Query：
学者姓名：江灏
Refining：
Year
Type
Indexed by
Source
Complex
Former Name
Co-
Language
Clean All
Abstract ：
In the field of multi-style textile defect detection, a common challenge is the difficulty of adapting the inherent detection model to different styles of textile defects. Changes in the color or style of the textile often result in a decrease in the accuracy of defect detection. Relying solely on the model for fine-tuning inspections can lead to catastrophic forgetting, which significantly impacts the performance of the textile defect detector. To address these challenges, a multi-task correlation distillation (MTCD) anomaly detection method based on knowledge distillation and representative sampling is proposed to detect multi-style textile defects. To enable MTCD to detect defects of new-style textiles while maintaining the detection of old-style textiles, two main modules are introduced. The distillation adaptation module (DAM) explores the intra-feature correlation in the feature space of the target detector, allowing the student model to acquire knowledge of new-style textile defect detection while inheriting the teacher model's detection ability for old-style textile defects. The representative sampling module (RSM) stores representative knowledge of textile defect detection for old-style textiles, facilitating the transfer of knowledge learned from detecting new-style textile defect styles and maintaining the ability to detect defects in old-style textiles. This increases the detection accuracy of the student model for new-style textile defects. The results show that the proposed MTCD method can adapt to the new textile defect detection while maintaining the accuracy of the old textile defect detection and avoiding the problem of catastrophic forgetting. Furthermore, it offers a better balance between stability and plasticity, making it a promising solution for defect detection of multi-style textiles in industrial production environments. © 2024 SPIE and IS&T.
Keyword ：
Anomaly detection Anomaly detection Defects Defects Distillation Distillation Knowledge management Knowledge management Textiles Textiles
Cite：
Copy from the list or Export to your reference management。
| GB/T 7714 | Jiang, Hao , Huang, Shicong , Jin, Zhiheng et al. Multi-style textile defect detection using distillation adaptation and representative sampling [J]. | Journal of Electronic Imaging , 2024 , 33 (3) . |
| MLA | Jiang, Hao et al. "Multi-style textile defect detection using distillation adaptation and representative sampling" . | Journal of Electronic Imaging 33 . 3 (2024) . |
| APA | Jiang, Hao , Huang, Shicong , Jin, Zhiheng , Zhang, Minggui , Chen, Jing , Miao, Xiren . Multi-style textile defect detection using distillation adaptation and representative sampling . | Journal of Electronic Imaging , 2024 , 33 (3) . |
| Export to | NoteExpress RIS BibTex |
Version ：
Abstract ：
超宽带(Ultra-Wideband, UWB)技术在室内外定位中应用广泛，针对传统多基站定位方案的局限性，提出了一种基于超宽带信号到达相位差(Ultra-Wideband Phase Difference of Arrival, UWB-PDOA)的少基站自适应定位系统。该系统利用UWB-PDOA技术和基于ESP32信号强度的权重自适应定位技术，大幅降低了对环境部署的依赖性，提高了定位的精度和稳定性。结合环境先验信息和目标高度的先验知识，构建了先验知识库，采用自适应定位技术，利用多个传感器的信息来调整对不同定位基站的置信度权重，进一步提高了定位精度和鲁棒性。实验结果表明，所提出的系统在视距(Line of Sight, LOS)和非视距(Non Line of Sight, NLOS)环境下都具有较高的定位精度和稳定性，并且仅需要不超过3个基站便可以满足室内环境定位的需求。
Keyword ：
UWB-PDOA UWB-PDOA 信号强度 信号强度 少基站定位 少基站定位 自适应 自适应
Cite：
Copy from the list or Export to your reference management。
| GB/T 7714 | 黄鑫 , 张成炜 , 韦周旺 et al. 基于UWB-PDOA的少基站自适应定位系统研究 [J]. | 测控技术 , 2024 , 43 (05) : 85-92 . |
| MLA | 黄鑫 et al. "基于UWB-PDOA的少基站自适应定位系统研究" . | 测控技术 43 . 05 (2024) : 85-92 . |
| APA | 黄鑫 , 张成炜 , 韦周旺 , 江灏 . 基于UWB-PDOA的少基站自适应定位系统研究 . | 测控技术 , 2024 , 43 (05) , 85-92 . |
| Export to | NoteExpress RIS BibTex |
Version ：
Abstract ：
热状态参量预测是特高压变压器绝缘老化评估及故障预警的重要技术方法.然而,现有预测方法侧重高维时间序列分析以构建数据驱动模型,未计及设备内部温度潜在的空间变化规律,为此,提出一种基于时空特征挖掘的特高压变压器热状态参量预测方法.首先,综合考虑多源数据间的相关度与冗余度,提出组合特征筛选策略寻找最优特征子集;其次,结合热状态参量的最优特征子集及相关系数,构建面向热状态参量预测的时空图数据;最后,建立双重自适应图卷积门控循环单元(dual adaptive graph convolution gate recurrent unit,DA-GCGRU)模型,采用节点自适应模块强化油箱内不同部位温度变化趋势的拟合,以适应特定温升趋势;采用图自适应模块自主学习热状态参量的空间温度分布关联性,以推断空间映射关系.实验结果表明,该方法可深度挖掘特高压变压器内部温度的时空变化特性,准确预测绕组温度和顶层油温的变化趋势,具有较好的鲁棒性和泛化性.
Keyword ：
图卷积网络 图卷积网络 特高压变压器 特高压变压器 绕组温度 绕组温度 自适应 自适应 门控循环单元 门控循环单元 顶层油温 顶层油温
Cite：
Copy from the list or Export to your reference management。
| GB/T 7714 | 林蔚青 , 缪希仁 , 肖洒 et al. 基于时空特征挖掘的特高压变压器热状态参量预测方法 [J]. | 中国电机工程学报 , 2024 , 44 (4) : 1649-1661,中插33 . |
| MLA | 林蔚青 et al. "基于时空特征挖掘的特高压变压器热状态参量预测方法" . | 中国电机工程学报 44 . 4 (2024) : 1649-1661,中插33 . |
| APA | 林蔚青 , 缪希仁 , 肖洒 , 江灏 , 卢燕臻 , 邱星华 et al. 基于时空特征挖掘的特高压变压器热状态参量预测方法 . | 中国电机工程学报 , 2024 , 44 (4) , 1649-1661,中插33 . |
| Export to | NoteExpress RIS BibTex |
Version ：
Abstract ：
针对当前高校实验室异常用电行为，提出一种基于Stacking相异模型融合的异常行为检测方法。考虑相异基学习器挖掘实验室用电行为规律的差异性，对相异基学习器进行优选。利用随机森林作为元学习器，充分融合相异基学习器的优势，弥补各基学习器的缺陷，构建基于Stacking相异模型融合的集成学习模型。通过算例对比分析，验证了基于Stacking相异模型融合的集成学习模型能有效提升单一分类器的异常检测效果，在准确率、F_1分数、ROC曲线下面积和误检率上均优于Bagging、Voting、Adaboost等集成学习方法并能适应样本不平衡的情况。
Keyword ：
Stacking结合策略 Stacking结合策略 实验室安全 实验室安全 异常用电行为 异常用电行为 集成学习 集成学习
Cite：
Copy from the list or Export to your reference management。
| GB/T 7714 | 陈静 , 王铭海 , 江灏 et al. Stacking相异模型融合的实验室异常用电行为检测 [J]. | 实验室研究与探索 , 2024 , 43 (01) : 231-237 . |
| MLA | 陈静 et al. "Stacking相异模型融合的实验室异常用电行为检测" . | 实验室研究与探索 43 . 01 (2024) : 231-237 . |
| APA | 陈静 , 王铭海 , 江灏 , 缪希仁 , 陈熙 , 郑垂锭 . Stacking相异模型融合的实验室异常用电行为检测 . | 实验室研究与探索 , 2024 , 43 (01) , 231-237 . |
| Export to | NoteExpress RIS BibTex |
Version ：
Abstract ：
Fault detection of electrical poles is part of the daily operation of power utilities to ensure the sustainability of power transmission. This paper develops a method for intelligent detection of fallen poles based on the improved YOLOX. The hyper-parameters in this method are optimized automatically by Particle Swarm Optimization (PSO) including batch size and input resolution. During parameter optimization, a specific comprehensive evaluation metric is presented as the fitness function to obtain optimal solutions with low labor cost and high method performance. In addition, virtual pole images are generated by 3D Studio Max to overcome the imbalance problem of normal and fault data. The results show that the proposed method can achieve 95.7% of recall and 98.9% of precision, which demonstrates the high accuracy of the method in fallen pole detection. In the comparative experiment, the proposed PSO-YOLOX method is superior to the existing methods including original YOLOX and Faster R-CNN, which verifies the effectiveness of automatic optimization and virtual data augmentation.
Keyword ：
Fallen poles detection Fallen poles detection Particle Swarm Optimization (PSO) Particle Swarm Optimization (PSO) UAV inspection UAV inspection YOLOX YOLOX
Cite：
Copy from the list or Export to your reference management。
| GB/T 7714 | Jiang, Hao , Wang, Ben , Wu, Li et al. Fallen detection of power distribution poles in UAV inspection using improved YOLOX with particle swarm optimization [J]. | MULTIMEDIA TOOLS AND APPLICATIONS , 2024 . |
| MLA | Jiang, Hao et al. "Fallen detection of power distribution poles in UAV inspection using improved YOLOX with particle swarm optimization" . | MULTIMEDIA TOOLS AND APPLICATIONS (2024) . |
| APA | Jiang, Hao , Wang, Ben , Wu, Li , Chen, Jing , Liu, Xinyu , Miao, Xiren . Fallen detection of power distribution poles in UAV inspection using improved YOLOX with particle swarm optimization . | MULTIMEDIA TOOLS AND APPLICATIONS , 2024 . |
| Export to | NoteExpress RIS BibTex |
Version ：
Abstract ：
For intelligent inspection of power lines, a core task is to detect components in aerial images. Currently, deep supervised learning, a data-hungry paradigm, has attracted great attention. However, considering real-world scenarios, labeled data are usually limited, and the utilization of abundant unlabeled data is rarely investigated in this field. This study deploys a pretrained model for power line component detection based on a self-supervised pretraining approach, which exploits useful information from unannotated data. Concretely, we design a new masking strategy based on the structural characteristic of power lines to guide the pretraining process with meaningful semantic content. Meanwhile, a Siamese architecture is proposed to extract complete global features by using dual reconstruction with semantic targets provided by the proposed masking strategy. Then, the knowledge distillation is utilized to enable the pretrained model to learn both domain-specific and general representations. Moreover, a feature pyramid mechanism is adopted to capture multiscale features, which can benefit the detection task. Experimental results show that the proposed approach can successfully improve the performance of a variety of detection frameworks for power line components, and outperforms other self-supervised pretraining methods.
Keyword ：
Component detection Component detection deep learning deep learning machine vision machine vision power line inspection power line inspection self-supervised pretraining self-supervised pretraining
Cite：
Copy from the list or Export to your reference management。
| GB/T 7714 | Liu, Xinyu , Miao, Xiren , Jiang, Hao et al. Tower Masking MIM: A Self-Supervised Pretraining Method for Power Line Inspection [J]. | IEEE TRANSACTIONS ON INDUSTRIAL INFORMATICS , 2024 , 20 (1) : 513-523 . |
| MLA | Liu, Xinyu et al. "Tower Masking MIM: A Self-Supervised Pretraining Method for Power Line Inspection" . | IEEE TRANSACTIONS ON INDUSTRIAL INFORMATICS 20 . 1 (2024) : 513-523 . |
| APA | Liu, Xinyu , Miao, Xiren , Jiang, Hao , Chen, Jing , Wu, Min , Chen, Zhenghua . Tower Masking MIM: A Self-Supervised Pretraining Method for Power Line Inspection . | IEEE TRANSACTIONS ON INDUSTRIAL INFORMATICS , 2024 , 20 (1) , 513-523 . |
| Export to | NoteExpress RIS BibTex |
Version ：
Abstract ：
[目的]针对现有的坐姿监测方法存在的接触式、隐私性低、成本高、部署不方便等问题对坐姿监测方法进行研究.[方法]提出基于Wi-Fi信道状态信息的坐姿监测方法.该方法在不同坐姿下采集商用路由器的Wi-Fi信道状态信息，结合卷积神经网络和长短期记忆神经网络建立坐姿分类模型，融合采样窗口内信道状态信息的幅值和相位数据，并充分提取数据的空间和时间特征，提高坐姿分类精度.在对原始相位数据进行预处理时，提出了近邻子载波差值阈值补偿方法，有效地解决了不同子载波的相位旋绕不同步的问题.[结果]搭建坐姿监测环境，对办公或学习场景下的5种常见坐姿进行分类.实验证明，该坐姿监测方法对坐姿分类有较高的准确率，对所有坐姿分类的平均准确率达到91.23%.[结论]本文提出的基于Wi-Fi信道状态信息的坐姿监测方法，具有非接触式、隐私性高、成本低、部署方便等特点，且对坐姿分类准确率高，在坐姿监测系统的研究上具有一定的实用价值.
Keyword ：
CNN-LSTM CNN-LSTM CSI CSI Wi-Fi感知 Wi-Fi感知 坐姿监测 坐姿监测
Cite：
Copy from the list or Export to your reference management。
| GB/T 7714 | 刘暾东 , 黄智斌 , 江灏 . 基于Wi-Fi信道状态信息的坐姿监测方法 [J]. | 厦门大学学报(自然科学版) , 2024 , 63 (04) : 649-658 . |
| MLA | 刘暾东 et al. "基于Wi-Fi信道状态信息的坐姿监测方法" . | 厦门大学学报(自然科学版) 63 . 04 (2024) : 649-658 . |
| APA | 刘暾东 , 黄智斌 , 江灏 . 基于Wi-Fi信道状态信息的坐姿监测方法 . | 厦门大学学报(自然科学版) , 2024 , 63 (04) , 649-658 . |
| Export to | NoteExpress RIS BibTex |
Version ：
Abstract ：
With the increasing demand for indoor location-based services (LBS), indoor positioning technology, especially the received signal strength (RSS)-based positioning technology, has attracted extensive attention. In the process of localization, the difference in RSS caused by heterogeneity between different devices cannot be ignored. It leads to the degradation of positioning accuracy. A comprehensive overview of device heterogeneity management methods in indoor positioning is presented in this chapter to deliver a superior solution. An analysis of the causes of device heterogeneity is conducted at the hardware and communication layers. The existing methods to deal with the device heterogeneity are summarized. The approaches for dealing with device heterogeneity are divided into three categories based on the development of technology. The methods are adjustment approach based on linear transformation, calibration-free function mapping method, and non-absolute fingerprint method, respectively. The principles of the implementation for these methods are presented in this chapter. Different evaluation metrics are utilized to participate in the comparison of these methods. The advantages and issues are summarized. Also, the future research trends are proposed at the end of this chapter. © The Editor(s) (if applicable) and The Author(s), under exclusive license to Springer Nature Switzerland AG 2023.
Keyword ：
Calibration-free function mapping Calibration-free function mapping Device heterogeneity Device heterogeneity Linear transformations Linear transformations Non-absolute fingerprint Non-absolute fingerprint Received signal strength (RSS) Received signal strength (RSS)
Cite：
Copy from the list or Export to your reference management。
| GB/T 7714 | Yin, C. , Jiang, H. , Chen, J. . Overview of Approaches for Device Heterogeneity Management During Indoor Localization [未知]. |
| MLA | Yin, C. et al. "Overview of Approaches for Device Heterogeneity Management During Indoor Localization" [未知]. |
| APA | Yin, C. , Jiang, H. , Chen, J. . Overview of Approaches for Device Heterogeneity Management During Indoor Localization [未知]. |
| Export to | NoteExpress RIS BibTex |
Version ：
Abstract ：
With the popularity of smart devices and growing demand of location-based services, machine learning methods have been attracting increasing attention for their potential in indoor localization. Since GPS signal has limited access in indoor environments, alternative sensing solutions have been employed, among which the integration of inertial measurement and WiFi Received Signal Strength (RSS) is the preferred choice for its low cost and pedestrian compatibility. Researchers have proposed various approaches incorporating machine learning algorithms to improve indoor localization performance, which can be broadly divided into the fingerprinting-based approaches and ranging-based approaches. However, these conventional methods still either cannot achieve satisfactory accuracy or need the assistance of other prerequisites to reduce the localization error. To address this issue, in this chapter, we propose a new indoor localization system that integrates the inertial sensing and RSS fingerprinting via a modified Particle Swarm Optimization (PSO)-based algorithm. Different from traditional methods, our proposed method improves the accuracy by a new optimization process, in which the Inertial Measurement Unit (IMU) data is translated into the displacement information serving as the soft constraints of the optimization, and the result of the RSS fingerprinting method provides a guide for the swarm search. Also, a fitness metric based on Gaussian Process Regression (GPR) is developed to evaluate the likelihood of each particle in finding the real position. Experiments are conducted in the real-world scenarios, and the results validate that the proposed approach outperforms the typical approaches by at least 15.8% with the mean localization error of 1.618 m. © The Editor(s) (if applicable) and The Author(s), under exclusive license to Springer Nature Switzerland AG 2023.
Keyword ：
Indoor localization Indoor localization Machine learning Machine learning Particle Swarm Optimization Particle Swarm Optimization Signal radio map Signal radio map WiFi RSS WiFi RSS
Cite：
Copy from the list or Export to your reference management。
| GB/T 7714 | Huang, H. , Yang, J. , Fang, X. et al. Fusion of WiFi and IMU Using Swarm Optimization for Indoor Localization [未知]. |
| MLA | Huang, H. et al. "Fusion of WiFi and IMU Using Swarm Optimization for Indoor Localization" [未知]. |
| APA | Huang, H. , Yang, J. , Fang, X. , Jiang, H. , Xie, L. . Fusion of WiFi and IMU Using Swarm Optimization for Indoor Localization [未知]. |
| Export to | NoteExpress RIS BibTex |
Version ：
Abstract ：
The top oil temperature in ultra-high voltage (UHV) reactors has attracted enormous interest due to its wide applications in fault diagnosis and insulation evaluation. In this work, the precise prediction method based on the Seq2Seq module with the convolutional block attention mechanism is proposed for the UHV reactor. To reduce the influence of vibratility and improve computational efficiency, a combination of the encoding layer and decoding layer named Seq2Seq is performed to reconstruct the complex raw data. The convolutional block attention mechanism (CBAM), composed of spatial attention and channel attention, is utilized to maximize the use of information in data. The Seq2Seq-CBAM is established to forecast the variation tendency of the oil temperatures in the UHV reactor. The experimental results show that the proposed method achieves high prediction accuracy for the top oil temperature in both single-step and multi-step.
Keyword ：
attention attention convolution block attention mechanism (CBAM) convolution block attention mechanism (CBAM) online detection scenario online detection scenario top oil temperature top oil temperature UHV reactor UHV reactor
Cite：
Copy from the list or Export to your reference management。
| GB/T 7714 | Jiang, Hao , Zhang, Hongwei , Chen, Jing et al. Forecasting top oil temperature for UHV reactor using Seq2Seq model with convolutional block attention mechanism [J]. | INTERNATIONAL JOURNAL OF APPLIED ELECTROMAGNETICS AND MECHANICS , 2023 , 73 (4) : 283-302 . |
| MLA | Jiang, Hao et al. "Forecasting top oil temperature for UHV reactor using Seq2Seq model with convolutional block attention mechanism" . | INTERNATIONAL JOURNAL OF APPLIED ELECTROMAGNETICS AND MECHANICS 73 . 4 (2023) : 283-302 . |
| APA | Jiang, Hao , Zhang, Hongwei , Chen, Jing , Xiao, Sa , Miao, Xiren , Lin, Weiqing . Forecasting top oil temperature for UHV reactor using Seq2Seq model with convolutional block attention mechanism . | INTERNATIONAL JOURNAL OF APPLIED ELECTROMAGNETICS AND MECHANICS , 2023 , 73 (4) , 283-302 . |
| Export to | NoteExpress RIS BibTex |
Version ：
Export
| Results： |
Selected to |
| Format： |
