您的检索:
学者姓名:韩红桂
精炼检索结果:
年份
成果类型
收录类型
来源
综合
合作者
语言
清除所有精炼条件
摘要 :
Transfer learning algorithms are capable to apply previously learned knowledge in source domain, which alleviates much expensive efforts of knowledge recollection in target domain. But the knowledge in source domain is always imperfect due to redundant or contaminated information. To solve this problem, an ensemble filtertransfer learning (EFTL) algorithm based on the source knowledge reconstruction is proposed in this paper. First, a knowledge partition strategy based on model is developed to classify the source knowledge into different types. Then, the positive knowledge can be identified, which contributes to target domain with a rejection of the negative transfer. Second, a knowledge filter algorithm is introduced to filter out the redundant information in non-positive knowledge. Then, the non-positive knowledge can be reconstructed by this algorithm to prevent the loss of available information. Third, an ensemble transfer mechanism is established to realize the synchronous transfer of omnidirectional knowledge for the target domain. Finally, comparative experiments on model prediction in practical applications are provided to illustrate the dependability of EFTL.
关键词 :
Knowledge filter algorithm Knowledge filter algorithm Transfer learning Transfer learning Model prediction Model prediction Ensemble transfer mechanism Ensemble transfer mechanism
引用:
复制并粘贴一种已设定好的引用格式,或利用其中一个链接导入到文献管理软件中。
GB/T 7714 | Han, Honggui , Li, Mengmeng , Yang, Hongyan et al. Ensemble filter-transfer learning algorithm [J]. | PATTERN RECOGNITION , 2024 , 154 . |
MLA | Han, Honggui et al. "Ensemble filter-transfer learning algorithm" . | PATTERN RECOGNITION 154 (2024) . |
APA | Han, Honggui , Li, Mengmeng , Yang, Hongyan , Wu, Xiaolong , Han, Huayun . Ensemble filter-transfer learning algorithm . | PATTERN RECOGNITION , 2024 , 154 . |
导入链接 | NoteExpress RIS BibTex |
摘要 :
Feature pyramids are widely adopted in visual detection models for capturing multiscale features of objects. However, the utilization of feature pyramids in practical object detection tasks is prone to complex background interference, resulting in suboptimal capture of discriminative multiscale foreground semantic features. In this article, a foreground capture feature pyramid network (FCFPN) for multiscale object detection is proposed, to address the problem of inadequate feature learning in complex backgrounds. FCFPN consists of a foreground dual attention (FDA) module and a pathway aggregation (PA) structure. Specifically, the FDA mechanism activates top-down foreground channel responses and lateral spatial foreground location features, so that channel and spatial foreground features are adequately captured. Then, the PA module adaptively learns the fusion weights of multiscale features at different levels of the feature pyramid, which enhances the complementarity of semantic information between different levels of the foreground feature maps. Since the fusion weights are learned adaptively based on different pyramid levels, the detection model accordingly retains the gained information of feature sizes and suppresses the conflicting information. The evaluations on public datasets and the self-built complex background dataset demonstrate that the detection average precision (AP) and the feature learning performance of the proposed method are superior compared with other FPNs, which proves the effectiveness of the proposed FCFPN.
关键词 :
Interference Interference Complex backgrounds Complex backgrounds feature pyramid feature pyramid object detection object detection Adaptation models Adaptation models Object detection Object detection Neck Neck Semantics Semantics Feature extraction Feature extraction Detectors Detectors foreground capture foreground capture
引用:
复制并粘贴一种已设定好的引用格式,或利用其中一个链接导入到文献管理软件中。
GB/T 7714 | Han, Honggui , Zhang, Qiyu , Li, Fangyu et al. Foreground Capture Feature Pyramid Network-Oriented Object Detection in Complex Backgrounds [J]. | IEEE TRANSACTIONS ON NEURAL NETWORKS AND LEARNING SYSTEMS , 2024 . |
MLA | Han, Honggui et al. "Foreground Capture Feature Pyramid Network-Oriented Object Detection in Complex Backgrounds" . | IEEE TRANSACTIONS ON NEURAL NETWORKS AND LEARNING SYSTEMS (2024) . |
APA | Han, Honggui , Zhang, Qiyu , Li, Fangyu , Du, Yongping . Foreground Capture Feature Pyramid Network-Oriented Object Detection in Complex Backgrounds . | IEEE TRANSACTIONS ON NEURAL NETWORKS AND LEARNING SYSTEMS , 2024 . |
导入链接 | NoteExpress RIS BibTex |
摘要 :
Distributed learning-based high-dimensional temporal modeling for the industrial Internet of Things (IIoT) has become a prevailing trend. However, traditional distributed learning inefficiently extracts information by straightforward architects, resulting in low modeling accuracy and high communication costs. We propose a distributed hierarchical temporal graph learning (DHTGL) approach. In terminal equipment, we construct an adaptive hierarchical dilation convolutional network to dynamically capture spatiotemporal features by adjusting the dilation factor at each layer. Next, we construct the adaptive graphs according to the connection similarity between dimensions to capture implicit connections. In the edge device, we design a node-edge graph distance calculation based on the Gromov-Wasserstein distance to group feature graphs and construct representative cluster feature graphs. Edge devices upload cluster feature graphs to reduce communication costs while minimizing information loss. In the central server, we incorporate graph attention networks into the graph neural networks for edge updating in training models on clustered feature graphs. Experiments using the public IIoT data sets and the self-built IIoT platform demonstrate the effectiveness of DHTGL in comparison with common distributed learning approaches. The results confirm that DHTGL consumes fewer communications while achieving higher accuracies.
关键词 :
graph convolutional network (GCN) graph convolutional network (GCN) Feature extraction Feature extraction Industrial Internet of Things Industrial Internet of Things Computational modeling Computational modeling Data models Data models Distance learning Distance learning Costs Costs industrial Internet of Things (IIoT) industrial Internet of Things (IIoT) Distributed learning Distributed learning Computer aided instruction Computer aided instruction
引用:
复制并粘贴一种已设定好的引用格式,或利用其中一个链接导入到文献管理软件中。
GB/T 7714 | Li, Fangyu , Lin, Junnuo , Wang, Yu et al. Distributed Hierarchical Temporal Graph Learning for Communication-Efficient High-Dimensional Industrial IoT Modeling [J]. | IEEE INTERNET OF THINGS JOURNAL , 2024 , 11 (17) : 28578-28590 . |
MLA | Li, Fangyu et al. "Distributed Hierarchical Temporal Graph Learning for Communication-Efficient High-Dimensional Industrial IoT Modeling" . | IEEE INTERNET OF THINGS JOURNAL 11 . 17 (2024) : 28578-28590 . |
APA | Li, Fangyu , Lin, Junnuo , Wang, Yu , Du, Yongping , Han, Honggui . Distributed Hierarchical Temporal Graph Learning for Communication-Efficient High-Dimensional Industrial IoT Modeling . | IEEE INTERNET OF THINGS JOURNAL , 2024 , 11 (17) , 28578-28590 . |
导入链接 | NoteExpress RIS BibTex |
摘要 :
本发明提出了一种基于延时策略的城市污水处理过程自适应随机采样模型预测控制方法,实现了城市污水处理过程中随机采样的溶解氧浓度和硝态氮浓度的稳定控制。构造具有随机时变延时的城市污水处理过程等效系统,建立模糊神经网络预测模型,设计自适应随机采样模型预测控制器,解决了城市污水处理过程中随机采样的溶解氧浓度和硝态氮浓度难以稳定控制的问题。实验结果表明该方法能够实现城市污水处理过程溶解氧浓度和硝态氮浓度的稳定控制,保证城市污水处理过程的安全稳定运行。
引用:
复制并粘贴一种已设定好的引用格式,或利用其中一个链接导入到文献管理软件中。
GB/T 7714 | 韩红桂 , 付世佳 , 孙浩源 et al. 一种基于延时策略的城市污水处理过程自适应随机采样模型预测控制方法 : CN202310723615.6[P]. | 2023-06-17 . |
MLA | 韩红桂 et al. "一种基于延时策略的城市污水处理过程自适应随机采样模型预测控制方法" : CN202310723615.6. | 2023-06-17 . |
APA | 韩红桂 , 付世佳 , 孙浩源 , 刘峥 , 乔俊飞 . 一种基于延时策略的城市污水处理过程自适应随机采样模型预测控制方法 : CN202310723615.6. | 2023-06-17 . |
导入链接 | NoteExpress RIS BibTex |
摘要 :
一种基于可变形卷积的印制线路板电子元器件目标检测方法属于计算机视觉领域。针对印制电路板(PCB)目标检测过程中由于拍摄角度多样化产生的检测对象形变问题,实现对PCB电子元器件的识别定位。该检测方法通过可变形卷积结构,使得卷积核根据目标的尺寸大小以及形状改变自适应地进行变形,提升神经网络对未知变化的适应性,增强泛化能力,从而提升PCB目标检测的精度;解决了当前基于深度神经网络目标检测难以解决物体空间形变的问题;实验结果表明该方法能够准确地对PCB电子元器件进行检测,并具有较强的自适应能力,提升目前废弃电器电子产品回收拆解过程中的智能化程度,提升工艺的自动化程度。
引用:
复制并粘贴一种已设定好的引用格式,或利用其中一个链接导入到文献管理软件中。
GB/T 7714 | 李方昱 , 牛末寒 , 刘峥 et al. 一种基于可变形卷积的印制线路板电子元器件目标检测方法 : CN202310621681.2[P]. | 2023-05-30 . |
MLA | 李方昱 et al. "一种基于可变形卷积的印制线路板电子元器件目标检测方法" : CN202310621681.2. | 2023-05-30 . |
APA | 李方昱 , 牛末寒 , 刘峥 , 韩红桂 . 一种基于可变形卷积的印制线路板电子元器件目标检测方法 : CN202310621681.2. | 2023-05-30 . |
导入链接 | NoteExpress RIS BibTex |
摘要 :
本发明提出了一种城市污水处理过程时滞鲁棒优化控制方法,该方法设计了城市污水处理过程时滞多目标优化模型,结合数据驱动建模方法与时滞补偿方法,建立了出水水质与能耗的实时优化目标函数,通过设计时滞鲁棒进化优化算法来求解鲁棒性强的溶解氧和硝态氮设定值,并利用比例积分微分控制器实时跟踪所求解的优化设定值,在污水处理过程时滞扰动的影响下,保证了出水水质的同时降低了运行能耗,实现了污水处理厂的高效稳定运行。
引用:
复制并粘贴一种已设定好的引用格式,或利用其中一个链接导入到文献管理软件中。
GB/T 7714 | 韩红桂 , 周昊 , 张嘉成 et al. 一种城市污水处理过程时滞鲁棒优化控制方法 : CN202310614392.X[P]. | 2023-05-29 . |
MLA | 韩红桂 et al. "一种城市污水处理过程时滞鲁棒优化控制方法" : CN202310614392.X. | 2023-05-29 . |
APA | 韩红桂 , 周昊 , 张嘉成 , 黄琰婷 . 一种城市污水处理过程时滞鲁棒优化控制方法 : CN202310614392.X. | 2023-05-29 . |
导入链接 | NoteExpress RIS BibTex |
摘要 :
本发明公开了基于自编码器和双向长短时记忆网络的加密货币交易追踪方法,实现对交易网络中未来有可能出现的交易预测,达到交易追踪的目的;该预测模型可以自动学习高维非线性的网络结构,并且可以学习时间特征捕捉网络的时变特性,通过微调网络结构可以使模型适应不同尺度的网络;解决了现有方法存在的不足,提出了一种通用模型;实验结果表明该方法能够取得较好的结果。
引用:
复制并粘贴一种已设定好的引用格式,或利用其中一个链接导入到文献管理软件中。
GB/T 7714 | 黄静 , 丁金飞 , 韩红桂 et al. 基于自编码器和双向长短时记忆网络的加密货币交易追踪方法 : CN202310479078.5[P]. | 2023-04-28 . |
MLA | 黄静 et al. "基于自编码器和双向长短时记忆网络的加密货币交易追踪方法" : CN202310479078.5. | 2023-04-28 . |
APA | 黄静 , 丁金飞 , 韩红桂 , 公备 , 牛裕茸 . 基于自编码器和双向长短时记忆网络的加密货币交易追踪方法 : CN202310479078.5. | 2023-04-28 . |
导入链接 | NoteExpress RIS BibTex |
摘要 :
本发明提出了一种基于部分标签特征学习的生活垃圾识别方法,针对数据缺少标签使模型难以挖掘数据特征,导致分类精度低的问题。本发明在真实垃圾图像数据的基础上,建立基于部分标签特征学习的生活垃圾识别方法,利用信息不确定性选择具有丰富未知信息的样本进行标注,实现特征的自适应学习,并降低学习成本,最终完成垃圾的准确分类。这种基于部分标签特征学习的生活垃圾识别方法在实际垃圾回收过程中,可以解决垃圾由于数据量大,标签难以获取,从而导致垃圾特征挖掘不充分的问题,实现了高精度垃圾分类,为垃圾回收行业提供技术支持。
引用:
复制并粘贴一种已设定好的引用格式,或利用其中一个链接导入到文献管理软件中。
GB/T 7714 | 韩红桂 , 范晓晔 , 李方昱 et al. 一种基于部分标签特征学习的生活垃圾识别方法 : CN202310557795.5[P]. | 2023-05-18 . |
MLA | 韩红桂 et al. "一种基于部分标签特征学习的生活垃圾识别方法" : CN202310557795.5. | 2023-05-18 . |
APA | 韩红桂 , 范晓晔 , 李方昱 , 杜永萍 . 一种基于部分标签特征学习的生活垃圾识别方法 : CN202310557795.5. | 2023-05-18 . |
导入链接 | NoteExpress RIS BibTex |
摘要 :
一种基于任务聚类的污水处理过程多工况双层优化控制方法,属于污水处理领域。为了实现污水处理过程中多工况双层优化控制,本发明建立基于数据的污水处理过程多工况双层优化模型,分别描述优化设定值与每个工况双层优化目标间的关系,包括每个工况中的领导层优化目标温室气体排放量模型和跟随层优化目标运行能耗模型,研究基于任务聚类的优化设定方法,求解污水处理过程溶解氧和硝态氮优化设定值,并设计模糊神经网络控制器完成优化设定值跟踪控制,从而促进污水处理过程多工况双层运行优化控制。
引用:
复制并粘贴一种已设定好的引用格式,或利用其中一个链接导入到文献管理软件中。
GB/T 7714 | 韩红桂 , 白星 , 侯莹 . 一种基于任务聚类的污水处理过程多工况双层优化控制方法 : CN202310884510.9[P]. | 2023-07-18 . |
MLA | 韩红桂 et al. "一种基于任务聚类的污水处理过程多工况双层优化控制方法" : CN202310884510.9. | 2023-07-18 . |
APA | 韩红桂 , 白星 , 侯莹 . 一种基于任务聚类的污水处理过程多工况双层优化控制方法 : CN202310884510.9. | 2023-07-18 . |
导入链接 | NoteExpress RIS BibTex |
摘要 :
本发明涉及一种生活垃圾收集过程臭气智能监测预警方法,实现了臭气浓度的预测,并确定预警级别,实现臭气智能监测预警,包括以下步骤:首先,采集数据并对数据进行预处理;确定模型的输入变量和输出变量;然后,采用模块化神经网络建立臭气浓度预测模型;最后,参考《恶臭污染物排放标准(GB 14554‑93)》及《室内空气质量标准(GB/T 18883‑2002)》,设置硫化氢H2S、氨气NH3浓度预警阈值,拟定臭气预警级别,实现臭气监测预警。本发明有效地实现了臭气浓度的准确预测以及监测预警,具有重要的理论意义和应用价值。
引用:
复制并粘贴一种已设定好的引用格式,或利用其中一个链接导入到文献管理软件中。
GB/T 7714 | 蒙西 , 王岩 , 孙子健 et al. 一种生活垃圾收集过程臭气智能监测预警方法 : CN202310657368.4[P]. | 2023-06-05 . |
MLA | 蒙西 et al. "一种生活垃圾收集过程臭气智能监测预警方法" : CN202310657368.4. | 2023-06-05 . |
APA | 蒙西 , 王岩 , 孙子健 , 韩红桂 . 一种生活垃圾收集过程臭气智能监测预警方法 : CN202310657368.4. | 2023-06-05 . |
导入链接 | NoteExpress RIS BibTex |
导出
数据: |
选中 到 |
格式: |