Query：
学者姓名：张振昌
Refining：
Year
Type
Indexed by
Source
Complex
Co-Author
Language
Clean All
Abstract ：
Mental fatigue detection plays an important role in preventing fatigue-related diseases and reducing traffic accidents caused by mental exhaustion. In this effort, existing studies have presented interesting results by using physiological signals. However, most works focus primarily on single physiological signal like electroencephalography (EEG). To address this gap, we propose an innovative cross-modal fusion method (CM-FusionNet) and conduct a multi-modal study using EEG and electrooculogram (EOG) for mental fatigue detection. Specifically, a variance channel attention (VCA) module is introduced to adaptively learn the optimal weights for each channel. Then, a Transformer fusion module is applied to extract and integrate the global features of EEG and EOG. Finally, we classify mental fatigue using the fused features. With this method, we conduct independent and cross-subject experiments on the public SEED-VIG dataset. The results of multi-modal experiment show an average accuracy of 84.62% and F1-score of 85.25%, an increase by 1.48% in accuracy 2.46% and in F1-score compared to the EOG-only experiment, and increase by 2.88% and 3.92% compared to the EEG-only experiment, respectively. This demonstrates the benefits of incorporating multi-modalities in fatigue detection and highlights the increased accuracy achieved with our CM-FusionNet approach. It also indicates that this method has potential for further exploration in the field of biomedical signal processing.
Keyword ：
Channel attention Channel attention Cross-modal fusion Cross-modal fusion Electroencephalogram Electroencephalogram Electrooculogram Electrooculogram Fatigue detection Fatigue detection
Cite：
Copy from the list or Export to your reference management。
| GB/T 7714 | Huang, Fuzhong , Yang, Chunfeng , Weng, Wei et al. CM-FusionNet: A cross-modal fusion fatigue detection method based on electroencephalogram and electrooculogram [J]. | COMPUTERS & ELECTRICAL ENGINEERING , 2025 , 123 . |
| MLA | Huang, Fuzhong et al. "CM-FusionNet: A cross-modal fusion fatigue detection method based on electroencephalogram and electrooculogram" . | COMPUTERS & ELECTRICAL ENGINEERING 123 (2025) . |
| APA | Huang, Fuzhong , Yang, Chunfeng , Weng, Wei , Chen, Zelong , Zhang, Zhenchang . CM-FusionNet: A cross-modal fusion fatigue detection method based on electroencephalogram and electrooculogram . | COMPUTERS & ELECTRICAL ENGINEERING , 2025 , 123 . |
| Export to | NoteExpress RIS BibTex |
Version ：
Abstract ：
Recent years have seen the significant potential of Artificial Intelligence (AI) techniques in monitoring and managing ocean ecosystem. Among these, deep learning (DL) solution demonstrated considerable performance in forecasting drifter trajectory, which provide critical scientific support in a variety of ways. However, most existing DL methods still suffer from stability and accuracy problems, due mainly to the fact that they rely primarily on historical memory of tracking target, and that attention-introduced bias further limits their applicability. To address these challenges, this study proposes an innovative drifter trajectory prediction framework (DriftNet) based on newly designed 1) Target-Area Differential Attention (TADA) mechanism and, 2) Direction-Distance Loss function (DDL). For TADA, we constructed a tempo-spatial structure by combining the initial target position and surrounding environment features, then we introduced a Differential Transformer to the integrated features. DDL is an improved loss function to jointly optimize directional alignment and spatial displacement in drift trajectory prediction. Utilizing this model, a case study was conducted in the Taiwan Strait, and the results showed that DrifterNet improved drifter prediction accuracy by over 50% compared to the Lag-fjhyj operational model. This achievement indicates that DriftNet is not only able to better capture the features of target and surrounding area, but better integrate the interaction of timestamp data and spatial data, which further deepens our understanding of the interaction between time-domain factors and spatial-domain factors in maritime environment. This study presents a novel and effective solution for marine drift prediction, offering both theoretical insights and practical value for real-world ocean monitoring and emergency response scenarios.
Keyword ：
direction-distance loss direction-distance loss DriftNet DriftNet Drift trajectory Drift trajectory TADA TADA Taiwan strait Taiwan strait
Cite：
Copy from the list or Export to your reference management。
| GB/T 7714 | Chen, Yongxiang , Chen, Haiqiang , Jiang, Yuwu et al. DriftNet: target-area differential attention mechanism for marine drifter trajectory prediction [J]. | ENGINEERING APPLICATIONS OF COMPUTATIONAL FLUID MECHANICS , 2025 , 19 (1) . |
| MLA | Chen, Yongxiang et al. "DriftNet: target-area differential attention mechanism for marine drifter trajectory prediction" . | ENGINEERING APPLICATIONS OF COMPUTATIONAL FLUID MECHANICS 19 . 1 (2025) . |
| APA | Chen, Yongxiang , Chen, Haiqiang , Jiang, Yuwu , Wang, Chiming , Wang, Ning , Zhang, Feng et al. DriftNet: target-area differential attention mechanism for marine drifter trajectory prediction . | ENGINEERING APPLICATIONS OF COMPUTATIONAL FLUID MECHANICS , 2025 , 19 (1) . |
| Export to | NoteExpress RIS BibTex |
Version ：
Abstract ：
Developing a universal model capable of effective cross-subject fatigue detection is crucial for alerting fatigued drivers in a timely manner and reducing traffic accidents. Although existing domain adaptation methods can reduce distribution differences between subjects, they primarily focus on inter-subject feature (inter-domain) alignment while neglecting category-level feature (intra-domain) alignment. To address this issue, this paper proposes a novel Multi-Level Domain Adaptation (MLDA) algorithm to enhance the model's generalization ability on unseen data. Specifically, for inter-domain alignment, we employ an improved Wasserstein distance, whose smooth properties more accurately measure inter-domain differences. For intra-domain alignment, we introduce intra-domain contrastive discrepancy, which enhances the discriminability of category features by maximizing inter-class distances and minimizing intra-class distances. The proposed method achieves cross-subject fatigue detection accuracies of 0.942 and 0.843 on the SEED-VIG and SADT public datasets, respectively. Experimental results demonstrate that MLDA offers significant advantages in cross-subject fatigue detection tasks, providing a promising solution fora generalized driver fatigue detection system.
Keyword ：
Brain-computer interface Brain-computer interface Cross-subject fatigue detection Cross-subject fatigue detection Domain adaptation Domain adaptation Electroencephalogram Electroencephalogram Transfer learning Transfer learning
Cite：
Copy from the list or Export to your reference management。
| GB/T 7714 | Huang, Fuzhong , Wang, Qicong , Chen, Lei et al. Multi-level domain adaptation for improved generalization in electroencephalogram-based driver fatigue detection [J]. | ENGINEERING APPLICATIONS OF ARTIFICIAL INTELLIGENCE , 2025 , 142 . |
| MLA | Huang, Fuzhong et al. "Multi-level domain adaptation for improved generalization in electroencephalogram-based driver fatigue detection" . | ENGINEERING APPLICATIONS OF ARTIFICIAL INTELLIGENCE 142 (2025) . |
| APA | Huang, Fuzhong , Wang, Qicong , Chen, Lei , Mei, Wang , Zhang, Zhenchang , Chen, Zelong . Multi-level domain adaptation for improved generalization in electroencephalogram-based driver fatigue detection . | ENGINEERING APPLICATIONS OF ARTIFICIAL INTELLIGENCE , 2025 , 142 . |
| Export to | NoteExpress RIS BibTex |
Version ：
Abstract ：
This study investigates the prediction of dissolved oxygen (DO) in the Taiwan Strait, a crucial indicator for marine ecosystems. Accurate prediction of DO in coastal waters is critical for various coastal engineering activities. Despite the availability of numerous prediction methods, including numerical models, statistical models, and machine learning approaches, these often fall short when addressing complex dynamic marine areas like the Taiwan Strait. This challenge is partly due to the complexity of the environment and issues with data scarcity. The research focuses on seven buoy stations in the Taiwan Strait and proposes the Attention-based Multi-teacher Knowledge Distillation (AMKD) model, which integrates a multi-channel attention mechanism with a multi-teacher knowledge distillation algorithm, specifically enhancing the accuracy of DO prediction and adaptability to missing data. The model is capable of forecasting DO hourly for the next 24 h, effectively mitigating the randomness and instability associated with DO. Experimental comparisons with state-of-the-art prediction models demonstrate that our approach outperforms commonly used methods in terms of accuracy. Overall, the AMKD model presents a novel and effective solution for predicting DO in complex marine areas, with significant implications for future marine environmental monitoring and management.
Keyword ：
Deep learning Deep learning Dissolved oxygen Dissolved oxygen Prediction methods Prediction methods Taiwan Strait Taiwan Strait
Cite：
Copy from the list or Export to your reference management。
| GB/T 7714 | Chen, Lei , Lin, Ye , Guo, Minquan et al. Dissolved oxygen prediction in the Taiwan Strait with the attention-based multi-teacher knowledge distillation model [J]. | OCEAN & COASTAL MANAGEMENT , 2025 , 265 . |
| MLA | Chen, Lei et al. "Dissolved oxygen prediction in the Taiwan Strait with the attention-based multi-teacher knowledge distillation model" . | OCEAN & COASTAL MANAGEMENT 265 (2025) . |
| APA | Chen, Lei , Lin, Ye , Guo, Minquan , Lu, Wenfang , Li, Xueding , Zhang, Zhenchang . Dissolved oxygen prediction in the Taiwan Strait with the attention-based multi-teacher knowledge distillation model . | OCEAN & COASTAL MANAGEMENT , 2025 , 265 . |
| Export to | NoteExpress RIS BibTex |
Version ：
Abstract ：
Accurate prediction of sea surface temperature (SST) plays a critical role in climate research and marine ecosystem management. Traditional models predict trends by analyzing and fitting data, but they struggle with capturing long-range dependencies and complex spatiotemporal patterns. The transformer's attention mechanism effectively addresses long-range dependencies, but its high computational complexity poses challenges. To overcome these limitations, this study proposes a novel spatiotemporal sequence prediction model: the spatiotemporal vision mamba recurrent neural network (SVRNN). The model innovatively integrates a bidirectional state-space processing mechanism and decoupled memory modules. The bidirectional mechanism maintains a global receptive field with linear computational complexity, while the decoupled memory modules explicitly separate spatiotemporal dependencies, enhancing the model's ability to capture complex spatiotemporal patterns. During the experiment on hourly SST prediction in the Taiwan Strait, where the SST of the next 12 h was predicted using data from the previous 12 h, the SVRNN model demonstrated superior performance, achieving a root mean square error (RMSE) of 0.159 degrees C, a mean absolute error (MAE) of 0.105 degrees C, and a mean absolute percentage error (MAPE) of 0.496%. Furthermore, our seasonal error analysis reveals that the model exhibits robust performance in different seasons, providing more reliable technical support for SST prediction in Taiwan Strait.
Keyword ：
Accuracy Accuracy Computational modeling Computational modeling Data models Data models Logic gates Logic gates Long short term memory Long short term memory Memory modules Memory modules Ocean temperature Ocean temperature Predictive models Predictive models recurrent neural network (RNN) recurrent neural network (RNN) sea surface temperature (SST) sea surface temperature (SST) Spatiotemporal phenomena Spatiotemporal phenomena spatiotemporal prediction model spatiotemporal prediction model Training Training Transformers Transformers vision mamba vision mamba
Cite：
Copy from the list or Export to your reference management。
| GB/T 7714 | Chen, Haiqiang , Chen, Yongxiang , Zhang, Zhenchang . SVRNN: A Spatiotemporal Prediction Model for Sea Surface Temperature Prediction in the Taiwan Strait [J]. | IEEE GEOSCIENCE AND REMOTE SENSING LETTERS , 2025 , 22 . |
| MLA | Chen, Haiqiang et al. "SVRNN: A Spatiotemporal Prediction Model for Sea Surface Temperature Prediction in the Taiwan Strait" . | IEEE GEOSCIENCE AND REMOTE SENSING LETTERS 22 (2025) . |
| APA | Chen, Haiqiang , Chen, Yongxiang , Zhang, Zhenchang . SVRNN: A Spatiotemporal Prediction Model for Sea Surface Temperature Prediction in the Taiwan Strait . | IEEE GEOSCIENCE AND REMOTE SENSING LETTERS , 2025 , 22 . |
| Export to | NoteExpress RIS BibTex |
Version ：
Abstract ：
The introduction of Network Function Virtualization (NFV) and Software-Defined Network (SDN) architectures has significantly reduced the Operational Expenditure (OPEX) and Capital Expenditure (CAPEX) of network system. However, NFV orchestration management also brings about challenges. After the initial deployment of VNFs, due to the volatility of network requests, the original deployment may not be able to meet user resource demands. The key issue is how to readjust resources dynamically to accommodate more network requests without violating Quality of Service (QoS) for users. Several existing techniques can be used to achieve this goal, such as horizontal scaling, vertical scaling, and virtual network function (VNF) migration. However, these techniques inevitably incur some overhead, such as the cost of instantiating VNF and link rerouting. Additionally, resource adjustment may also result in unbalanced distribution of network resources. In this paper, an Intelligent Service Function Chain Dynamic Adjustment Algorithm (ISFCDAA) is proposed to address the above challenges. Firstly, an Integer Linear Programming (ILP) model is established with the objective of minimizing the long-term adjustment cost and reducing the imbalance of resource distribution. Then we transform the optimization process into a Markov Decision Process (MDP). Secondly, to solve the problems that the state and action space is too large and the state transition probability is uncertain in MDP, a SFC dynamic adjustment algorithm based on deep reinforcement learning is proposed. This algorithm can obtain an approximate optimal adjustment strategy for ILP model. The simulation results show that ISFCDAA can reduce the adjustment overhead and maintain a balanced distribution of network resources while ensuring the QoS. Compared with the existing algorithms, the average standard deviation of resource distribution of ISFCDAA is reduced by up to 9.90%, the average acceptance rate of ISFCDAA is improved by up to 39.57%, and the average long-term profit is improved by up to 42.92%. The incorporation of cost and demand-sensitive considerations into ISFCDAA enhances its responsiveness to fluctuating network demands, solidifying its effectiveness in dynamic resource management scenarios.
Keyword ：
Network function virtualization orchestrator Network function virtualization orchestrator Resource management Resource management Service function chain Service function chain
Cite：
Copy from the list or Export to your reference management。
| GB/T 7714 | Wang, Yuantao , Shu, Zhaogang , Chen, Shuwu et al. A cost and demand sensitive adjustment algorithm for service function chain in data center network [J]. | COMPUTER NETWORKS , 2024 , 242 . |
| MLA | Wang, Yuantao et al. "A cost and demand sensitive adjustment algorithm for service function chain in data center network" . | COMPUTER NETWORKS 242 (2024) . |
| APA | Wang, Yuantao , Shu, Zhaogang , Chen, Shuwu , Lin, Jiaxiang , Zhang, Zhenchang . A cost and demand sensitive adjustment algorithm for service function chain in data center network . | COMPUTER NETWORKS , 2024 , 242 . |
| Export to | NoteExpress RIS BibTex |
Version ：
Abstract ：
本发明涉及物联网技术领域，公开了一种基于AI和物联网的洗衣机管理系统、方法及介质，包括：物联网传感器模块，其用于采集洗衣机的运行信息；物联网图像采集模块，其用于采集衣物的图像信息和使用者的面部信息；物联网通讯模块，其用于将物联网传感器模块和物联网图像采集模块采集的信息发送到数据处理模块；数据处理模块包括：运行信息处理模块，其基于洗衣机的运行信息来处理获得运行特征数据；异常分析模块，其用于将洗衣机的运行信息和衣物的图像信息输入异常分析模型，输出洗衣机的故障类型；本发明能够通过衣物特征信息的识别来降低仅依靠洗衣机运行数据来判断故障类型的误差。
Cite：
Copy from the list or Export to your reference management。
| GB/T 7714 | 张振昌 , 林甲祥 , 李小林 et al. 一种基于AI和物联网的洗衣机管理系统、方法及介质 : CN202410677583.5[P]. | 2024-05-29 . |
| MLA | 张振昌 et al. "一种基于AI和物联网的洗衣机管理系统、方法及介质" : CN202410677583.5. | 2024-05-29 . |
| APA | 张振昌 , 林甲祥 , 李小林 , 陈宏方 , 林清波 , 方艳 et al. 一种基于AI和物联网的洗衣机管理系统、方法及介质 : CN202410677583.5. | 2024-05-29 . |
| Export to | NoteExpress RIS BibTex |
Version ：
Abstract ：
本发明涉及了一种基于YOLOV7目标检测算法的实时违规旗帜检测方法，包括以下步骤：通过编写爬虫收集违规旗帜图片，确定违规旗帜检测的特征，利用图片标注工具对收集到的图片进行标注，制作违规旗帜数据集，同时，收集与违规旗帜具有相似特征的正常旗帜，制作混淆特征旗帜数据集；对模型算法进行优化，基于焦点和全局目标检测蒸馏，对YOLOV7目标检测算法进行改进，获得自适应孪生蒸馏YOLOV7‑tiny目标检测算法模型；区别于现有技术，本发明通过制作混淆特征旗帜数据集，提高违规旗帜识别能力，通过基于焦点和全局目标检测蒸馏，对YOLOV7目标检测算法进行改进，获得孪生蒸馏YOLOV7‑tiny目标检测算法模型，提高检测精度，可以达到实时检测。
Cite：
Copy from the list or Export to your reference management。
| GB/T 7714 | 张振昌 , 刘金强 , 薛弘晖 et al. 基于YOLOV7目标检测算法的实时违规旗帜检测方法 : CN202310327582.3[P]. | 2023-03-30 . |
| MLA | 张振昌 et al. "基于YOLOV7目标检测算法的实时违规旗帜检测方法" : CN202310327582.3. | 2023-03-30 . |
| APA | 张振昌 , 刘金强 , 薛弘晖 , 林甲祥 , 林清波 , 李小林 et al. 基于YOLOV7目标检测算法的实时违规旗帜检测方法 : CN202310327582.3. | 2023-03-30 . |
| Export to | NoteExpress RIS BibTex |
Version ：
Abstract ：
为解决头皮脑电信号电位信号微弱、易受干扰、具有非平稳性和随机性、手动提取特难度大的问题，提出了CNN+BiGRU的网络模型，充分提取了EEG信号前后序列之间的关联信息。实验结果表明，提出双流网络模型准确率达到95.70%。与现有的研究方法相比，显著提高了单通道脑电信号进行疲劳检测的准确性与可行性，为疲劳检测研究提供了新思路。
Keyword ：
BiGRU BiGRU 深度学习 深度学习 脑电信号 脑电信号
Cite：
Copy from the list or Export to your reference management。
| GB/T 7714 | 张怀勇 , 张振昌 . 单通道脑电信号疲劳分类检测 [J]. | 闽江学院学报 , 2024 , 45 (05) : 31-41 . |
| MLA | 张怀勇 et al. "单通道脑电信号疲劳分类检测" . | 闽江学院学报 45 . 05 (2024) : 31-41 . |
| APA | 张怀勇 , 张振昌 . 单通道脑电信号疲劳分类检测 . | 闽江学院学报 , 2024 , 45 (05) , 31-41 . |
| Export to | NoteExpress RIS BibTex |
Version ：
Abstract ：
Recently, in order to pursue better detection results, more convolutional layers and deeper networks are a direction pursued by everyone. However, more and more down-sampling convolution or up-sampling operations generate feature maps of different scales, which makes it difficult to avoid the loss of detailed information of the image, and the distribution of different scales features will be misaligned. In particular, the loss and dislocation of the target boundary information will affect the features learned by the model and reduce the accuracy. This paper proposes a feature alignment method based on non-local idea, and designed two modules—Non Local Align Module (NLA) and Channel Fusion Augment Module (CFA). At the same time, the neighborhood calculation algorithm is also designed for it, which strengthens the binding force on the calculation of boundary information. These two modules can be easily embedded into the current mainstream object detection network to improve the detection effect of the model. Compared to the previous model, the network using our NLA module and CFA module achieves better results than the original model. © 2023, The Author(s), under exclusive license to Springer Nature Switzerland AG.
Keyword ：
Convolution Convolution Feature extraction Feature extraction Network layers Network layers Object detection Object detection Object recognition Object recognition Signal sampling Signal sampling
Cite：
Copy from the list or Export to your reference management。
| GB/T 7714 | Xue, Honghui , Ma, Jinshan , Cai, Zheyi et al. NLFA: A Non Local Fusion Alignment Module for Multi-Scale Feature in Object Detection [J]. | Mechanisms and Machine Science , 2023 , 138 : 155-173 . |
| MLA | Xue, Honghui et al. "NLFA: A Non Local Fusion Alignment Module for Multi-Scale Feature in Object Detection" . | Mechanisms and Machine Science 138 (2023) : 155-173 . |
| APA | Xue, Honghui , Ma, Jinshan , Cai, Zheyi , Fu, Junfang , Guo, Feng , Weng, Wei et al. NLFA: A Non Local Fusion Alignment Module for Multi-Scale Feature in Object Detection . | Mechanisms and Machine Science , 2023 , 138 , 155-173 . |
| Export to | NoteExpress RIS BibTex |
Version ：
Export
| Results： |
Selected to |
| Format： |
