期刊文献+
共找到155,669篇文章
< 1 2 250 >
每页显示 20 50 100
Potential mechanisms of non-coding RNA regulation in Alzheimer's disease 被引量:1
1
作者 Yue Sun Xinping Pang +5 位作者 Xudong Huang Dinglu Liu Jingyue Huang Pengtao Zheng Yanyu Wei Chaoyang Pang 《Neural Regeneration Research》 2026年第1期265-280,共16页
Alzheimer's disease,a progressively degenerative neurological disorder,is the most common cause of dementia in the elderly.While its precise etiology remains unclear,researchers have identified diverse pathologica... Alzheimer's disease,a progressively degenerative neurological disorder,is the most common cause of dementia in the elderly.While its precise etiology remains unclear,researchers have identified diverse pathological characteristics and molecular pathways associated with its progression.Advances in scientific research have increasingly highlighted the crucial role of non-coding RNAs in the progression of Alzheimer's disease.These non-coding RNAs regulate several biological processes critical to the advancement of the disease,offering promising potential as therapeutic targets and diagnostic biomarkers.Therefore,this review aims to investigate the underlying mechanisms of Alzheimer's disease onset,with a particular focus on microRNAs,long non-coding RNAs,and circular RNAs associated with the disease.The review elucidates the potential pathogenic processes of Alzheimer's disease and provides a detailed description of the synthesis mechanisms of the three aforementioned non-coding RNAs.It comprehensively summarizes the various non-coding RNAs that have been identified to play key regulatory roles in Alzheimer's disease,as well as how these noncoding RNAs influence the disease's progression by regulating gene expression and protein functions.For example,miR-9 targets the UBE4B gene,promoting autophagy-mediated degradation of Tau protein,thereby reducing Tau accumulation and delaying Alzheimer's disease progression.Conversely,the long non-coding RNA BACE1-AS stabilizes BACE1 mRNA,promoting the generation of amyloid-βand accelerating Alzheimer's disease development.Additionally,circular RNAs play significant roles in regulating neuroinflammatory responses.By integrating insights from these regulatory mechanisms,there is potential to discover new therapeutic targets and potential biomarkers for early detection and management of Alzheimer's disease.This review aims to enhance the understanding of the relationship between Alzheimer's disease and non-coding RNAs,potentially paving the way for early detection and novel treatment strategies. 展开更多
关键词 Alzheimer's disease biomarkers circular RNA long non-coding RNA MICRORNA ncRNA regulation NEURODEGENERATION non-coding RNA PATHOGENESIS therapeutic targets
暂未订购
基于WSS-Pointnet的变电站点云弱监督语义分割方法
2
作者 裴少通 孙海超 +2 位作者 胡晨龙 王玮琦 兰博 《电工技术学报》 北大核心 2026年第1期234-245,共12页
现有的变电站点云语义分割算法均采用完全监督学习,需要大量人工标注点云数据,导致分割任务耗时长且成本高昂。为解决这一问题,该文提出一种基于PointNet改进的弱监督语义分割PointNet(WSS-PointNet)算法。首先,通过构建多层降采样结构... 现有的变电站点云语义分割算法均采用完全监督学习,需要大量人工标注点云数据,导致分割任务耗时长且成本高昂。为解决这一问题,该文提出一种基于PointNet改进的弱监督语义分割PointNet(WSS-PointNet)算法。首先,通过构建多层降采样结构,结合采样层与分组层对输入点云数据进行多尺度特征提取,从而捕捉点云在不同尺度上的几何和拓扑信息。在此基础上,引入PointNet结构以进一步提取区域特征,优化局部特征整合与全局特征表示;针对粗粒度语义特征的优化,提出膨胀式语义信息嵌入与浸染式语义信息嵌入两种模块,分别采用“由内而外”和“由外而内”的信息传递策略对点云语义信息进行细致处理,两种嵌入机制均基于图卷积神经网络,通过捕捉局部连接模式与信息共享实现语义特征的高效传播。其次,构建变电站点云数据集,并对WSS-PointNet算法进行消融实验,同时与主流的完全监督学习算法和弱监督学习算法进行对比。经实验验证,WSS-PointNet相比于改进前将变电站点云分割的总体精度(OA)提高了10.3个百分点,平均交并比(mIoU)提高了10.1个百分点,平均准确率(mAcc)提高了10.5个百分点,同时在标注所需时间方面缩短了90%,接近完全监督算法中最好的分割效果。该模型可显著降低处理变电站点云数据的时间与成本,同时保持点云分割的高精度。 展开更多
关键词 点云语义分割 弱监督方法 膨胀式语义信息嵌入 浸染式语义信息嵌入 变电站
在线阅读 下载PDF
Helicobacter pylori-related non-coding RNAs in gastric cancer screening:Emerging evidence and translational challenges
3
作者 Zuo-Po Lv Muhammad Haris Sultan Yi-Gang Wang 《World Journal of Gastrointestinal Oncology》 2026年第1期1-7,共7页
Gastric cancer(GC)has high morbidity and mortality worldwide.Due to the absence of noticeable symptoms,diagnosing GC at an early stage is very difficult,which consequently leads to advanced GC and poor prognosis.Effec... Gastric cancer(GC)has high morbidity and mortality worldwide.Due to the absence of noticeable symptoms,diagnosing GC at an early stage is very difficult,which consequently leads to advanced GC and poor prognosis.Effective biomarkers are essential for prolonging patients’survival.Helicobacter pylori(H.pylori)infection represents the most significant risk factor for GC,with nearly all cases linked to this infection.Many non-coding RNAs(ncRNAs)are dysregulated in H.pylori-infected GC,indicating that ncRNAs may serve as biomarkers of early-stage GC.In this editorial,we discuss the study by Chen et al.Although previous studies have identified roles for miR-136 in gastric cancer proliferation,apoptosis,and invasion,none have specifically explored its relationship with H.pylori-associated gastric carcinogenesis. 展开更多
关键词 Helicobacter pylori Gastric cancer non-coding RNA BIOMARKER Clinical challenges
暂未订购
Novel insights into non-coding RNAs and their role in hydrocephalus
4
作者 Zhiyue Cui Jian He +8 位作者 An Li Junqiang Wang Yijian Yang Kaiyue Wang Zhikun Liu Qian Ouyang Zhangjie Su Pingsheng Hu Gelei Xiao 《Neural Regeneration Research》 2026年第2期636-647,共12页
A large body of evidence has highlighted the role of non-coding RNAs in neurodevelopment and neuroinflammation.This evidence has led to increasing speculation that non-coding RNAs may be involved in the pathophysiolog... A large body of evidence has highlighted the role of non-coding RNAs in neurodevelopment and neuroinflammation.This evidence has led to increasing speculation that non-coding RNAs may be involved in the pathophysiological mechanisms underlying hydrocephalus,one of the most common neurological conditions worldwide.In this review,we first outline the basic concepts and incidence of hydrocephalus along with the limitations of existing treatments for this condition.Then,we outline the definition,classification,and biological role of non-coding RNAs.Subsequently,we analyze the roles of non-coding RNAs in the formation of hydrocephalus in detail.Specifically,we have focused on the potential significance of non-coding RNAs in the pathophysiology of hydrocephalus,including glymphatic pathways,neuroinflammatory processes,and neurological dysplasia,on the basis of the existing evidence.Lastly,we review the potential of non-coding RNAs as biomarkers of hydrocephalus and for the creation of innovative treatments. 展开更多
关键词 HYDROCEPHALUS NEURODEVELOPMENT NEUROINFLAMMATION non-coding RNA therapeutic target
暂未订购
RP11-Derived Long Non-Coding RNAs in Hepatocellular Carcinoma:Hidden Treasures in Plain Sight
5
作者 Se Ha Jang Hyung Seok Kim Jung Woo Eun 《Oncology Research》 2026年第1期89-104,共16页
Hepatocellular carcinoma(HCC)remains one of the most prevalent and lethal malignancies worldwide.Long non-coding RNAs(lncRNAs)have emerged as crucial regulators of gene expression and cancer progression,yet the functi... Hepatocellular carcinoma(HCC)remains one of the most prevalent and lethal malignancies worldwide.Long non-coding RNAs(lncRNAs)have emerged as crucial regulators of gene expression and cancer progression,yet the functional diversity of RP11-derived lncRNAs—originally mapped to bacterial artificial chromosome(BAC)clones from the Roswell Park Cancer Institute—has only recently begun to be appreciated.This mini-review aims to systematically synthesize current findings on RP11-derived lncRNAs in HCC,outlining their genomic origins,molecular mechanisms,and biological significance.We highlight their roles in metabolic reprogramming,microRNA network modulation,and tumor progression,as well as their diagnostic and prognostic value in tissue and serum-based analyses.Finally,we discuss therapeutic opportunities and propose future directions to translate RP11-derived lncRNAs into clinically actionable biomarkers and targets for precision liver cancer therapy. 展开更多
关键词 Hepatocellular carcinoma long non-coding RNA RP11-derived lncRNA BIOMARKER therapeutic target
暂未订购
基于Pointnet++的花生植株三维模型器官分割研究
6
作者 孟兆凡 程曼 +1 位作者 袁洪波 赵欢 《中国农机化学报》 北大核心 2026年第1期118-127,共10页
基于点云进行三维重构并进行器官分割对植物学研究至关重要,为研究花生植株茎叶器官分割训练样本的数量和类型对分割结果的影响规律,基于Pointnet++构建花生植株三维模型茎叶分割网络模型,并对比分析训练集类型以及数量对分割效果的影... 基于点云进行三维重构并进行器官分割对植物学研究至关重要,为研究花生植株茎叶器官分割训练样本的数量和类型对分割结果的影响规律,基于Pointnet++构建花生植株三维模型茎叶分割网络模型,并对比分析训练集类型以及数量对分割效果的影响。当训练集为10株花生幼苗期数据时,模型分割效果最好,准确率、类平均准确率、类平均交并比、F1分数分别为94.5%、81.9%、76.9%、85.7%。其中,在花生荚果期训练集中加入20株开花期数据,类平均准确率、类平均交并比分别上升19.55%、20.75%。试验结果表明,Pointnet++可以有效分割花生植株茎叶器官,训练集的多样性和数据量的增加有利于模型学习花生植株不同生长阶段的形态特征,在训练集中加入相近生长阶段和生长特征的模型数据,并增加数据量对模型分割效果提高更明显。 展开更多
关键词 花生植株 三维建模 点云 器官分割 训练集
在线阅读 下载PDF
Ultrastructure and key identification points of fossilized Os Draconis in traditional Chinese medicine
7
作者 Dong-Han Bai Zi Xing +5 位作者 Zi-Hao Zhang Zhi-Jie Zhang Da-Jun Lu Nan-Xi Huang Qiao-Chu Wang Lu Luo 《Traditional Medicine Research》 2026年第1期39-46,共8页
Background:The medicinal material known as Os Draconis(Longgu)originates from fossilized remains of ancient mammals and is widely used in treating emotional and mental conditions.However,fossil resources are nonrenewa... Background:The medicinal material known as Os Draconis(Longgu)originates from fossilized remains of ancient mammals and is widely used in treating emotional and mental conditions.However,fossil resources are nonrenewable,and clinical demand is increasingly difficult to meet,leading to a proliferation of counterfeit products.During prolonged geological burial,static pressure from the surrounding strata severely compromises the microstructural integrity of osteons in Os Draconis,but Os Draconis still largely retains the structural features of mammalian bone.Methods:Using verified authentic Os Draconis samples over 10,000 years old as a baseline,this study summarizes the ultrastructural characteristics of genuine Os Draconis.Employing electron probe microanalysis and optical polarized light microscopy,we examined 28 batches of authentic Os Draconis and 31 batches of counterfeits to identify their ultrastructural differences.Key points for ultrastructural identification of Os Draconis were compiled,and a new identification approach was proposed based on these differences.Results:Authentic Os Draconis exhibited distinct ultrastructural markers:irregularly shaped osteons with traversing fissures,deformed/displaced Haversian canals,and secondary mineral infill(predominantly calcium carbonate).Counterfeits showed regular osteon arrangements,absent traversal fissures,and homogeneous hydroxyapatite composition.Lab-simulated samples lacked structural degradation features.EPMA confirmed calcium carbonate infill in fossilized Haversian canals,while elemental profiles differentiated lacunae types(void vs.mineral-packed).Conclusion:The study established ultrastructural criteria for authentic Os Draconis identification:osteon deformation,geological fissures penetrating bone units,and heterogenous mineral deposition.These features,unattainable in counterfeits or modern processed bones,provide a cost-effective,accurate identification method.This approach bridges gaps in TCM material standardization and supports quality control for clinical applications. 展开更多
关键词 Os Draconis ULTRASTRUCTURE identification points electron probe polarized light microscope
暂未订购
基于PointNet与曲率约束的导丝配准方法
8
作者 邓子寒 胡陟 +2 位作者 辛绍宗 李树凡 张贝朗 《中国医学物理学杂志》 2026年第2期204-210,共7页
针对血管介入手术过程中,传统配准算法处理柔性导丝的动态变化时,高度依赖导丝的初始位姿,配准精度不足、容易陷入局部最优解和效率低下的问题,提出一种基于PointNet与曲率约束的导丝配准方法。首先,建立导丝和血管的物理模型,获取运动... 针对血管介入手术过程中,传统配准算法处理柔性导丝的动态变化时,高度依赖导丝的初始位姿,配准精度不足、容易陷入局部最优解和效率低下的问题,提出一种基于PointNet与曲率约束的导丝配准方法。首先,建立导丝和血管的物理模型,获取运动点位置。然后,利用PointNet提取关键点特征,通过姿态回归网络预测导丝点云和血管中心线点云的变换矩阵,使用KD-Tree加速搜索点云目标并基于曲率特征约束的迭代最近点算法对配准结果进行优化。实验结果表明,相较于传统方法和基于学习的方法,本文方法的均方误差、平均绝对误差最小,证明该方法在血管介入导丝配准中的有效性。 展开更多
关键词 血管介入 pointNet 回归预测 最近迭代点 曲率
在线阅读 下载PDF
基于改进PointNet++的中压电力线点云分类方法 被引量:1
9
作者 雒建艳 《应用激光》 北大核心 2025年第3期146-158,共13页
针对中压电力线点云分类中存在的噪声干扰、分类精度低和鲁棒性不足的问题,提出一种基于改进PointNet++的中压电力线点云分类方法。首先,通过多种手段提取点云空间信息、几何特征以及局部几何特征等多维度特征,为点云单点构造40维特征向... 针对中压电力线点云分类中存在的噪声干扰、分类精度低和鲁棒性不足的问题,提出一种基于改进PointNet++的中压电力线点云分类方法。首先,通过多种手段提取点云空间信息、几何特征以及局部几何特征等多维度特征,为点云单点构造40维特征向量;然后对PointNet++进行改进,引入了点注意力模块(point attention module,PAM)和组注意力模块(group attention module,GAM),同时与层归一化(layer norm)和残差连接结构组合使用,用以增强其特征的细节捕捉能力,降低复杂环境对分类效果影响;最后采用某地机载采集的10 kV中压电力线走廊数据构建数据集,进行了方法验证。实验结果表明,所提方法在Precision、Recall和F_1-score上均优于传统机器学习方法和基于PointNet、PointNet++的深度学习方法。相较于PointNet++(XYZ+Features),所提方法在Precision、Recall和F_1-score上分别高出1.6个百分点、5.3个百分点和4.6个百分点,且通过可视化结果进一步验证了PAM和GAM的有效性。验证了所提方法在中压电力线点云的提取上更为精确,其结构特征更加清晰,且与周围环境的区分度更高。 展开更多
关键词 激光点云 注意力机制 pointNet++ 中压电力线 点云分类
原文传递
基于SwinPoinTr的视角受限下杏鲍菇表型参数测量方法 被引量:2
10
作者 谢立敏 黄轶 +2 位作者 吴昊宇 叶大鹏 方兵 《农业机械学报》 北大核心 2025年第3期148-157,共10页
针对菇房内杏鲍菇表型参数测量任务中,由于扫描设备视角受限,扫描的杏鲍菇点云出现残缺问题,基于AdaPoinTr(Adaptive geometry-aware point transformers)提出了改进的SwinPoinTr模型,实现了对残缺杏鲍菇点云的准确补全和杏鲍菇表型参... 针对菇房内杏鲍菇表型参数测量任务中,由于扫描设备视角受限,扫描的杏鲍菇点云出现残缺问题,基于AdaPoinTr(Adaptive geometry-aware point transformers)提出了改进的SwinPoinTr模型,实现了对残缺杏鲍菇点云的准确补全和杏鲍菇表型参数的测量。该方法在使用提出的特征重塑模块的基础上,构建具有几何感知能力的层次化Transformer编码模块,提高了模型对输入点云的利用率和模型捕捉点云细节特征的能力。然后基于泊松重建方法完成了补全点云表面重建,并测量到杏鲍菇表型参数。实验结果表明,本文所提算法在残缺杏鲍菇点云补全任务中,模型倒角距离为1.316×10^(-4),地球移动距离为21.3282,F1分数为87.87%。在表型参数估测任务中,模型对杏鲍菇菌高、体积、表面积估测结果的决定系数分别为0.9582、0.9596、0.9605,均方根误差分别为4.4213 mm、10.8185 cm^(3)、7.5778 cm^(2)。结果证实了该研究方法可以有效地补全残缺的杏鲍菇点云,可以为菇房内杏鲍菇表型参数测量提供基础。 展开更多
关键词 杏鲍菇 智慧菇房 表型参数 点云补全 泊松重建 Swinpointr
在线阅读 下载PDF
基于Point Transformer方法的鱼类三维点云模型分类 被引量:1
11
作者 胡少秋 段瑞 +3 位作者 张东旭 鲍江辉 吕华飞 段明 《水生生物学报》 北大核心 2025年第2期146-155,共10页
为实现对不同鱼类的精准分类,研究共采集110尾真实鱼类的三维模型,对获取的3D模型进行基于预处理、旋转增强和下采样等操作后,获取了1650尾实验样本。然后基于Point Transformer网络和2个三维分类的对比网络进行数据集的分类训练和验证... 为实现对不同鱼类的精准分类,研究共采集110尾真实鱼类的三维模型,对获取的3D模型进行基于预处理、旋转增强和下采样等操作后,获取了1650尾实验样本。然后基于Point Transformer网络和2个三维分类的对比网络进行数据集的分类训练和验证。结果表明,利用本实验的目标方法Point Transformer获得了比2个对比网络更好的分类结果,整体的分类准确率能够达到91.9%。同时对所使用的三维分类网络进行有效性评估,3个模型对于5种真实鱼类模型的分类是有意义的,其中Point Transformer的模型ROC曲线准确率最高,AUC面积最大,对于三维鱼类数据集的分类最为有效。研究提供了一种可以实现对鱼类三维模型进行精准分类的方法,为以后的智能化渔业资源监测提供一种新的技术手段。 展开更多
关键词 点云处理 point Transformer 三维模型 鱼类分类
在线阅读 下载PDF
基于DI-PointNet的变电站主设备点云高精度语义分割方法 被引量:2
12
作者 裴少通 孙海超 +2 位作者 孙志周 胡晨龙 祝雨馨 《电工技术学报》 北大核心 2025年第9期2917-2930,共14页
在变电站机器人巡检任务中,三维点云数据的高精度语义分割是关键技术之一,有助于机器人理解电力设备、障碍物和其他物体的空间布局。然而,现有的点云分割算法在变电站场景中的应用效果有限,准确度较低、计算复杂度高,难以实现对变电站... 在变电站机器人巡检任务中,三维点云数据的高精度语义分割是关键技术之一,有助于机器人理解电力设备、障碍物和其他物体的空间布局。然而,现有的点云分割算法在变电站场景中的应用效果有限,准确度较低、计算复杂度高,难以实现对变电站主设备点云的准确分割。为了解决这一问题,该文提出了一种基于PointNet++的DI-PointNet算法。首先,采用双层连续变换器模块增强点云之间的信息交互,有效地聚合长距离上下文,增大网络有效感受野;其次,通过分层键采样策略生成自注意力机制所需的键值,降低算法复杂度;最后,使用倒置残差模块,通过倒置瓶颈设计和残差连接缓解梯度消失,有效地增加模型的深度,同时降低计算复杂度。此外,该文构建了变电站点云数据集,对DI-PointNet算法进行详细的消融实验,并与主流深度学习算法和电力领域典型点云分割算法进行对比。实验验证结果表明,DI-PointNet算法对变电站主设备点云分割的平均交并比达到82.5%,相比PointNet++算法提高了2.1个百分点,且总体精度提高了3.4个百分点,达到90.1%。DI-PointNet算法为智能电力设备巡检和维护提供了有效的解决方案。 展开更多
关键词 点云语义分割 双层连续变换器 分层键采样 倒置残差 变电站
在线阅读 下载PDF
基于改进PointNet++的城市道路点云分类方法
13
作者 田晟 熊辰崟 龙安洋 《广西师范大学学报(自然科学版)》 北大核心 2025年第4期1-14,共14页
城市道路场景的点云数据量巨大、类别分布不平衡且密度极不均匀,导致现有的点云分类方法难以满足高精度分类的需求。为了解决现有PointNet++网络对局部特征提取不充分的问题,本文充分考虑场景的上下文信息和点之间的全局依赖性,构建融... 城市道路场景的点云数据量巨大、类别分布不平衡且密度极不均匀,导致现有的点云分类方法难以满足高精度分类的需求。为了解决现有PointNet++网络对局部特征提取不充分的问题,本文充分考虑场景的上下文信息和点之间的全局依赖性,构建融合上下文信息的PointNet++点云分类网络模型。首先,基于注意力机制设计局部特征聚合模块,通过动态地融合邻域点特征以充分捕获局部信息。其次,考虑现有的分类模型不能顾及上下文信息,导致复杂场景下的分类性能受限,本文构建上下文感知模块和双注意力模块,从多个维度提取上下文信息,进一步增强特征的表达能力。实验结果表明:改进模型在大型点云数据集下具有更高的分类精度及更强的泛化性能(总体分类精度在Oakland和Paris公开数据集上分别为98.70%和96.84%),更适用于大规模点云分类。 展开更多
关键词 点云分类 pointNet++ 局部特征 注意力机制 上下文信息 城市道路
在线阅读 下载PDF
Non-coding RNAs in acute ischemic stroke:from brain to periphery 被引量:1
14
作者 Shuo Li Zhaohan Xu +7 位作者 Shiyao Zhang Huiling Sun Xiaodan Qin Lin Zhu Teng Jiang Junshan Zhou Fuling Yan Qiwen Deng 《Neural Regeneration Research》 SCIE CAS 2025年第1期116-129,共14页
Acute ischemic stroke is a clinical emergency and a condition with high morbidity,mortality,and disability.Accurate predictive,diagnostic,and prognostic biomarkers and effective therapeutic targets for acute ischemic ... Acute ischemic stroke is a clinical emergency and a condition with high morbidity,mortality,and disability.Accurate predictive,diagnostic,and prognostic biomarkers and effective therapeutic targets for acute ischemic stroke remain undetermined.With innovations in high-throughput gene sequencing analysis,many aberrantly expressed non-coding RNAs(ncRNAs)in the brain and peripheral blood after acute ischemic stroke have been found in clinical samples and experimental models.Differentially expressed ncRNAs in the post-stroke brain were demonstrated to play vital roles in pathological processes,leading to neuroprotection or deterioration,thus ncRNAs can serve as therapeutic targets in acute ischemic stroke.Moreover,distinctly expressed ncRNAs in the peripheral blood can be used as biomarkers for acute ischemic stroke prediction,diagnosis,and prognosis.In particular,ncRNAs in peripheral immune cells were recently shown to be involved in the peripheral and brain immune response after acute ischemic stroke.In this review,we consolidate the latest progress of research into the roles of ncRNAs(microRNAs,long ncRNAs,and circular RNAs)in the pathological processes of acute ischemic stroke–induced brain damage,as well as the potential of these ncRNAs to act as biomarkers for acute ischemic stroke prediction,diagnosis,and prognosis.Findings from this review will provide novel ideas for the clinical application of ncRNAs in acute ischemic stroke. 展开更多
关键词 acute ischemic stroke apoptosis blood–brain barrier damage circular RNAs excitatory toxicity long non-coding RNAs MICRORNAS NEUROINFLAMMATION non-coding RNAs oxidative stress
暂未订购
Point-GBLS:结合深宽度学习的三维点云分类网络
15
作者 张国有 左嘉欣 +3 位作者 潘理虎 郝志祥 郭伟 张雪楠 《计算机系统应用》 2025年第3期1-13,共13页
基于点云的三维物体识别和检测是计算机视觉和自主导航领域的一个重要研究课题.如今,深度学习算法大大提高了三维点云分类的准确性和鲁棒性.然而,深度学习网络通常存在网络结构复杂、训练过程耗时等问题.本文提出了一种三维点云分类网络... 基于点云的三维物体识别和检测是计算机视觉和自主导航领域的一个重要研究课题.如今,深度学习算法大大提高了三维点云分类的准确性和鲁棒性.然而,深度学习网络通常存在网络结构复杂、训练过程耗时等问题.本文提出了一种三维点云分类网络Point-GBLS,它将深度学习和宽度学习系统结合在一起.网络结构简单,训练时间短.首先通过基于深度学习的特征提取网络提取点云特征,然后用改进的宽度学习系统对其进行分类.ModelNet40和ScanObjectNN数据集上的实验表明,Point-GBLS识别准确率分别达到92%以上和78%以上,训练时间低于同类深度学习方法的50%以上,优于具有相同骨干的深度学习网络. 展开更多
关键词 三维模型分类 点云 深度学习 宽度学习系统
在线阅读 下载PDF
基于Point-Attention点云分类的激光雷达故障诊断方法研究
16
作者 谭光兴 程星 陈海峰 《现代电子技术》 北大核心 2025年第20期10-17,共8页
在智能车辆和自主机器人领域,激光雷达传感器因高精度和可靠性,被广泛应用于环境感知和物体检测,因此其故障诊断尤为重要。激光雷达内部的故障往往有固件提醒,而外部环境因素导致的故障检测挑战较大,比如车辆形变、污垢等导致的激光点... 在智能车辆和自主机器人领域,激光雷达传感器因高精度和可靠性,被广泛应用于环境感知和物体检测,因此其故障诊断尤为重要。激光雷达内部的故障往往有固件提醒,而外部环境因素导致的故障检测挑战较大,比如车辆形变、污垢等导致的激光点云遮挡故障,难以直接在固件层面体现,需通过外部检测进行诊断。为此,提出一种基于Point-Attention激光雷达遮挡故障诊断方法。首先,结合多头几何注意力机制模块与CBAM模块、残差连接机制,增强了模型对点云数据中关键特征的提取能力,提高了分类准确性和鲁棒性;在真实的ScanObjectNN数据集和ModelNet40基准数据集上对Point-Attention模型进行了实验。该模型在分类任务上准确率分别达到了93.7%、82.5%。其次,融合了一种时间特征捕捉机制,从而使模型能够更好地适应现实场景中的时间相关性,进而更准确地处理激光雷达的遮挡故障。实验结果表明,所提方法能有效诊断激光雷达遮挡故障,最佳总体精度达99%以上,为激光雷达故障诊断提供了一种高效准确的解决方案。 展开更多
关键词 激光雷达 故障诊断 点云分类 残差连接 遮挡检测 时间特征捕捉
在线阅读 下载PDF
Nomenclature and location of acupuncture points for laboratory animals Part 3:Mouse 被引量:3
17
作者 《World Journal of Acupuncture-Moxibustion》 2025年第2期160-162,共3页
This paper introduces part of the content in the association standard,T/CAAM0002–2020 Nomenclature and Location of Acupuncture Points for Laboratory Animals Part 3:Mouse.This standard was released by the China Associ... This paper introduces part of the content in the association standard,T/CAAM0002–2020 Nomenclature and Location of Acupuncture Points for Laboratory Animals Part 3:Mouse.This standard was released by the China Association of Acupuncture and Moxibustion on May 15,2020,implemented on October 31,2020,and published by Standards Press of China.The standard was drafted by the Institute of Acupuncture and Moxibustion,China Academy of Chinese Medical Sciences,and the Nanjing University of Chinese Medicine.Principal draftsmen:Xiang-hong JING and Xing-bang HUA.Participating draftsmen:Wan-zhu BAI,Bin XU,Dong-sheng XU,Yi GUO,Tie-ming MA,Xin-jun WANG,and Sheng-feng LU. 展开更多
关键词 acupuncture points STANDARD MOUSE NOMENCLATURE location acupuncture points association standardt caam LOCATION acupuncture moxibustion
原文传递
Mechanisms underlying hepatocellular carcinoma progression through N6-methyladenosine modifications of long non-coding RNA 被引量:1
18
作者 Ning Wang Fei-Tian Min +1 位作者 Wei-Bo Wen Huan-Tian Cui 《World Journal of Gastroenterology》 2025年第21期135-139,共5页
Hepatocellular carcinoma(HCC)is a highly lethal malignancy with limited treatment options,particularly for patients with advanced stages of the disease.Sorafenib,the standard first-line therapy,faces significant chall... Hepatocellular carcinoma(HCC)is a highly lethal malignancy with limited treatment options,particularly for patients with advanced stages of the disease.Sorafenib,the standard first-line therapy,faces significant challenges due to the development of drug resistance.Yu et al explored the mechanisms by which lncRNA KIF9-AS1 regulates the stemness and sorafenib resistance in HCC using a combination of cell culture,transfection,RNA immunoprecipitation,co-immunoprecipitation,and xenograft tumor models.They demonstrate that N6-methyladenosine-modified long non-coding RNA KIF9-AS1 acts as an oncogene in HCC.This modification involves methyltransferase-like 3 and insulin-like growth factor 2 mRNA-binding protein 1,which play critical roles in regulating KIF9-AS1.Furthermore,KIF9-AS1 stabilizes and upregulates short stature homeobox 2 by promoting its deubiquitination through ubiquitin-specific peptidase 1,thereby enhancing stemness and contributing to sorafenib resistance in HCC cells.These findings provide a theoretical basis for KIF9-AS1 as a diagnostic marker and therapeutic target for HCC,highlighting the need for further investigation into its clinical application potential. 展开更多
关键词 Hepatocellular carcinoma STEMNESS Sorafenib resistance Long non-coding RNA KIF9-AS1 Short stature homeobox 2 N6-methyladenosine
暂未订购
Multi-sensor missile-borne LiDAR point cloud data augmentation based on Monte Carlo distortion simulation 被引量:1
19
作者 Luda Zhao Yihua Hu +4 位作者 Fei Han Zhenglei Dou Shanshan Li Yan Zhang Qilong Wu 《CAAI Transactions on Intelligence Technology》 2025年第1期300-316,共17页
Large-scale point cloud datasets form the basis for training various deep learning networks and achieving high-quality network processing tasks.Due to the diversity and robustness constraints of the data,data augmenta... Large-scale point cloud datasets form the basis for training various deep learning networks and achieving high-quality network processing tasks.Due to the diversity and robustness constraints of the data,data augmentation(DA)methods are utilised to expand dataset diversity and scale.However,due to the complex and distinct characteristics of LiDAR point cloud data from different platforms(such as missile-borne and vehicular LiDAR data),directly applying traditional 2D visual domain DA methods to 3D data can lead to networks trained using this approach not robustly achieving the corresponding tasks.To address this issue,the present study explores DA for missile-borne LiDAR point cloud using a Monte Carlo(MC)simulation method that closely resembles practical application.Firstly,the model of multi-sensor imaging system is established,taking into account the joint errors arising from the platform itself and the relative motion during the imaging process.A distortion simulation method based on MC simulation for augmenting missile-borne LiDAR point cloud data is proposed,underpinned by an analysis of combined errors between different modal sensors,achieving high-quality augmentation of point cloud data.The effectiveness of the proposed method in addressing imaging system errors and distortion simulation is validated using the imaging scene dataset constructed in this paper.Comparative experiments between the proposed point cloud DA algorithm and the current state-of-the-art algorithms in point cloud detection and single object tracking tasks demonstrate that the proposed method can improve the network performance obtained from unaugmented datasets by over 17.3%and 17.9%,surpassing SOTA performance of current point cloud DA algorithms. 展开更多
关键词 data augmentation LIDAR missile-borne imaging Monte Carlo simulation point cloud
在线阅读 下载PDF
Optimizing electronic structure through point defect engineering for enhanced electrocatalytic energy conversion
20
作者 Wei Ma Jiahao Yao +6 位作者 Fang Xie Xinqi Wang Hao Wan Xiangjian Shen Lili Zhang Menggai Jiao Zhen Zhou 《Green Energy & Environment》 SCIE EI CAS 2025年第1期109-131,共23页
Point defect engineering endows catalysts with novel physical and chemical properties,elevating their electrocatalytic efficiency.The introduction of defects emerges as a promising strategy,effectively modifying the e... Point defect engineering endows catalysts with novel physical and chemical properties,elevating their electrocatalytic efficiency.The introduction of defects emerges as a promising strategy,effectively modifying the electronic structure of active sites.This optimization influences the adsorption energy of intermediates,thereby mitigating reaction energy barriers,altering paths,enhancing selectivity,and ultimately improving the catalytic efficiency of electrocatalysts.To elucidate the impact of defects on the electrocatalytic process,we comprehensively outline the roles of various point defects,their synthetic methodologies,and characterization techniques.Importantly,we consolidate insights into the relationship between point defects and catalytic activity for hydrogen/oxygen evolution and CO_(2)/O_(2)/N_(2) reduction reactions by integrating mechanisms from diverse reactions.This underscores the pivotal role of point defects in enhancing catalytic performance.At last,the principal challenges and prospects associated with point defects in current electrocatalysts are proposed,emphasizing their role in advancing the efficiency of electrochemical energy storage and conversion materials. 展开更多
关键词 point defect engineering DOPING VACANCY ELECTROCATALYSIS Electronic structure
在线阅读 下载PDF
上一页 1 2 250 下一页 到第
使用帮助 返回顶部