Results 61 to 70 of about 861 (140)
Dianci bileiqi, 2023 配电线路中绝缘子与避雷器的故障影响供电系统的可靠性，为了实现航拍图像中绝缘子与避雷器目标的快速、精准检测，笔者提出一种基于改进YOLOv4的配电线路绝缘子与避雷器快速检测方法。针对无人机航拍的配电线路图像中各目标数目不均衡的问题，采用基于GraphCut与图像融合的样本扩充方法，建立含真实和虚拟的配电线路图像样本数据库。对于目标检测网络参数多、检测速度慢导致应用性能差的问题，通过迁移轻量级ShuffleNet网络对YOLOv4主干网络进行替换，利用深度可分离卷积原理对PANet模块进行轻量化调整 ...
王佰川, 王聪
doaj
Dynamic Adaptive Cross-Domain Mean Approximation [PDF]
Cross-domain mean approximation (CDMA) is an efficient measure of distributional differences between domains. It measures the sample distribution difference between domains by calculating the distance from a sample in one domain to the sample mean in ...
LI Huimin, MA Jianwei, ZANG Shaofei, SONG Yanbing
core +1 more source
Kongjian kexue xuebao不同等级耀斑发生的频次存在数量级上的差别, 使基于常规卷积神经网络的耀斑预报模型通常难以捕捉M和X类耀斑先兆特征, 导致高等级耀斑预报精度低的问题. 本文对于这种耀斑预报中的长尾分布问题, 通过控制变量法讨论不同深度长尾学习方法对于耀斑预报精度提升. 尝试从训练集优化、损失函数优化、网络权重优化等角度改进模型对于M和X类耀斑的预报性能.
周 俊 +3 more
doaj +1 more source
Research on Drift Elimination Algorithm for Inertial Navigation System Based on EKF [PDF]
, 2016 我国的捷联式惯性导航系统领域虽然仍处于初步发展阶段，不能满足商业化应用对产品的长时间高精度的定位要求，但该方案能够解决全球定位系统在室内和极端环境下的定位问题，因此越来越多的研究机构参与其中。捷联式惯性导航系统的误差源，主要是指器件本身的误差、系统模型误差以及工程计算环节引入的误差，此类误差虽然在短时间内不会对系统精度造成太大影响，但随着误差累计会使系统出现方向漂移。本文拟提出了一种基于扩展卡尔曼滤波器的复合型航向误差消除算法以解决上述缺陷。 本文使用基于扩展卡尔曼滤波器的惯性导航系统模型。其次 ...
唐玮玮
core
Key method of digitization of power distribution panel with artificial intelligence identification for power communication network [PDF]
To enhance the operational management efficiency of the power communication network, artificial intelligence technology was utilized to develop a digitalization method for distribution panels within the communication power system.
CHEN Yitong +4 more
core +1 more source
Image data-driven intelligent recognition of permafrost strength and feature visualization based analysis [PDF]
Ensuring the stability of the frozen wall is critical in freezing construction, but traditional onsite detection methods, due to their intermittent nature, fail to provide real-time monitoring, limiting timely responses to potential catastrophic events ...
Hang WEI +3 more
core +1 more source
[Automatic sleep staging algorithm for stochastic depth residual networks based on transfer learning]. [PDF]
Sheng Wu Yi Xue Gong Cheng Xue Za Zhi, 2023 Tian Y, Zhou Q, Li W.
europepmc +1 more source
Lightweight Recognition Method Based on Edge Computing [PDF]
The emergence of the "new retail" model has significant implications in transforming the traditional retail industry and providing enhanced consumer experience.
LI Zhijin, TANG Jiahui, YAN Jinfeng
core +1 more source
Zhendong Ceshi yu Zhenduan针对滚动轴承早期故障特征边缘分布与条件分布权重难以调节，导致分类任务精度低等问题，提出利用动态对抗自适应网络进行滚动轴承早期故障诊断。首先，将源域样本和目标域样本视作不同阶段的故障样本，利用对抗网络减小领域间的分布差异；其次，在全局领域辨别器和局部子领域辨别器的基础上引入动态对抗因子ω，动态调节2种分布的权重，减小领域间分布差异；然后，耦合特征转换层实现特征多样化，提升模型泛化性；最后，使用转子实验台轴承数据集和凯斯西储大学轴承数据集来验证模型性能。结果表明 ...
doaj +1 more source
[Parameter transfer learning based on shallow visual geometry group network and its application in motor imagery classification]. [PDF]
Sheng Wu Yi Xue Gong Cheng Xue Za Zhi, 2022 Xu D, Li M.
europepmc +1 more source