检索结果-内蒙古大学图书馆

All-in-one Weather-degraded Image Restoration via Adaptive Degradation-aware Self-prompting Model

IEEE Transactions on Multimedia 2025年 27卷 3343-3355页

作者： Wen, Yuanbo Gao, Tao Li, Ziqi Zhang, Jing Zhang, Kaihao Chen, Ting Chang'an University School of Information Engineering Xi'an China Chang'an University School of Data Science and Artificial Intelligence Xi'an China Australian National University School of Computing CanberraACT Australia Harbin Institute of Technology School of Computer Science and Technology Shenzhen China

Existing approaches for all-in-one weather-degraded image restoration suffer from inefficiencies in leveraging degradation-aware priors, resulting in sub-optimal performance in adapting to different weather conditions. To this end, we develop an adaptive degradation-aware self-prompting model (ADSM) for all-in-one weather-degraded image restoration. Specifically, our model employs the contrastive language-image pre-training model (CLIP) to facilitate the training of our proposed latent prompt generators (LPGs), which represent three types of latent prompts to characterize the degradation type, degradation property and image caption. Moreover, we integrate the acquired degradation-aware prompts into the time embedding of diffusion model to improve degradation perception. Meanwhile, we employ the latent caption prompt to guide the reverse sampling process using the cross-attention mechanism, thereby guiding the accurate image reconstruction. Furthermore, to accelerate the reverse sampling procedure of diffusion model and address the limitations of frequency perception, we introduce a wavelet-oriented noise estimating network (WNE-Net). Extensive experiments conducted on eight publicly available datasets demonstrate the effectiveness of our proposed approach in both task-specific and all-in-one applications. © 1999-2012 IEEE.

关键词： Restoration

来源：评论

学校读者我要写书评

暂无评论

Adaptive Edge Caching in mmWave Integrated Access and Backhaul Networks 11

Adaptive Edge Caching in mmWave Integrated Access and Backha...

引用

11th International Symposium on Telecommunication, IST 2024

作者： Rashidi, Zahra Nazarifard, Fatemeh Sadat Hashemi Hakami, Vesal Iran University of Science and Technology School of Computer Engineering Tehran Iran Iran University of Science and Technology Center of Excellence in Future Networks School of Computer Engineering Tehran Iran

ISBN: (纸本)9798350356250

Caching popular files at the small base stations has proved to be an effective strategy for reducing the content delivery delay in cellular networks and alleviating backhaul congestion. The challenging characteristics of radio propagation, the use of highly directional transmission in future-generation cellular networks, and the popularity of content lead to more complex and critical problems in content placement. In this paper, we propose a mathematical formulation for the centralized optimization of content placement at integrated access and backhaul (IAB) nodes to minimize the average content delivery latency in millimeter wave (mmWave) IAB cellular communications. We consider the dynamics of links and the time-varying popularity of contents as Markov decision processes (MDP) and propose a deep reinforcement learning framework to obtain a solution with low computation complexity. Simulation experiments are conducted to investigate the effectiveness of the proposed learning algorithm as well as to compare it against some schemes with different levels of adaptation to the system dynamics. © 2024 IEEE.

关键词： Markov processes

来源：评论

学校读者我要写书评

暂无评论

GENOCARE PROGNOSTICATOR MODEL: HOST GENETICS PREDICT SEVERITY OF INFECTIOUS DISEASE

引用

Scalable Computing 2025年第2期26卷 924-939页

作者： DUBEY, SHIVENDRA SINGH, SHWETA VERMA, DINESH KUMAR LODHI, SUDHEER KUMAR DUBEY, SAKSHI Department of Artificial Intelligence and Machine Learning Manipal University Rajasthan Jaipur303007 India School of Engineering and Technology Jagran Lakecity University Madhya Pradesh Bhopal462042 India Department of Computer Science and Engineering Jaypee University of Engineering and Technology Madhya Pradesh Guna473226 India Department of Computer Science and Engineering Parul Institute of Engineering and Technology Gujarat Vadodara391760 India

Scientific community understanding of the variance in severity of infectious disease like COVID-19 across patients is an important area of focus. The article presents an innovative voting ensemble GenoCare Prognosticator (GCP) model that incorporates XGBoost and Random Forest classifiers, two cutting-edge machine learning approaches. A large dataset that incorporates medical covariates like gender and age along with biological WES (Whole Exome Sequencing) data was used to train these models. Five-fold stratified cross-validation was used to process the dataset in order to improve model stability and avoid overfitting. Two medical covariates and sixteen recognized candidate gene variants were among the eighteen major features on which our GCP model had been verified using data from earlier studies. Specific post-hoc clarification of the model's predictions was provided by ExplainerDashboard, a Python open-source library, to improve interpretability. Furthermore, we utilized OpenTarget and Enrichr, two bioinformatic resources, to establish connections between the discovered variations in genetics and pertinent ontologies, biological pathways, and possible drug/disease relationships. Unsupervised clustering of SHAP key feature values was included in the analysis, which revealed intricate genetic interactions that affect the severity of the disease. Our results show that although gender and age are the main factors influencing the severity of COVID-19, complex genetic interactions cause severe symptoms in a specific subset of patients. This work contributes to our comprehension of the biological variables influencing the severity of COVID-19 and offers a reliable, comprehensible model that can help recognize patients at high risk and guide individualized treatment plans. © (2025), (West University of Timisoara). All rights reserved.

关键词： COVID-19

来源：评论

学校读者我要写书评

暂无评论

Prediction of railway track irregularity based on TCN-BiLSTMs-Attention model 24

Prediction of railway track irregularity based on TCN-BiLSTM...

引用

3rd International Conference on Signal Processing, computer Networks and Communications, SPCNC 2024

作者： Wu, Chen Lu, Xiaofeng Wang, Yiyun Pang, Tiantian School of Computer Science and Engineering Xi’an University of Technology Shaanxi Xi’an China

ISBN: (纸本)9798400710834

Track irregularities can significantly reduce the comfort and safety of train operation. If the development trend of track irregularities can be predicted, the railway management department can issue early warnings to ensure the safe and smooth operation of trains. The sequence data associated with track irregularities exhibit complex nonlinear and non-stationary characteristics, and predictions based solely on the track quality index (TQI) do not take into account the influence of other factors. To address this issue, this paper proposes a model based on TCN-BiLSTMs-Attention for predicting the development trends of track irregularities. First, the Time Convolutional Network (TCN) is employed to extract features from the preprocessed multivariate input data of track irregularities, so as to capture the impact of various indicators on TQI values and provide richer inputs for subsequent model stages. The data is then fed into a multi-layer bidirectional long short-term memory network (BiLSTM) for training. The stacked BiLSTM layers utilize the non-linear changes and long-term dependencies of TQI sequence data that LSTM networks can learn, as well as the bidirectional LSTM, to simultaneously process past and future information. This increases the depth of the model, further improving its learning ability. Subsequently, the attention mechanism module is used for information extraction, enhancing the model’s sensitivity to key information and its ability to capture long-distance dependencies. Finally, the prediction results are output through a fully connected layer. Comparative experiments with multiple models have shown that the proposed TCN-BiLSTMs-Attention model yields more accurate predictions and has strong robustness. © 2024 Copyright held by the owner/author(s).

关键词： Railroads

来源：评论

学校读者我要写书评

暂无评论

Automatic Positioning of Intelligent Mobile Robot Based on Improved Front-End Scanning Matching

Journal of Network Intelligence

引用

Journal of Network Intelligence 2025年第1期10卷 559-572页

作者： Wang, Hong Wang, Yi-Gui Zhang, Jie Department of Computer and Software Engineering Shandong College of Electronic Technology Jinan250200 China School of Computer Science and Technology Shandong Jianzhu University Jinan250101 China College of Agricultural Science and Technology Yana Royal Polytechnic University Chiang Mai50000 Thailand

Traditional autonomous navigation methods for mobile robots mainly rely on geometric feature-based LiDAR scan-matching algorithms, but in complex environments, this method is often affected due to the presence of moving objects, occlusions, and other interfering factors, resulting in a decrease in positioning accuracy. With the growing demand for robotics applications in logistics, security, exploration and other fields, the need for robust autonomous navigation and high-precision mapping in highly dynamic and complex environments is becoming more and more urgent. To solve this problem, an improved front-end scan matching algorithm based on Frequency Modulated Continuous Wave (FMCW) LiDAR is proposed in this paper. Firstly, by using the target Doppler velocity information provided by the FMCW LiDAR, we design a novel point cloud segmentation algorithm based on velocity clustering, which is able to effectively distinguish between stationary and moving objects, and avoid the dynamic interference affecting the position estimation. Secondly, we introduce the Gaussian Mixture Model Sampling Consistency (GMMSC) algorithm, which is more robust to reject the mis-matched pairs in the scanning matching process and improve the alignment accuracy. Finally, based on the residual high-quality matched pairs, we combine the classical ICP algorithm with the robust kernel function to further enhance the stability of the position estimation in the case of occlusion and local mismatch. The experimental results show that the proposed improved algorithm significantly improves the mapping capability of mobile robots in complex environments compared with the existing techniques. The average relative error of the improved front-end scanning matching algorithm is reduced by 40.6 % compared with the pre-improved one. © 2025, Taiwan Ubiquitous Information CO LTD. All rights reserved.

关键词： Mobile robots

来源：评论

学校读者我要写书评

暂无评论

Joint Mode Selection and Beamforming Designs for Hybrid-RIS Assisted ISAC Systems

引用

IEEE Wireless Communications Letters 2025年第6期14卷 1718-1722页

作者： Lin, Yingbin Wang, Feng Zhang, Xiao Han, Guojun Lau, Vincent K. N. Guangdong University of Technology School of Information Engineering Guangzhou510006 China South-Central Minzu University College of Computer Science Wuhan430079 China The Hong Kong University of Science and Technology Department of Electronic and Computer Engineering Hong Kong Hong Kong

This letter considers a hybrid reconfigurable intelligent surface (RIS) assisted integrated sensing and communication (ISAC) system, where each RIS element can flexibly switch between the active and passive modes. Subject to the signal-to-interference-plus-noise ratio (SINR) constraint for each communication user (CU) and the transmit power constraints for both the base station (BS) and the active RIS elements, with the objective of maximizing the minimum beampattern gain among multiple targets, we jointly optimize the BS transmit beamforming for ISAC and the mode selection of each RIS reflecting element, as well as the RIS reflection coefficient matrix. Such formulated joint hybrid-RIS assisted ISAC design problem is a mixed-integer nonlinear program, which is decomposed into two low-dimensional subproblems being solved in an alternating manner. Specifically, by using the semidefinite relaxation (SDR) technique along with the rank-one beamforming construction process, we efficiently obtain the optimal ISAC transmit beamforming design at the BS. Via the SDR and successive convex approximation (SCA) techniques, we jointly determine the active/passive mode selection and reflection coefficient for each RIS element. Numerical results demonstrate that the proposed design solution is significantly superior to the existing baseline solutions. © 2012 IEEE.

关键词： Beamforming

来源：评论

学校读者我要写书评

暂无评论

Emotion-oriented Cross-modal Prompting and Alignment for Human-centric Emotional Video Captioning

引用

IEEE Transactions on Multimedia 2025年 27卷 3766-3780页

作者： Wang, Yu Liu, Yuanyuan Zhou, Shunping Huang, Yuxuan Tang, Chang Zhou, Wujie Chen, Zhe China University of Geosciences School of Computer Science School of Geography and Information Engineering Wuhan430074 China China University of Geosciences School of Computer Science Wuhan430074 China Zhejiang University of Science and Technology School of Information and Electronic Engineering Hangzhou310018 China La Trobe University Cisco-La Trobe Centre for AI and IoT School of Computing Engineering and Mathematical Sciences BundooraVIC3086 Australia

Human-centric Emotional Video Captioning (H-EVC) aims to generate fine-grained, emotion-related sentences for human-based videos, enhancing the understanding of human emotions and facilitating human-computer emotional interaction. However, existing video captioning methods primarily focus on overall event content, often overlooking sufficient subtle emotional clues and interactions in videos. As a result, the generated captions frequently lack emotional information. To address this, we propose a novel Emotion-oriented Cross-modal Prompting and Alignment (ECPA) approach for large foundation models to enhance H-EVC accuracy by effectively modeling fine-grained visual-textual emotion clues and interactions. Using large foundation models, our ECPA introduces two learnable prompting strategies: visual emotion prompting (VEP) and textual emotion prompting (TEP), as well as an emotion-oriented cross-modal alignment (ECA) module. In VEP, we develop two-level learnable visual prompts, i.e., emotion recognition (ER)-level and action unit (AU)-level prompting, to assist pre-trained vision-language foundation models to attend to both coarse and fine emotion-related visual information in videos. In TEP, we correspondingly devise two-level learnable textual prompts, i.e., sentence-level emotional tokens, and word-level masked tokens, for obtaining both whole and local textual prompt representations related to emotions. To further facilitate the interaction and alignment of visual-textual emotion prompt representations, our ECA introduces another two levels of emotion-oriented prompt alignment learning mechanisms: the ER-sentence level and the AU-word level alignment losses. Both enhance the model's ability to capture and integrate both global and local cross-modal emotion semantics, thereby enabling the generation of fine-grained emotional linguistic descriptions in video captioning. Extensive experiments not only demonstrate that our ECPA outperforms existing state-of-the-art ap

关键词： Semantics

来源：评论

学校读者我要写书评

暂无评论

Revolutionizing wildlife protection: a novel approach combining deep learning and night-time surveillance

引用

Multimedia Tools and Applications 2024年 1-35页

作者： Madhasu, Nithya Pande, Sagar Dhanraj VIT-AP University Andhra Pradesh India Department of Computer Science and Engineering School of Engineering and Technology Pimpri Chinchwad University Maharashtra Pune India

The increasing instances of animals encroaching on human settlements, as well as the illicit trafficking of wildlife, have prompted immediate actions to protect the natural heritage. In addition to this, the difficulties of night-time animal surveillance are also being faced. This paper highlights the urgent necessity for comprehensive animal welfare monitoring and effective anti-illegal trafficking prevention. The expensive cost of installing night vision cameras heightens the necessity of locating an affordable and effective solution. To overcome these issues, a unique strategy, combining colorization utilizing Customised Conditional GAN (Generative Adversarial Net) for night-time applications and YOLO-CNAS (You Only Look Once-Customised Neural Architecture Search) classification for intelligent wildlife detection is proposed. Colorization during the night enables the discovery of critical nocturnal behaviors, encouraging a greater knowledge and relationship with nature. Using these powerful deep learning algorithms helps to recognize the wildlife throughout both daylight and night-time hours, as well as smugglers trespassing into the forest. The suggested method obtains an amazing testing accuracy rate that has improved from 55.73% to 72.54% and then finally to 94.67%, demonstrating the revolutionary approach's potential for animal protection. The extensive dataset used for training and assessment was meticulously sourced from various websites including iNaturalist, Unsplash, and Pexels. These platforms provided a rich array of diverse images. During the pressing need to protect the nation's irreplaceable heritage, the intelligent model provides a ray of hope. Conservationists and policymakers may work together to safeguard the natural heritage by enabling effective wildlife surveillance and maintaining a happy coexistence between humans and animals for future generations. © The Author(s), under exclusive licence to Springer science+Business Media, LLC, part of S

关键词： Animals

来源：评论

学校读者我要写书评

暂无评论

Music source separation via hybrid waveform and spectrogram based generative adversarial network

引用

Multimedia Tools and Applications 2024年 1-15页

作者： Wu, Qiuxia Deng, Haipeng Hu, Kun Wang, Zhiyong School of Software Engineering South China University of Technology Guangdong Province Guangzhou China School of Computer Science The University of Sydney SydneyNSW Australia

Music source separation aims to disentangle individual sources from the mixture of musical signals. Existing generative adversarial network (GAN) based methods generally work on the spectrogram domain only. However, this practice ignores the patterns from the waveform domain, which are more informative for modelling some categories of sources. In this paper, we propose a fully hybrid GAN framework to integrate knowledge from both domains. In particular, the generator formulates acoustical patterns from waveform and spectrogram domains, while the discriminator provides discriminative information based on the local patch-level spectrograms such that the generator can produce more plausible separation results. Furthermore, to enhance the quality of estimated sources, we devise a perceptual spectrogram loss term, which is a complement of the waveform-level loss. The proposed method is evaluated on two widely used music source separation datasets, producing music sources of high signal-to-distortion ratio (12.03 in MIR-1K dataset and 8.08 in MUSDB18 dataset). These results demonstrate the superiority of the proposed method compared with the state-of-the-art methods. © The Author(s), under exclusive licence to Springer science+Business Media, LLC, part of Springer Nature 2024.

关键词： Generative adversarial networks

来源：评论

学校读者我要写书评

暂无评论

An Enhanced Linearly Homomorphic Network Coding Signature Scheme for Secure Data Delivery in IoT Networks

引用

IEEE Transactions on Information Forensics and Security 2025年 20卷 5534-5548页

作者： Huang, Hao Wang, Xiaofen Au, Man Ho Cao, Sheng Zhao, Qinglin Yu, Jiguo University of Electronic Science and Technology of China Chengdu611731 China The Hong Kong Polytechnic University Department of Computing Hong Kong Hong Kong Macau University of Science and Technology School of Computer Science and Engineering China

Recently, Li et al. proposed an identity-based linearly homomorphic network coding signature (IB-HNCS) scheme for secure data delivery in Internet of Things (IoT) networks, and they claimed that the IB-HNCS scheme can resist pollution attacks. However, this paper shows that the IB-HNCS scheme is vulnerable to pollution attacks, as anyone who only has the public parameter can forge a new file identifier or a valid signature on a corrupted data packet to pollute legitimate sensor data. To enhance security and performance in network coding-based IoT networks, we propose a secure and efficient certificateless linearly homomorphic network coding signature scheme for IoT data delivery, which is free of burdensome certificate management and key escrow issue. In addition, our scheme is proved to be secure against adaptive chosen identity and adaptive chosen subspace attacks under two types of adversaries in the algebraic group model and random oracle model. Therefore, our scheme can verify the validity of data packets and allow data packets to be computed, so as to resist pollution attacks. The performance evaluation demonstrates that our scheme is more efficient and practical than existing secure schemes. Specifically, for a 73-dimensional data vector, the costs of signature generation and verification in our scheme are reduced by 38.588%-86.076% and 38.570%-85.664% respectively under the symmetric bilinear pairing setting, and the costs of signature generation and verification in our scheme are reduced by 17.740%-49.752% and 29.697%-58.645% respectively under the asymmetric bilinear pairing setting. © 2005-2012 IEEE.

关键词： Internet of things

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：