检索结果-内蒙古大学图书馆

arXiv 2025年

作者： Chen, Ziyang Zhang, Yongjun Li, Wenting Wang, Bingshu Wu, Yabo Zhao, Yong Chen, C.L. Philip College of Computer Science The State Key Laboratory of Public Big Data Guizhou University Guiyang550025 China School of Information Engineering Guizhou University of Commerce Guiyang550021 China School of Software Northwestern Polytechnical University Xi’an710129 China Key Laboratory of Integrated Microsystems Shenzhen Graduate School Peking University Shenzhen518055 China School of Computer Science and Engineering South China University of Technology Guangzhou510641 China

In light of the advancements in transformer technology, extant research posits the construction of stereo transformers as a potential solution to the binocular stereo matching challenge. However, constrained by the low-rank bottleneck and quadratic complexity of attention mechanisms, stereo transformers still fail to demonstrate sufficient nonlinear expressiveness within a reasonable inference time. The lack of focus on key homonymous points renders the representations of such methods vulnerable to challenging conditions, including reflections and weak textures. Furthermore, a slow computing speed is not conducive to the application. To overcome these difficulties, we present the Hadamard Attention Recurrent Stereo Transformer (HART) that incorporates the following components: 1) For faster inference, we present a Hadamard product paradigm for the attention mechanism, achieving linear computational complexity. 2) We designed a Dense Attention Kernel (DAK) to amplify the differences between relevant and irrelevant feature responses. This allows HART to focus on important details. DAK also converts zero elements to non-zero elements to mitigate the reduced expressiveness caused by the low-rank bottleneck. 3) To compensate for the spatial and channel interaction missing in the Hadamard product, we propose MKOI to capture both global and local information through the interleaving of large and small kernel convolutions. Experimental results demonstrate the effectiveness of our HART. In reflective area, HART ranked 1st on the KITTI 2012 benchmark among all published methods at the time of submission. Code is available at https://***/ ZYangChen/HART. Copyright © 2025, The Authors. All rights reserved.

关键词： Hadamard transforms

来源：评论

学校读者我要写书评

暂无评论

A novel high-fidelity reversible data hiding scheme based on multi-classification pixel value ordering

引用

Journal of Visual Communication and Image Representation 2025年 110卷

作者： Cui, Chen Li, Li Lu, Jianfeng Zhang, Shanqing Chang, Chin-Chen College of Computer Science and Technology Hangzhou Dianzi University Hangzhou310018 China Key Laboratory of Public Security Information Application Based on Big-Data Architecture Zhejiang Police College Hangzhou310018 China Department of Information Engineering and Computer Science Feng Chia University Taichung407 Taiwan

Pixel value ordering (PVO) is a highly effective technique that employs a pixel block partitioning and sorting for reversible data hiding (RDH). However, its embedding performance is significantly impacted by block size. To address this, an improved pixel-based PVO (IPPVO) was developed adopting a per-pixel approach and adaptive context size. Nevertheless, IPPVO only considers pixels below and to the right for prediction, neglecting other closer neighboring regions, leading to inaccurate predictions. This study presents a RDH strategy using multi-classification embedding to enhance performance. First, pixels are categorized into four classes based on parity coordinates, obtaining higher correlation prediction values using an adaptive nearest neighbor content size. Second, a new complexity calculation method is introduced, the complexity frequency of pixel regions to better differentiate between complex and flat regions. Finally, an effective embedding ratio and index value constraint are introduced to mitigate the challenge of excessive distortion when embedding large capacities. Experimental results indicate that the proposed scheme offers superior embedding capacity with low distortion compared to state-of-the-art PVO-based RDH methods. © 2025 Elsevier Inc.

关键词： Adaptive modification High-fidelity Multi-classification RDH

来源：评论

学校读者我要写书评

暂无评论

Nucleic Acid Cascade Amplification Empowered Self-Powered Sensing Platform Based on the Confined Bioconjugate Catalysis Strong Redox Boosting Mechanism

SSRN

引用

SSRN 2025年

作者： Jin, Chenchen Wu, Qingnian Ma, Yunzhi Tan, Xue-Cai Huang, Ke-Jing Xie, Shengyu Duan, Yun Education Department of Guangxi Zhuang Autonomous Region Laboratory of Optic-electric Chemo/Biosensing and Molecular Recognition Guangxi Collaborative Innovation Center for Chemistry and Engineering of Forest Products Guangxi Key Laboratory of Chemistry and Engineering of Forest Products Key Laboratory of Chemistry and Engineering of Forest Products State Ethnic Affairs Commission School of Chemistry and Chemical Engineering Guangxi Minzu University Nanning530006 China Analysis and Test Center Chinese Academy of Tropical Agricultural Sciences Haikou571101 China School of Big Data Guangxi Vocational and Technical College Nanning530226 China

Patulin is one of the most common mycotoxins found in moldy fruits and poses a significant threat to human health. This study develops a confined bioconjugate efficiently catalysis self-powered sensing system based on CoOOH nanosheets and glucose oxidase,. In this sensing system, the spider-type Walker is activated when its aptamer is captured by the target analyte to release DNA through an equivalent DNA transducer under the action of the auxiliary factor Mg2+. This process opens a series of hairpin probes, self-assembles into a DNA framework, and activates the catalytic hairpin assembly-hybridization chain reaction cascade amplification. The construction of a DNA network structure on the electrode surface significantly increases the probe loading capacity, thereby greatly enhancing the sensitivity and selectivity of target detection. Under optimized conditions, the detection linear range of this method is from 10-14 to 10-6 mg/mL, with a detection limit of 9.05×10-15 mg/mL (S/N = 3). The application of the confined bioconjugate catalytic system, the equivalent DNA transducer constructed by the spider-type Walker, and the nucleic acid cascade amplification technology has increased the sensitivity of this method by more than 1000 times compared to other electrochemical sensing methods. Additionally, this method exhibits excellent selectivity and a broad detection range, providing a reliable solution for the on-site rapid detection of ultra-trace patulin in food. © 2025, The Authors. All rights reserved.

关键词： Glucose oxidase

来源：评论

学校读者我要写书评

暂无评论

Buckling analysis of GPL RC micro smart plates resting on elastic foundation subjected to thermal loads

引用

Archive of Applied Mechanics 2025年第6期95卷 1-23页

作者： Chen, Detong Wang, Qiyu Zhang, Zilin School of Rail Transportation Shandong Jiaotong University Jinan People’s Republic of China School of Big Data and Fundamental Sciences Shandong Institute of Petroleum and Chemical Technology Dongying People’s Republic of China Guangxi Key Laboratory of Green Chemical Materials and Safety Technology College of Petroleum and Chemical Engineering Beibu Gulf University Qinzhou People’s Republic of China

This study investigates the thermal buckling behavior of functionally graded microplates reinforced with graphene, incorporating two piezoelectric layers, resting on an elastic foundation, and subjected to an externally applied voltage. An artificial neural network (ANN) is utilized to analyze this behavior. The governing equations are formulated based on the modified couple stress theory to account for microscale effects. The material properties of the graphene-reinforced composite layer are determined using the Halpin–Tsai micromechanical model. The Ritz method is employed to solve the governing equations and generate the dataset for training the ANN. Specifically, the Levenberg–Marquardt algorithm is implemented within the ANN framework to significantly reduce computational costs in the buckling analysis. The input parameters include nanofiller dimensions, weight fraction, and piezoelectric layer thickness, while the output is the thermal buckling load. The results demonstrate that ANN-based predictions of the critical buckling temperature for functionally graded graphene-reinforced microplates with piezoelectric layers not only achieve high accuracy but also substantially decrease computational time compared to conventional numerical approaches. The obtained results denote that aspect ratio has the most significant impact on critical buckling temperature, with longer plates showing much greater resistance to buckling. Also, elastic foundation stiffness also plays a moderate but important role, particularly at higher stiffness values. In addition, the X pattern is the most effective in enhancing buckling resistance, but the differences between patterns are relatively small.

关键词：

来源：评论

学校读者我要写书评

暂无评论

SMS Spam Detection Using BERT and Multi-Graph Convolutional Networks

SSRN

引用

SSRN 2025年

作者： Shen, Linjie Wang, Yanbin Li, Zhao Ma, Wenrui School of Computer Science and Technology Zhejiang Gongshang University Zhejiang Hangzhou China Department of Engineering Shenzhen MSU-BIT University Guangdong Shenzhen518172 China Zhejiang Lab Zhejiang Hangzhou China Department of Big Data and Future E-Commerce Technology Zhejiang Key Laboratory Zhejiang Hangzhou China

The surge in smartphone usage has significantly increased SMS traffic and, consequently, SMS spam, posing risks such as phishing, financial losses, and privacy breaches. Traditional rule-based and blacklist methods fail against evolving spamming techniques, prompting the adoption of machine learning and deep learning approaches. However, models like CNNs and RNNs struggle to capture global co-occurrence patterns and complex semantics, while transformer-based models like BERT lack explicit syntactic and co-occurrence modeling. To address these limitations, we propose the BERT with Triple-Graph Convolutional Networks (BERT-G3CN) model, the first framework to integrate BERT word embeddings with graph embeddings from Cooccurrence, Heterogeneous, and Integrated Syntactic Graphs. This multi-graph approach captures diverse features and models both global and local structures using tailored Graph Convolutional Networks (GCNs). Experiments on two benchmark datasets demonstrate that BERT-G3CN achieves superior accuracy of 99.28% and 93.78%. © 2025, The Authors. All rights reserved.

关键词： Graph embeddings

来源：评论

学校读者我要写书评

暂无评论

Optimized learning pipeline for predicting future dissolved oxygen levels in aquaculture using time-series data

引用

Computers and Electronics in Agriculture 2025年 236卷

作者： Wang, Mingyan Hassan, Shahbaz Gul Sohel, Ferdous Liu, Tonglai He, Min Gao, Xuekai Xie, Caijin Deng, Xiwen Liu, Shuangyin Xu, Longqin Guangzhou Key Laboratory of Agricultural Product Quality Safety Traceability Information Technology Zhongkai University of Agriculture and Engineering Guangzhou510225 China Smart Agriculture Engineering Technology Research Center of Guangdong Higher Education Institutes Zhongkai University of Agriculture and Engineering Guangzhou510225 China Guangdong Provincial Agricultural Products Safety Big Data Engineering Technology Research Center Zhongkai University of Agriculture and Engineering Guangzhou510225 China School of Information Technology Murdoch University MurdochWA6150 Australia

To achieve long-term accurate predictions of dissolved oxygen in aquaculture environments. In this paper, we propose the Time varying filter based empirical mode decomposition (TVFEMD) - Permutation Entropy (PE) -Temporal Convolutional Networks (TCN) - bidirectional gated recurrent units (bigRU)-Improved Sparrow Search Algorithm (ISSA) (TPBI) model. First, dissolved oxygen data are applied to the Time-Variant Filtered Empirical Mode Decomposition (TVFEMD) model to remove noise factors in the data. Permutation Entropy (PE) is then applied to reconstruct the data and reduce its complexity. Additionally, Temporal Convolutional Networks (TCN) and Bidirectional Gated Recurrent Units (bigRU) are combined to extract features from the denoised data, improving the model's learning efficiency and prediction accuracy. Based on this, the Dynamic Opposite Learning Strategy Improved Sparrow Search Algorithm (ISSA) is introduced to optimize hyperparameters such as batch size and the number of hidden layer units. The framework was applied to predict dissolved oxygen data from an aquaculture farm in Guangdong Province, with future 1-step, 3-step, and 6-step prediction experiments. The experimental results show that the proposed model outperforms the comparison models in predicting dissolved oxygen, particularly excelling in long-term predictions (6 steps). In terms of Mean Absolute Error (MAE), compared to models such as RNN, bigRU, CEEMDAN-PE-TCN-bigRU-ISSA, and TVFEMD-PE-TCN-bigRU-SSA, the proposed model improved dissolved oxygen prediction by 50%, 55.2%, 50%, and 27.7%, respectively. Ablation experiments were conducted to verify the effectiveness of all components. In terms of Mean Absolute Error (MAE) and Root Mean Square Error (RMSE), dissolved oxygen prediction improved by 41.3% and 21.0%, respectively. The outstanding performance of this framework in long-term predictions of dissolved oxygen provides effective support for precise environmental control and early warning in aqu

关键词： Deep learning

来源：评论

学校读者我要写书评

暂无评论

Jointly Optimizing the Energy and Time for Multi-UAV 3-D Coverage of Terrestrial Regions

引用

IEEE Transactions on Mobile Computing 2025年

作者： Gong, Hao Huang, Baoqi Jia, Bing Hao, Lifei Shi, Zhenwei Inner Mongolia University Inner Mongolia Key Laboratory of Wireless Networking and Mobile Computing Engineering Research Center of Ecological Big Data Ministry of Education College of Computer Science Hohhot010021 China Inner Mongolia University College of Computer Science Hohhot010021 China Beihang University Image Processing Center School of Astronautics Beijing100191 China

Multi-rotor unmanned aerial vehicles (UAVs) have been widely employed in various sensing tasks, e.g., environmental monitoring and disaster rescuing, many of which often require full coverage of terrestrial regions by UAVs. Efforts have been devoted to minimizing one of two objectives, i.e., energy consumptions and time costs of UAVs fulfilling such tasks, whereas it is still challenging to jointly optimize both objectives due to their complicated interdependent relationship. Therefore, this paper deals with the tasks of sensing terrestrial regions with multiple UAVs, and focuses on the three-dimensional (3-D) coverage problem by formulating a multi-objective optimization problem of jointly minimizing both objectives. Specifically, in order to optimize energy consumption effectively, an advanced closed-form energy consumption model for multi-rotor UAVs is developed based on a rigorous theoretical analysis by introducing the influences of torque and acceleration, which are often ignored by existing heuristic models. Moreover, considering the NP-hardness of the problem, an innovative swarm intelligence optimization framework is established by leveraging a multitasking learning pattern to exploit cross-task knowledge transfer and adopting an improved multi-objective salp swarm algorithm. Therein, two novel operators, i.e., a variable characteristic-guided hybrid solution initialization operator and a large-scale search-space-oriented multi-mechanism solution update operator, are designed to handle continuous, discrete and even high-dimensional variables involved. Real-world experiments validate the proposed energy model due to the reduction of power consumption estimation error by up to 59% compared to baselines, and besides, extensive simulations demonstrate that the proposed algorithm significantly outperforms the benchmarks in terms of both energy consumptions and time costs. © 2002-2012 IEEE.

关键词： Energy utilization

来源：评论

学校读者我要写书评

暂无评论

Detecting Difference Between Coefficients in Linear Model Using Jackknife Empirical Likelihood

引用

Journal of Systems Science & Complexity 2016年第2期29卷 542-556页

作者： WU Xinqi ZHANG Qingzhao ZHANG Sanguo School of Mathematical Sciences University of Chinese Academy of Sciences Key Laboratory of Big Data Mining and Knowledge Management Chinese Academy of Sciences

Empirical likelihood has been found very useful in many different occasions. It usually runs into serious computational difficulties while jackknife empirical likelihood （JEL） is shown to be effective when applied to some complicated statistics. In this paper, to test the difference between coefficients of two linear regression models, the authors apply JEL to construct the confidence regions. Based on the 3EL ratio test, a version of Wilks＇ theorem is developed. Furthermore, to improve the coverage accuracy of confidence regions, a Bartlett correction is applied. Simulation studies are carried out to show the effectiveness of the proposed method in aspects of coverage accuracy. A real data set is analyzed with the proposed method as an example.

关键词： Bartlett correction coverage accuracy Jackknife empirical likelihood linear regression model.

来源：评论

学校读者我要写书评

暂无评论

Unsupervised learning for lake underwater vegetation classification: Constructing high-precision, large-scale aquatic ecological datasets

引用

Science of the Total Environment 2025年 958卷 177895页

作者： Liu, Lei Bao, Zhengsen Liang, Ying Deng, Huanxi Zhang, Xiaolin Cao, Te Zhou, Chichun Zhang, Zhenyu School of Engineering Dali University Yunnan 671003 China National Observation and Research Station of Erhai Lake Ecosystem in Yunnan Dali671006 China Air-Space-Ground Integrated Intelligence and Big Data Application Engineering Research Center of Yunnan Provincial Department of Education Yunnan 671003 China State Key Laboratory of Freshwater Ecology and Biotechnology Institute of Hydrobiology Chinese Academy of Sciences Wuhan430072 China

Monitoring underwater vegetation is vital for evaluating lake ecosystem health. Automated data collection and analysis play key roles in achieving large-scale, high-precision, and high-frequency monitoring. While technologies such as unmanned vessels have made data collection more efficient, challenges persist in the analysis process, particularly in addressing the varied needs of different lake environments. Supervised AI methods can automatically identify underwater vegetation but are heavily reliant on labeled datasets. In practice, models trained on public datasets often struggle with generalization due to differences in vegetation types, collection environments, and equipment, resulting in discrepancies between training and testing datasets. Moreover, traditional dataset construction methods that rely on manual annotation are time-consuming and costly, limiting their scalability and application. This study aims to overcome these challenges by proposing an unsupervised method for automatically classifying underwater vegetation data, aiming to reduce manual annotation efforts and construct unbiased datasets at lower costs with greater efficiency. Compared with existing unsupervised, self-supervised, and unsupervised domain adaptation methods, this method introduces two key innovations: 1) a two-step dimensionality reduction method that combines pre-trained model and manifold learning to extract key features and 2) a multialgorithm voting mechanism to increase classification confidence. These features enable high-accuracy classification without prior data annotation. Experiments show 97.32 % accuracy on public dataset and 92.43 % and 96.15 % accuracy on private datasets from Erhai Lake and Wuhan East Lake, respectively, surpassing supervised methods and matching manual classification. Additionally, it drastically reduces the annotation effort, requiring only approximately 20 labeled images to classify thousands of points. By integrating unmanned vessel technology,

关键词： Lake ecological monitoring Species classification Unbiased datasets Underwater vegetation Unsupervised methods

来源：评论

学校读者我要写书评

暂无评论

EPSCL: zero-shot segmentation with embedded pre-trained model and supervised contrastive learning on NIR-II vessel images

EPSCL: zero-shot segmentation with embedded pre-trained mode...

引用

Medical Imaging 2025: Image Processing

作者： Han, keyi Xiao, Anqi Li, Chunzhao Fu, Lidan Tian, Jie Hu, Zhenhua CAS Key Laboratory of Molecular Imaging Institute of Automation Chinese Academy of Sciences Beijing100190 China School of Artificial Intelligence University of Chinese Academy of Sciences Beijing100049 China Department of Neurosurgery Beijing Tiantan Hospital Capital Medical University Beijing100070 China China National Clinical Research Center for Neurological Diseases Beijing Tiantan Hospital Beijing100070 China Beijing Advanced Innovation Center for Big Data-based Precision Medicine School of Engineering Medicine Beihang University Beijing100191 China Engineering Research Center of Molecular and Neuro Imaging of Ministry of Education School of Life Science and Technology Xidian University Xi'an710071 China National Key Laboratory of Kidney Diseases Beijing100853 China

ISBN: (纸本)9781510685901

Blood vessel analysis is significant in many clinical domains, such as laryngology, ophthalmology, and neurosurgery. The second near-infrared window (NIR-II) fluorescence imaging is a promising technique in providing high-resolution and high-contrast blood vessel images, with improved depth penetration for better visualization of deeper blood vessels and structures. However, NIR-II vessel segmentation task is a zero-shot problem, since the lack of NIR-II blood vessel datasets and lots of labor to build the datasets. Moreover, existing methods face challenges in segmenting NIR-II fluorescence images due to the complexity introduced by the imaging modality's capacity to reveal intricate details, such as overlapping deeper vessels and mixed signals from deeper tissue organs and vessels. Therefore, we propose a zero-shot segmentation method for NIR-II fluorescence vessel images, which is an encoder-decoder framework with an embedded pre-trained model and a supervised contrastive learning strategy (EPSCL). Experiments on complex NIR-II fluorescence images show that EPSCL gets better performance and segments more vascular details compared with other popular methods. The proposed EPSCL has a great potential to serve as an effective segmentation method for NIR-II fluorescent images and assist blood vessel analysis. © 2025 SPIE.

关键词： Image segmentation

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：