检索结果-内蒙古大学图书馆

A transfer learning-based long short-term memory model for the prediction of river water temperature

ENGINEERING APPLICATIONS OF ARTIFICIAL INTELLIGENCE 2024年第PartF期133卷

作者： Chen, Jinzhou Xue, Xinhua Sichuan Univ Coll Water Resource & Hydropower State Key Lab Hydraul & Mt River Engn Chengdu 610065 Peoples R China

Water temperature affects many physical, chemical and biological processes in rivers and plays a crucial role in determining the quality of aquatic ecosystems. Due to the complexity and nonlinear characteristics of most factors affecting river water temperature, it is difficult for traditional models to accurately predict river water temperature. In this context, accurate prediction of river water temperature calls for new and innovative machine learning techniques. This paper presents a new hybrid model, called LSTM-encoder, that combines long shortterm memory (LSTM) with transformer encoder. To improve the accuracy of the hybrid LSTM-encoder model, wavelet threshold denoising (WTD) method was used to denoise the data. The proposed model was compared with the Air2water model as well as eight other artificial intelligence (AI) models. The results show that the mean absolute percentage error (MAPE), mean absolute error (MAE), root mean squared error (RMSE) and coefficient of determination (R 2 ) values of the WTD-LSTM-encoder model are 2.9, 0.279 degrees C, 0.567 degrees C and 0.867, respectively, for the training datasets and 3.2, 0.312 degrees C, 0.625 degrees C and 0.714, respectively, for the testing datasets, indicating that the proposed WTD-LSTM-encoder model has the best comprehensive performance in predicting the river water temperature.

关键词： Water temperature prediction Long short-term memory Wavelet threshold denoising transformer encoder Air2water model Artificial intelligence

来源：评论

学校读者我要写书评

暂无评论

A transformer-based multi-modal fusion network for 6D pose estimation

引用

INFORMATION FUSION 2024年 105卷

作者： Hong, Jia-Xin Zhang, Hong-Bo Liu, Jing-Hua Lei, Qing Yang, Li-Jie Du, Ji-Xiang Huaqiao Univ Dept Comp Sci & Technol Xiamen 361021 Peoples R China Huaqiao Univ Fujian Key Lab Big Data Intelligence & Secur Xiamen 361021 Peoples R China Huaqiao Univ Xiamen Key Lab Comp Vis & Pattern Recognit Xiamen 361021 Peoples R China Huaqiao Univ Key Lab Comp Vis & Machine Learning Fujian Prov Univ Xiamen 361021 Peoples R China

6D pose estimation has garnered significant attention and research. RGB images and point clouds converted from RGB-D images provide complementary color and geometry information, making them the mainstream data sources for object 6D pose estimation. However, due to the fact that RGB image and point cloud belong to different dimensional spaces and have different distribution characteristics, the fusion of these two complementary data sources remains a key technical challenge for 6D pose estimation. In contrast to prior approaches that simply concatenate separately processed RGB images and point clouds, this work introduces a transformer -based multi -modal fusion network to address this challenge. More precisely, We build a transformer architecture based pixel -wise feature extraction to optimize feature extraction from RGB images and point clouds. Subsequently, we investigate various multi -modal feature fusion methods to process these features, enabling deeper fusion of complementary data. Additionally, during the experimental phase, we design a 6D pose estimation network based on depth prediction to assess the impact of point cloud accuracy on the multi -modal fusion module. Finally, the proposed method is verified on four datasets: LineMOD, Occlusion Linemod, MP6D and YCB-Video. Experimental results show that the proposed method outperforms similar methods on these datasets.

关键词： 6D pose estimation Multi-modal feature fusion Pixel-wise feature extraction transformer encoder

来源：评论

学校读者我要写书评

暂无评论

transformer-BASED ESTIMATION OF SPOKEN SENTENCES USING ELECTROCORTICOGRAPHY 47

TRANSFORMER-BASED ESTIMATION OF SPOKEN SENTENCES USING ELECT...

引用

47th IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)

作者： Komeiji, Shuji Shigemi, Kai Mitsuhashi, Takumi Iimura, Yasushi Suzuki, Hiroharu Sugano, Hidenori Shinoda, Koichi Tanaka, Toshihisa Tokyo Univ Agr & Technol Dept Elect & Informat Engn Tokyo Japan Juntendo Univ Dept Neurosurg Sch Med Tokyo Japan Tokyo Inst Technol Dept Comp Sci Tokyo Japan

ISBN: (纸本)9781665405409

Invasive brain-machine interfaces (BMIs) are a promising neurotechnological venture for achieving direct speech communication from a human brain, but it faces many challenges. In this paper, we measured the invasive electrocorticogram (ECoG) signals from seven participating epilepsy patients as they spoke a sentence consisting of multiple phrases. A transformer encoder was incorporated into a "sequence-to-sequence" model to decode spoken sentences from the ECoG. The decoding test revealed that the use of the transformer model achieved a minimum phrase error rate (PER) of 16.4%, and the median (+/- standard deviation) across seven participants was 31.3% (+/- 10:0%). Moreover, the proposed model with the transformer achieved significantly better decoding accuracy than a conventional long short-term memory model.

关键词： Electrocorticogram (ECoG) Brain-machine interface (BMI) transformer encoder Sequence to sequence

来源：评论

学校读者我要写书评

暂无评论

Depression level prediction via textual and acoustic analysis

引用

Computers in Biology and Medicine 2025年 190卷 110009-110009页

作者： Hong, Jisun Lee, Jihun Choi, Daegil Jung, Jaehyo AI Healthcare Research Center Department of IT Fusion Technology Chosun University 309 Pilmun-daero Dong-gu Gwangju61452 Korea Republic of Department of Bigdata and Artificial Intelligence Eulji Universtiy Seongnam-si13135 Korea Republic of

Extensive research on automatic depression diagnosis has utilized video data to capture related cues, but data collection is challenging because of privacy concerns. By contrast, voice data offer a less-intrusive assessment method and can be analyzed for features such as simple tones, the expression of negative emotions, and a focus on oneself. Recent advancements in multimodal depression-level prediction using speech and text data have gained traction, but most studies overlook the temporal alignment of these modalities, limiting their analysis of the interaction between speech content and intonation. To overcome these limitations, this study introduces timestamp-integrated multimodal encoding for depression (TIMEX-D) which synchronizes the acoustic features of human speech with corresponding text data to predict depression levels on the basis of their relationship. TIMEX-D comprises three main components: a timestamp extraction block that extracts timestamps from speech and text, a multimodal encoding block that extends positional encoding from transformers to mimic human speech recognition, and a depression analysis block that predicts depression levels while reducing model complexity compared with existing transformers. In experiments using the DAIC-WOZ and EDAIC datasets, TIMEX-D achieved accuracies of 99.17 % and 99.81 %, respectively, outperforming previous methods by approximately 13 %. The effectiveness of TIMEX-D in predicting depression levels can enhance mental health diagnostics and monitoring across various contexts. © 2025 Elsevier Ltd

关键词： Depression analysis block Human speech recognition system Multimodal encoding block Positional encoding Timestamp extraction block TIMEX-D transformer encoder

来源：评论

学校读者我要写书评

暂无评论

Hyperspectral Target Detection Using Diffusion Model and Convolutional Gated Linear Unit

引用

IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING 2025年 63卷

作者： Li, Zhaokui Mo, Fan Zhao, Xiaobin Liu, Cuiwei Gong, Xuewei Li, Wei Du, Qian Yuan, Bo Shenyang Aerosp Univ Sch Comp Sci Shenyang 110136 Peoples R China Univ Sci & Technol Beijing Sch Comp & Commun Engn Beijing 100083 Peoples R China Chinese Acad Sci CAS Key Lab Forest Ecol & Silviculture Inst Appl Ecol Shenyang 110016 Peoples R China Beijing Inst Technol Sch Informat & Elect Beijing 100081 Peoples R China Beijing Inst Technol Beijing Key Lab Fract Signals & Syst Beijing 100081 Peoples R China Mississippi State Univ Dept Elect & Comp Engn Starkville MS 39762 USA

Deep learning can effectively extract latent information from data to enhance target-background separation in hyperspectral target detection (HTD). However, these models typically require extensive labeled samples, while available target spectra in hyperspectral images (HSIs) are scarce. In addition, existing deep models struggle with target detection in complex backgrounds due to subtle spectral differences. To address these issues, we propose a novel HTD method based on diffusion model and convolutional gated linear unit (HTD-DMCG). First, the diffusion model is integrated with MixUp for data augmentation to generate a diverse and sufficiently large sample set. Next, a transformer architecture using a convolutional gated linear unit (GLU) is designed to effectively capture global dependencies and local feature correlations, leading to more discriminative feature representations. In addition, a new target aggregation and background separation loss is introduced, which emphasizes target sample aggregation while increasing the distance between targets and background samples to enhance separability. The HTD-DMCG method is compared against classical and state-of-the-art HTD methods on four real HSI datasets. Extensive experiments show that it can effectively outperform existing methods in target detection performance. The code is available at https://***/Li-ZK/ HTD-DMCG

关键词： Feature extraction transformers Diffusion models Object detection Hyperspectral imaging Logic gates Detectors Training Learning systems Ecology Denoising diffusion probabilistic model (DDPM) hyperspectral target detection (HTD) target intraclass aggregation transformer encoder

来源：评论

学校读者我要写书评

暂无评论

Precision forecasting of grinding wheel Wear: A TransBiGRU model for advanced industrial predictive maintenance

引用

MEASUREMENT 2024年 234卷

作者： Si, Zekai Si, Sumei Mu, Deqiang Changchun Univ Technol Coll Electromech Engn Changchun 130012 Peoples R China Changchun Univ Technol Changchun 130012 Peoples R China

In intelligent manufacturing, accurate prediction of grinding wheel wear is essential to reduce maintenance costs and improve production efficiency. To achieve accurate forecasts, this paper introduces a grinding wheel wear prediction model, TransBiGRU, incorporating a transformer encoder, Bidirectional Gated Recurrent Unit (BiGRU), positional encoding layer, and position-wise feedforward layer. The model extracts features by analyzing current signal characteristics in basic statistical, time, and frequency domains during the machining process. Training is conducted through K-fold cross-validation to ensure model stability. The experimental results indicate that the model achieved good performance with Mean Absolute Error (MAE), Root Mean Square Error (RMSE), and Coefficient of Determination (R2) evaluation metrics, obtaining values of 2.0898, 3.262, and 0.9338, respectively. Systematically reducing modules validates the importance of each module in enhancing predictive capabilities. This research achieves the practical application of low-cost current sensors in optimizing maintenance plans and reducing downtime in predicting grinding wheel wear.

关键词： Grinding wheel wear prediction transformer encoder Bidirectional Gated Recurrent Unit K-fold cross-validation

来源：评论

学校读者我要写书评

暂无评论

Are Graphs and GCNs necessary for short-term metro ridership forecasting?

引用

EXPERT SYSTEMS WITH APPLICATIONS 2024年 254卷

作者： Yang, Qiong Xu, Xianghua Wang, Zihang Yu, Juan Hu, Xiaodong Hangzhou Dianzi Univ Sch Comp Sci Hangzhou 310018 Peoples R China Zhejiang Ind Polytech Coll Dept Art & Design Shaoxing 312000 Peoples R China Zhejiang Normal Univ Sch Comp Sci & Technol Jinhua 321004 Peoples R China

Short-term metro ridership prediction is of great significance to efficient and economic operation of Urban Rail Transit (URT) systems. With the popularity of Graph Convolution Networks (GCN) and transformers, the recent notable metro ridership forecasting methods are GCN-based and transformer -based models. However, existing methods face the following drawbacks. First, GCN-based models fail to effectively capture global spatial correlations which are significant for accurate prediction. Second, transformer -based models are prone to loss temporal information due to the permutation -invariant and anti -order properties of the self -attention which they used for capturing temporal correlations. To overcome the drawbacks, we propose a novel sequence -tosequence metro ridership prediction model, named SDT-GRU, with Stacked DT-GRU layers as both encoder and decoder. The core component of our model is DT-GRU, which integrates Dual -branch transformer decoder into the GRU to effectively capture global spatial correlations and temporal correlations with transformer decoder and GRU, separately. In particular, the DT-GRU module uses one branch transformer encoder layer to capture spatial correlations within the same timestamp, and adopts another transformer encoder layer to implicitly capture spatio-temporal correlations among previous timestamps. Then, outputs of the two transformer encoder layers are fed into a GRU layer to capturing spatio-temporal patterns. To evaluate the effectiveness of the proposed SDT-GRU, we conduct comprehensive experiments on three real -world metro ridership datasets from Beijing, Shanghai and Hangzhou. Experimental results demonstrate that our SDT-GRU achieves better prediction performance than the state-of-the-art baselines.

关键词： Metro ridership prediction transformer encoder GRU encoder-decoder architecture Graph convolutional networks (GCNs)

来源：评论

学校读者我要写书评

暂无评论

transformer-Based Feature Compensation Network for Aerial Photography Person and Ground Object Recognition

引用

REMOTE SENSING 2024年第2期16卷 268-268页

作者： Zhang, Guoqing Zheng, Chen Ye, Zhonglin Nanjing Univ Informat Sci & Technol Sch Comp Sci Nanjing 210044 Peoples R China Nanjing Univ Informat Sci & Technol Engn Res Ctr Digital Forens Minist Educ Nanjing 210044 Peoples R China Nanjing Univ Informat Sci & Technol Jiangsu Collaborat Innovat Ctr Atmospher Environm Nanjing 210044 Peoples R China Qinghai Normal Univ State Key Lab Tibetan Intelligent Informat Proc & Xining 810008 Peoples R China

Visible-infrared person re-identification (VI-ReID) aims at matching pedestrian images with the same identity between different modalities. Existing methods ignore the problems of detailed information loss and the difficulty in capturing global features during the feature extraction process. To solve these issues, we propose a transformer-based Feature Compensation Network (TFCNet). Firstly, we design a Hierarchical Feature Aggregation (HFA) module, which recursively aggregates the hierarchical features to help the model preserve detailed information. Secondly, we design the Global Feature Compensation (GFC) module, which exploits transformer's ability to capture long-range dependencies in sequences to extract global features. Extensive results show that the rank-1/mAP of our method on the SYSU-MM01 and RegDB datasets reaches 60.87%/58.87% and 91.02%/75.06%, respectively, which is better than most existing excellent methods. Meanwhile, to demonstrate our method's transferability, we also conduct related experiments on two aerial photography datasets.

关键词： person re-identification aerial photography remote sensing transformer encoder semantic information

来源：评论

学校读者我要写书评

暂无评论

TARA-Net: A Fusion Network for Detecting Takeaway Rider Accidents

引用

ACM TRANSACTIONS ON INTELLIGENT SYSTEMS AND TECHNOLOGY 2021年第6期12卷 1–19页

作者： He, Yifan Li, Zhao Fu, Lei Wang, Anhui Zhang, Peng Zhou, Shuigeng Zhang, Ji Yu, Ting Fudan Univ 2005 Songhu Rd Shanghai 200238 Peoples R China Alibaba Grp 969 Wenyi West Rd Hangzhou 310023 Peoples R China Univ Southern Queensland West St Darling Hts Qld 4350 Australia Zhejiang Lab 1818 Wenyi West Rd Hangzhou 311121 Peoples R China

In the emerging business of food delivery, rider traffic accidents raise financial cost and social traffic burden. Although there has been much effort on traffic accident forecasting using temporal-spatial prediction models, none of the existing work studies the problem of detecting the takeaway rider accidents based on food delivery trajectory data. In this article, we aim to detect whether a takeaway rider meets an accident on a certain time period based on trajectories of food delivery and riders' contextual information. The food delivery data has a heterogeneous information structure and carries contextual information such as weather and delivery history, and trajectory data are collected as a spatial-temporal sequence. In this article, we propose a TakeAway Rider Accident detection fusion network TARA-Net to jointly model these heterogeneous and spatial-temporal sequence data. We utilize the residual network to extract basic contextual information features and take advantage of a transformer encoder to capture trajectory features. These embedding features are concatenated into a pyramidal feed-forward neural network. We jointly train the above three components to combine the benefits of spatial-temporal trajectory data and sparse basic contextual data for early detecting traffic accidents. Furthermore, although traffic accidents rarely happen in food delivery, we propose a sampling mechanism to alleviate the imbalance of samples when training the model. We evaluate the model on a transportation mode classification dataset Geolife and a real-world *** dataset with over 3 million riders. The experimental results show that the proposed model is superior to the state-of-the-art.

关键词： Traffic accident detection trajectory data transformer encoder residual network deep learning

来源：评论

学校读者我要写书评

暂无评论

Complementary two-branch transformer for multi-label image retrieval

引用

PATTERN RECOGNITION 2025年 168卷

作者： Li, Ying Deng, Shuaiyu Guan, Chunming Gao, Jiaquan Nanjing Normal Univ Nanjing 210023 Peoples R China Samsung Elect China R&D Ctr Nanjing 210012 Peoples R China

Pattern images are widely utilized in daily life related domains such as home furnishings, apparel design, and artistic creation. Retrieving the required images from a large number of multi-label patterns has practical value in both the commercial and industrial fields. Currently, multi-label pattern image retrieval techniques primarily focus on learning effective features, similar to single-label image retrieval. Given the complicated similarity relationships between multiple semantics, merely extracting effective visual features may not yield satisfactory performance for multi-label pattern image retrieval. Although Graph Convolutional Networks have been used to enhance the semantic relationships among labels, they still rely on prior information from the database. To eliminate the need for additional prior information, in this paper, we introduce a complementary two-branch transformer model specifically designed for multi-label image retrieval, which is especially suitable for fine-grained patterns. In the model, we design a transformer encoder to reveal the intricate dependencies between visual features and labels. Except for the encoder, the image feature embedding module is also based on transformer, so that the optimal model weights could be learned end-to-end in a pure transformer model. To be specific, inputs of the transformer encoder mainly consist of a vision transformer branch and a label embedding branch, which generates discriminative image features and label descriptions, respectively. Extensive experiments show that our approach not only achieves competitive results on Pattern2 dataset, but also establishes the state-of-the-art performance on other multi-label datasets including NUS-WIDE, MS-COCO and VOC2007.

关键词： Image retrieval Multi-label Pattern image transformer encoder

来源：评论

学校读者我要写书评

暂无评论

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案：

请选择收藏分类：

通借通还

建议与咨询 留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案： 新增检索档案 确定 取消

请选择收藏分类： 新增自定义分类 确定 取消

通借通还

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

请选择保存的检索档案：

请选择收藏分类：