliferation.transcriptomic analysis identified Pdk4,Ttn,Csrp3,Actn2,Myoz2,Tnnc2,Myod1,Myog,Myot,and Myf6 as key regulatory proteins involved in promoting wound healing.Conclusion:Sanhuang Oil promotes wound healing in diabetic ulcer mice,potentially by mitigating inflammation and regulating key targets such as Pdk4 to enhance fibroblast function.These findings provide novel insights into the multi-target,multi-pathway mechanism of Sanhuang Oil for treating diabetic foot ulcers. 展开更多
关键词 Sanhuang Oil diabetic foot high-resolution mass spectrometry molecular network analysis mechanism of action
暂未订购
Recent Advances in Regulation Strategy and Catalytic Mechanism of Bi-Based Catalysts for CO_(2) Reduction Reaction
3
作者 Jianglong Liu Yunpeng Liu +5 位作者 Shunzheng Zhao Baotong Chen Guang Mo Zhongjun Chen Yuechang Wei Zhonghua Wu 《Nano-Micro Letters》 2026年第1期647-697,共51页
Using photoelectrocatalytic CO_(2) reduction reaction(CO_(2)RR)to produce valuable fuels is a fascinating way to alleviate environmental issues and energy crises.Bismuth-based(Bi-based)catalysts have attracted widespr... Using photoelectrocatalytic CO_(2) reduction reaction(CO_(2)RR)to produce valuable fuels is a fascinating way to alleviate environmental issues and energy crises.Bismuth-based(Bi-based)catalysts have attracted widespread attention for CO_(2)RR due to their high catalytic activity,selectivity,excellent stability,and low cost.However,they still need to be further improved to meet the needs of industrial applications.This review article comprehensively summarizes the recent advances in regulation strategies of Bi-based catalysts and can be divided into six categories:(1)defect engineering,(2)atomic doping engineering,(3)organic framework engineering,(4)inorganic heterojunction engineering,(5)crystal face engineering,and(6)alloying and polarization engineering.Meanwhile,the corresponding catalytic mechanisms of each regulation strategy will also be discussed in detail,aiming to enable researchers to understand the structure-property relationship of the improved Bibased catalysts fundamentally.Finally,the challenges and future opportunities of the Bi-based catalysts in the photoelectrocatalytic CO_(2)RR application field will also be featured from the perspectives of the(1)combination or synergy of multiple regulatory strategies,(2)revealing formation mechanism and realizing controllable synthesis,and(3)in situ multiscale investigation of activation pathways and uncovering the catalytic mechanisms.On the one hand,through the comparative analysis and mechanism explanation of the six major regulatory strategies,a multidimensional knowledge framework of the structure-activity relationship of Bi-based catalysts can be constructed for researchers,which not only deepens the atomic-level understanding of catalytic active sites,charge transport paths,and the adsorption behavior of intermediate products,but also provides theoretical guiding principles for the controllable design of new catalysts;on the other hand,the promising collaborative regulation strategies,controllable synthetic paths,and the in situ multiscale characterization techniques presented in this work provides a paradigm reference for shortening the research and development cycle of high-performance catalysts,conducive to facilitating the transition of photoelectrocatalytic CO_(2)RR technology from the laboratory routes to industrial application. 展开更多
关键词 Bismuth-based catalysts CO_(2)reduction reaction Regulation strategy Catalytic mechanism REVIEW
在线阅读 下载PDF
Towards mechanism-based tau-targeted therapies
4
作者 Lidia Bakota Roland Brandt 《Neural Regeneration Research》 2026年第2期687-688,共2页
Tau plays a crucial role in several neurodegenerative diseases,collectively referred to as tauopathies.Therefore,targeting potential pathological changes in tau could enable useful therapeutic interventions.However,ta... Tau plays a crucial role in several neurodegenerative diseases,collectively referred to as tauopathies.Therefore,targeting potential pathological changes in tau could enable useful therapeutic interventions.However,tau is not an easy target because it dynamically interacts with microtubules and other cellular components,which presents a challenge for tau-targeted drugs.New cellular models could aid the development of mechanism-based tau-targeted therapies. 展开更多
关键词 tau targeted therapies cellular components mechanism based therapies cellular componentswhich cellular models MICROTUBULES TAUOPATHIES neurodegenerative diseasescollectively
暂未订购
Regulatory T cells in neurological disorders and tissue regeneration:Mechanisms of action and therapeutic potentials
5
作者 Jing Jie Xiaomin Yao +5 位作者 Hui Deng Yuxiang Zhou Xingyu Jiang Xiu Dai Yumin Yang Pengxiang Yang 《Neural Regeneration Research》 2026年第4期1277-1291,共15页
Regulatory T cells,a subset of CD4^(+)T cells,play a critical role in maintaining immune tolerance and tissue homeostasis due to their potent immunosuppressive properties.Recent advances in research have highlighted t... Regulatory T cells,a subset of CD4^(+)T cells,play a critical role in maintaining immune tolerance and tissue homeostasis due to their potent immunosuppressive properties.Recent advances in research have highlighted the important therapeutic potential of Tregs in neurological diseases and tissue repair,emphasizing their multifaceted roles in immune regulation.This review aims to summarize and analyze the mechanisms of action and therapeutic potential of Tregs in relation to neurological diseases and neural regeneration.Beyond their classical immune-regulatory functions,emerging evidence points to non-immune mechanisms of regulatory T cells,particularly their interactions with stem cells and other non-immune cells.These interactions contribute to optimizing the repair microenvironment and promoting tissue repair and nerve regeneration,positioning non-immune pathways as a promising direction for future research.By modulating immune and non-immune cells,including neurons and glia within neural tissues,Tregs have demonstrated remarkable efficacy in enhancing regeneration in the central and peripheral nervous systems.Preclinical studies have revealed that Treg cells interact with neurons,glial cells,and other neural components to mitigate inflammatory damage and support functional recovery.Current mechanistic studies show that Tregs can significantly promote neural repair and functional recovery by regulating inflammatory responses and the local immune microenvironment.However,research on the mechanistic roles of regulatory T cells in other diseases remains limited,highlighting substantial gaps and opportunities for exploration in this field.Laboratory and clinical studies have further advanced the application of regulatory T cells.Technical advances have enabled efficient isolation,ex vivo expansion and functionalization,and adoptive transfer of regulatory T cells,with efficacy validated in animal models.Innovative strategies,including gene editing,cell-free technologies,biomaterial-based recruitment,and in situ delivery have expanded the therapeutic potential of regulatory T cells.Gene editing enables precise functional optimization,while biomaterial and in situ delivery technologies enhance their accumulation and efficacy at target sites.These advancements not only improve the immune-regulatory capacity of regulatory T cells but also significantly enhance their role in tissue repair.By leveraging the pivotal and diverse functions of Tregs in immune modulation and tissue repair,regulatory T cells–based therapies may lead to transformative breakthroughs in the treatment of neurological diseases. 展开更多
关键词 demyelinating diseases gene editing immune regulation immune tolerance neural regeneration neurological diseases non-immune mechanisms regulatory T cells stem cells STROKE tissue homeostasis tissue repair
暂未订购
基于I_CBAM-DenseNet模型的小麦发育期识别研究
6
作者 付景枝 马悦 +4 位作者 宏观 刘云平 吴文宇 丁明明 尹泽凡 《南京信息工程大学学报》 北大核心 2025年第1期42-52,共11页
针对我国农作物发育期人工观测效率低、识别准确率不高等问题,提出一种基于I_CBAM-DenseNet模型的小麦发育期识别方法.该方法以密集连接卷积网络(DenseNet)为主干提取网络,融入卷积块注意模块CBAM.先将CBAM中的空间注意力模块(SAM)与通... 针对我国农作物发育期人工观测效率低、识别准确率不高等问题,提出一种基于I_CBAM-DenseNet模型的小麦发育期识别方法.该方法以密集连接卷积网络(DenseNet)为主干提取网络,融入卷积块注意模块CBAM.先将CBAM中的空间注意力模块(SAM)与通道注意力模块(CAM)由传统的串联连接改为并行连接,并将改进的CBAM(I_CBAM)插入到DenseNet最后一个密集网络中,构建一种I_CBAM-DenseNet模型,再选取小麦7个重要发育时期进行自动识别.为最大化提取小麦的特征信息,将超绿特征(ExG)因子和最大类间方差法(Otsu)相结合对采集到的小麦图像进行分割处理.对比分析了I_CBAM-DenseNet、AlexNet、ResNet、DenseNet、CBAM-DenseNet以及VGG等模型的准确率和损失值的变化.结果表明,采取基于I_CBAM-DenseNet的卷积神经网络建立的模型,准确率达到99.64%,高于对比模型. 展开更多
关键词 小麦 发育期 DenseNet 卷积块注意模块(cbam)
在线阅读 下载PDF
一种基于CBAM注意力机制优化YOLOv8n的滑坡检测方法
7
作者 瞿伟 宫明利 +3 位作者 徐荣堂 陈沛男 李久元 唐兴友 《应用基础与工程科学学报》 北大核心 2025年第5期1231-1238,共8页
高效准确的滑坡检测方法对于滑坡灾害的防灾预警具有重要的参考价值.当前滑坡目标检测研究方法易受复杂背景和小目标特征干扰,尤其在高分辨率遥感影像检测中,上述问题更加突出.鉴于此,发展了一种基于CBAM(Convolutional Block Attention... 高效准确的滑坡检测方法对于滑坡灾害的防灾预警具有重要的参考价值.当前滑坡目标检测研究方法易受复杂背景和小目标特征干扰,尤其在高分辨率遥感影像检测中,上述问题更加突出.鉴于此,发展了一种基于CBAM(Convolutional Block Attention Module)注意力机制优化YOLOv8n模型的高分辨率遥感影像滑坡目标检测方法(YOLOv8n-CBAM),并利用高分辨率遥感影像数据集对该方法进行了有效性验证.研究结果表明:(1)在复杂地形、植被等场景下对滑坡的检测,YOLOv8n-CBAM模型可有效提高模型对关键特征的关注度,从而提升小目标滑坡的检测精度,并且显著改善YOLOv8n模型存在的漏检和误检现象;(2)混淆矩阵归一化、精确率-召回率曲线、Loss曲线结果均表明,YOLOv8n-CBAM模型在滑坡目标的检测能力、准确性和鲁棒性方面均显著高于YOLOv8n模型;(3)相较于其他不同的目标检测方法,YOLOv8n-CBAM模型在准确率、召回率、mAP@0.5、mAP@0.5:0.95这4个评价指标中均表现出最优的检测效果. 展开更多
关键词 滑坡检测 YOLOv8n cbam 高分辨率遥感影像 目标检测 深度学习 注意力机制
原文传递
融合CBAM注意力机制的敦煌壁画风格迁移
8
作者 贵向泉 曹磊 李立 《计算机系统应用》 2025年第4期276-285,共10页
敦煌壁画是人类世界文明史中耀眼的瑰宝.然而,现有对敦煌壁画的算法研究主要集中在壁画修复方面,很少有针对敦煌壁画的色彩风格迁移研究.因此,提出一种基于循环生成对抗网络的融合CBAM注意力机制的敦煌壁画风格迁移方法.通过提取输入图... 敦煌壁画是人类世界文明史中耀眼的瑰宝.然而,现有对敦煌壁画的算法研究主要集中在壁画修复方面,很少有针对敦煌壁画的色彩风格迁移研究.因此,提出一种基于循环生成对抗网络的融合CBAM注意力机制的敦煌壁画风格迁移方法.通过提取输入图像的特征,将其输入到添加CBAM注意力机制的生成器中,应用注意力机制提升重点区域的风格迁移效果,抑制边界伪影的产生;为了更好地保留图像内容的结构信息,在下采样区和上采样区之间添加了残差网络模块;并且在损失函数中加入色彩损失,约束模型提高生成图像的风格化效果.通过自建的敦煌壁画数据集上进行的实验验证,所提出的模型在敦煌壁画艺术风格迁移任务中展现出了相较于现有方法的优越性.该模型能够生成视觉效果更为卓越、艺术韵味更为浓厚的敦煌壁画风格化图像,为敦煌壁画的创新研究提供了新思路. 展开更多
关键词 风格迁移 循环生成对抗网络 cbam注意力机制 敦煌壁画
在线阅读 下载PDF
基于U^(2)-Net和CBAM融合注意力的双模态睡眠分期研究 被引量:1
9
作者 赵倩 李锦 +2 位作者 凤飞龙 强宁 胡静 《陕西师范大学学报(自然科学版)》 北大核心 2025年第1期1-11,共11页
针对当前自动睡眠分期方法存在的难点问题,提出了一种结合U^(2)-Net和CBAM融合注意力对EEG-ECG双模态信号进行自动睡眠分期的方法。首先,采用MIT-BIH公开数据集中的EEG-ECG信号进行预处理;然后,利用添加了多尺度特征提取模块的U^(2)-Ne... 针对当前自动睡眠分期方法存在的难点问题,提出了一种结合U^(2)-Net和CBAM融合注意力对EEG-ECG双模态信号进行自动睡眠分期的方法。首先,采用MIT-BIH公开数据集中的EEG-ECG信号进行预处理;然后,利用添加了多尺度特征提取模块的U^(2)-Net网络并行提取EEG和ECG中的波形特征;其次,利用CBAM融合注意力对全部特征进行权重分配;最后,使用Softmax激活函数对睡眠时期进行六分类。结果表明:基于U^(2)-Net和CBAM融合注意力模型进行睡眠分期时,使用ECG单模态信号的六分类总体准确率为80.2%,F1分数为75.3%;使用EEG单模态信号的六分类总体准确率为85.8%,F1分数为81.7%;使用EEG-ECG双模态信号的六分类总体准确率为90.4%,F1分数为85.6%。提出的双模态睡眠分期模型是可行有效的,并且为自动睡眠分期提供了一种新的思路。 展开更多
关键词 自动睡眠分期 EEG-ECG双模态信号 U^(2)-Net网络 cbam融合注意力
在线阅读 下载PDF
基于CBAM-TransUNet的地震断层识别方法 被引量:2
10
作者 王新 张薇 +2 位作者 陈同俊 张傲 赵砀 《煤炭学报》 北大核心 2025年第2期1192-1202,共11页
断层的检测和识别在煤炭勘探开采过程中至关重要,传统的人工解释断层方法已经无法满足实际生产的需求,基于深度学习的地震断层解释方法在断层分割领域表现较为出色。常规卷积神经网络(CNN)感受野有限,不能很好地利用全局信息,会导致一... 断层的检测和识别在煤炭勘探开采过程中至关重要,传统的人工解释断层方法已经无法满足实际生产的需求,基于深度学习的地震断层解释方法在断层分割领域表现较为出色。常规卷积神经网络(CNN)感受野有限,不能很好地利用全局信息,会导致一些预测的断层存在连续性不足和断层缺失等问题。Transformer具有提取全局信息的优势,引入CNN和Transformer融合的TransUNet网络,构建一种基于CBAM-TransUNet的地震断层识别方法对二维地震断层图像进行识别。首先,将CBAM-Block注意力模块融入TransUNet网络,将该模块分别加入CNN断层编码器部分和连接断层编码器与断层解码器的3层跳跃连接部分,同时从通道和空间2个维度增强地震断层图像的识别能力;其次,选择Dice损失函数和交叉熵损失函数联合优化的损失函数,使得断层图像分割更为准确,CBAM-TransUNet断层识别网络在合成地震数据集上获得的DICE值和IOU值分别提高到0.84和0.75,试验结果表明断层识别的连续性更强,明显优于其他经典分割方法;最后,利用构建的模型对荷兰近海北海F3区块真实地震数据集进行了断层解释。试验结果表明:基于CBAM-TransUNet的地震断层识别方法在去除冗余断层信息的同时能够有效识别出断层,在断层识别准确度和断层识别连续性方面表现优异,识别出的断层细节更加丰富,提高了断层识别的精度,可以有效应用于实际地震数据中识别断层。 展开更多
关键词 地震图像 断层识别 机器学习 注意力机制 TRANSFORMER
在线阅读 下载PDF
融合CBAM的违法犯罪类安卓恶意软件检测与分类模型研究 被引量:1
11
作者 刘红玉 高见 《计算机工程与应用》 北大核心 2025年第6期317-327,共11页
针对公安工作领域移动终端APP违法犯罪日益频发的情况,为解决Android恶意违法犯罪软件检测领域中相关数据集数量少、分类不清晰,识别Android恶违法软件可行性方法匮乏等情况,提出了一种基于安卓违法犯罪APP数据集,融合CBAM注意力机制的... 针对公安工作领域移动终端APP违法犯罪日益频发的情况,为解决Android恶意违法犯罪软件检测领域中相关数据集数量少、分类不清晰,识别Android恶违法软件可行性方法匮乏等情况,提出了一种基于安卓违法犯罪APP数据集,融合CBAM注意力机制的深度学习模型。收集6181个违法犯罪类APP并整理划分为4个家族;对违法APP软件进行灰度图、RGB以及RGBA三种图像可视化处理;利用融合CBAM注意力机制的深度模型进行家族检测分类。在违法犯罪APP数据集上的实验表明,融合CBAM机制的Resnet18模型在RGBA图像上与未引入该机制的灰度图图像相比,准确度提升了4.04%,达到93.52%。融合CBAM机制的模型在公开Drebin数据集上进行了验证,引入CBAM深度学习模型VGG16在RGBA图像上取得了96.35%的准确率。 展开更多
关键词 违法犯罪 安卓恶意软件 RGBA图像 可视化处理 卷积块注意力模块(cbam) 深度学习
在线阅读 下载PDF
基于轻量化CBAM—GoogLeNet的辣椒病虫害识别 被引量:2
12
作者 戴敏 孙文靖 缪宏 《中国农机化学报》 北大核心 2025年第2期224-229,252,共7页
针对GoogLeNet模型在自然环境下进行辣椒叶片病虫害识别时存在网络参数多、模型内存大以及训练时间长的问题,提出一种融合CBAM机制的轻量化GoogLeNet模型(CBAM—GoogLeNet)。采用CBAM注意力机制替换Inception(4b)和Inception(4c)模块,... 针对GoogLeNet模型在自然环境下进行辣椒叶片病虫害识别时存在网络参数多、模型内存大以及训练时间长的问题,提出一种融合CBAM机制的轻量化GoogLeNet模型(CBAM—GoogLeNet)。采用CBAM注意力机制替换Inception(4b)和Inception(4c)模块,将该注意力机制插入到平均池化层之后,在全连接层中添加L2正则化,达到减小训练模型和缩短训练时长的目的,同时保证网络模型的高准确率和验证率,并结合MATLAB平台设计一款可视化的辣椒病虫害识别系统。结果表明,CBAM—GoogLeNet的模型大小相比AlexNet、VGG16、VGG19和GoogLeNet分别缩小91.2%、96.2%、96.3%和15.0%,训练时长分别减少12.7%、26.5%、62.2%和8.8%,此外,该模型的识别准确率达到99.5%,验证准确率达到97.3%,实现模型轻量化和快速精准识别的目标。为辣椒及时防治、减少损失提供一种有效的技术支持。 展开更多
关键词 辣椒病虫害 精准识别 轻量化模型 注意力机制 深度学习
在线阅读 下载PDF
基于CBAM改进YOLOv7的电力设备红外图像分类检测 被引量:1
13
作者 陈佳 余成波 +3 位作者 王士彬 蒋启超 何鑫 张未 《红外技术》 北大核心 2025年第1期72-80,共9页
针对复杂环境下电力设备红外图像的深度学习目标检测数据繁杂、检测精度较低等问题,本文提出一种基于卷积块注意模块(Convolutional Block Attention Module,CBAM)改进YOLOv7的电力设备红外图像分类算法。首先将已有数据集进行标注,并... 针对复杂环境下电力设备红外图像的深度学习目标检测数据繁杂、检测精度较低等问题,本文提出一种基于卷积块注意模块(Convolutional Block Attention Module,CBAM)改进YOLOv7的电力设备红外图像分类算法。首先将已有数据集进行标注,并按一定比例划分成训练集、验证集以及测试集,然后在YOLOv7的主干网络中引入CBAM,使模型能对感兴趣的区域进行强调并抑制无用信息,其次将划分好的数据集放入改进后的YOLOv7进行模型训练,同时对比了6种改进的YOLOv5s模型。实验结果表明,在相同实验条件下改进YOLOv7模型优于YOLOv7模型、YOLOv5s模型和基于YOLOv5s的6种注意力模型。改进YOLOv7性能有明显提升,可实现快速、精准的红外图像分类。 展开更多
关键词 电力设备 YOLOv7 红外图像 cbam
在线阅读 下载PDF
基于CBAM-YOLOv4的东巴象形文识别方法研究 被引量:1
14
作者 黄颢 吴国新 +1 位作者 徐小力 赵西伟 《计算机应用与软件》 北大核心 2025年第2期196-201,226,共7页
东巴象形文字是中国早期使用的一种象形文字,对该文字的识别和数字化保护等方面的研究对传承我国文化具有深远意义。针对从东巴古籍提取的象形文字结构复杂、存在异体字、记录该文字的特殊东巴纸的纹理特征干扰识别的情况,提出一种基于C... 东巴象形文字是中国早期使用的一种象形文字,对该文字的识别和数字化保护等方面的研究对传承我国文化具有深远意义。针对从东巴古籍提取的象形文字结构复杂、存在异体字、记录该文字的特殊东巴纸的纹理特征干扰识别的情况,提出一种基于CBAM-YOLOv4的图像识别改进算法,该算法添加注意力机制模块CBAM(Convolutional Block Attention Module)和特征融合模块,通过CBAM中的通道和空间注意力子模块依次对图像推断出注意力图,并结合特征融模块对输入的东巴象形文字图片进行更深的特征提取,从而实现对YOLOv4图像检测识别算法的优化。将改进后的CBAM-YOLOv4算法应用于东巴象形文字识别,相比YOLOv4算法mAP值提高了4.42百分点,表明该算法具有较好的东巴文字识别性能。 展开更多
关键词 东巴文识别 YOLOv4 cbam 特征提取
在线阅读 下载PDF
基于CBAM残差块结合纹理采样器的水墨迁移算法
15
作者 刘雪莉 杜洪波 +1 位作者 袁雪丰 朱立军 《电子科技》 2025年第6期65-73,共9页
针对水墨风格迁移算法出现的线条不清晰、纹理杂乱及颜色重建效果差等问题,文中提出了一种新型轻量级水墨风格迁移神经网络IWSGAN(Ink and Wash Style Generative Adversarial Networks)。使用CBAM(Convolutional Block Attention Modu... 针对水墨风格迁移算法出现的线条不清晰、纹理杂乱及颜色重建效果差等问题,文中提出了一种新型轻量级水墨风格迁移神经网络IWSGAN(Ink and Wash Style Generative Adversarial Networks)。使用CBAM(Convolutional Block Attention Module)残差块对特征图进行通道空间注意力计算,提高了算法对有效信息的采取率。结合水墨纹理显著性采样器ITSS(Ink-Texture-Saliency-Sampler)从训练数据中采样水墨显著性局部图像块,使迁移的水墨图片线条纹理的水墨质感更明显。采用5种不同的损失函数约束内容图像和生成图像的高层语义,以促进风格特征的一致性。实验结果表明,IWSGAN生成图像保留了更多的内容特征,线条的纹理细节更形象,色彩重建表现较好,图像质量得到了显著提升。 展开更多
关键词 风格迁移 生成对抗网络 cbam残差块 注意力机制 ITSS 局部对抗 水墨迁移 消融实验
在线阅读 下载PDF
基于GoogLeNet-CBAM船载图像的滑坡识别方法 被引量:2
16
作者 郭江波 李浩然 +1 位作者 冼进业 祝敏刚 《水利水电快报》 2025年第6期28-32,共5页
为应对河岸消落带常见的滑坡灾害问题,提出了一种基于多尺度特征融合(GoogLeNet)和空间-通道注意力机制(CBAM)的船载图像库区滑坡识别方法。以GoogLeNet作为基准网络结构,通过Inception模块实现多尺度特征并行卷积,有效捕捉从宏观到微... 为应对河岸消落带常见的滑坡灾害问题,提出了一种基于多尺度特征融合(GoogLeNet)和空间-通道注意力机制(CBAM)的船载图像库区滑坡识别方法。以GoogLeNet作为基准网络结构,通过Inception模块实现多尺度特征并行卷积,有效捕捉从宏观到微观的多层次特征信息。进一步结合CBAM模块,动态调整特征图的通道和空间维度权重,显著提升网络的表达能力和滑坡区域的显著性。实验结果表明:该方法在三峡库区巫山段约30 km的滑坡图像识别中取得了95.93%的高准确率,与VGG19、ResNet50和DenseNet53等其他模型相比,准确率、精确度和F1分数方面均有显著提升;尽管召回率略低于VGG19,但整体性能卓越。研究成果可为地质灾害预防与管理提供技术支持。 展开更多
关键词 多尺度特征融合 GoogLeNet cbam 船载图像 滑坡识别 遥感技术
在线阅读 下载PDF
基于CBAM的弱监督目标检测 被引量:1
17
作者 刘均 潘妍 邓海航 《化工自动化及仪表》 2025年第2期191-197,共7页
由于实例级类别标注的缺失,弱监督目标检测网络在精确预测目标位置时面临显著挑战。当前主流策略倾向于采用分阶段学习,然而这一过程可能导致特定对象类别陷入局部最优。为克服此难题,提出一种新的端到端联合训练框架,即构建了一个集成... 由于实例级类别标注的缺失,弱监督目标检测网络在精确预测目标位置时面临显著挑战。当前主流策略倾向于采用分阶段学习,然而这一过程可能导致特定对象类别陷入局部最优。为克服此难题,提出一种新的端到端联合训练框架,即构建了一个集成多实例学习与边界框回归分支的统一网络架构,两者共享一个高效的主干网络以促进协同。同时,引入注意力机制于主干,深化特征中的位置信息挖掘。通过在基准数据集PASCAL VOC 2007、2012上的广泛实验验证,证实所提方法达到了较高的性能。 展开更多
关键词 弱监督目标检测 边界框回归 全监督检测 注意力机制
在线阅读 下载PDF
基于改进CBAM注意力机制的MobileNetV3风扇异常状况识别研究
18
作者 刘明 王荣燕 +3 位作者 王汝旭 武高旭 张佳宁 梁俊祥 《工业控制计算机》 2025年第3期90-92,共3页
工业风扇在生产设施中起着至关重要的作用,关键风扇的突然停机对安全生产影响巨大。通过分析在-6 dB噪声环境中的故障风扇发出的声音,提取声音样本的语谱图,采用MobileNetV3模型,针对该模型注意力模块SE(Squeeze-and-Excitation)存在的... 工业风扇在生产设施中起着至关重要的作用,关键风扇的突然停机对安全生产影响巨大。通过分析在-6 dB噪声环境中的故障风扇发出的声音,提取声音样本的语谱图,采用MobileNetV3模型,针对该模型注意力模块SE(Squeeze-and-Excitation)存在的参数化程度较低问题,采用空洞卷积(Dilated Convolution)优化的卷积块注意力模块CBAM(Convolutional Block Attention Module)予以替代,提出了改进后的MobileNetV3模型。实验结果显示,该模型的分类准确率达到了98%,相较于原MobileNetV3模型,准确率提升了2.07个百分点。 展开更多
关键词 空洞卷积 cbam MobileNetV3 迁移学习 SPECTROGRAM
在线阅读 下载PDF
欧盟CBAM对中国铝产品出口的影响及对策
19
作者 李薇 《北方经贸》 2025年第3期32-35,共4页
随着全球对气候变化问题的关注度日益增加,欧盟实施了碳边境调节机制(CBAM),预防“碳泄露”并减少温室气体排放。而欧盟作为中国铝产品出口的主要目的地,通过分析中国对欧盟铝产品的出口现状,包括出口概况、增长趋势和市场地位,以及从... 随着全球对气候变化问题的关注度日益增加,欧盟实施了碳边境调节机制(CBAM),预防“碳泄露”并减少温室气体排放。而欧盟作为中国铝产品出口的主要目的地,通过分析中国对欧盟铝产品的出口现状,包括出口概况、增长趋势和市场地位,以及从贸易关联度、碳排放水平以及碳价差异三个维度,探讨CBAM对中国铝产品出口的影响。从而提出中国的应对策略,即发展再生铝、完善碳交易市场、调整能源消耗结构,保持国际市场上的竞争力。 展开更多
关键词 欧盟 碳边境调节机制 cbam 中国铝产品出口
在线阅读 下载PDF
Pyroptosis,ferroptosis,and autophagy in spinal cord injury:regulatory mechanisms and therapeutic targets 被引量:4
20
作者 Qingcong Zheng Du Wang +1 位作者 Rongjie Lin Weihong Xu 《Neural Regeneration Research》 SCIE CAS 2025年第10期2787-2806,共20页
Regulated cell death is a form of cell death that is actively controlled by biomolecules.Several studies have shown that regulated cell death plays a key role after spinal cord injury.Pyroptosis and ferroptosis are ne... Regulated cell death is a form of cell death that is actively controlled by biomolecules.Several studies have shown that regulated cell death plays a key role after spinal cord injury.Pyroptosis and ferroptosis are newly discovered types of regulated cell deaths that have been shown to exacerbate inflammation and lead to cell death in damaged spinal cords.Autophagy,a complex form of cell death that is interconnected with various regulated cell death mechanisms,has garnered significant attention in the study of spinal cord injury.This injury triggers not only cell death but also cellular survival responses.Multiple signaling pathways play pivotal roles in influencing the processes of both deterioration and repair in spinal cord injury by regulating pyroptosis,ferroptosis,and autophagy.Therefore,this review aims to comprehensively examine the mechanisms underlying regulated cell deaths,the signaling pathways that modulate these mechanisms,and the potential therapeutic targets for spinal cord injury.Our analysis suggests that targeting the common regulatory signaling pathways of different regulated cell deaths could be a promising strategy to promote cell survival and enhance the repair of spinal cord injury.Moreover,a holistic approach that incorporates multiple regulated cell deaths and their regulatory pathways presents a promising multi-target therapeutic strategy for the management of spinal cord injury. 展开更多
关键词 AUTOPHAGY cell death ferroptosis INFLAMMATION pathological mechanisms PYROPTOSIS regulated cell death regulatory pathways spinal cord injury therapeutic targets
暂未订购
上一页 1 2 250 下一页 到第
使用帮助 返回顶部