检索结果-内蒙古大学图书馆

A diabetic retinopathy classification method based on image-text contrastive learning

Pattern Recognition Letters 2025年 193卷 142-148页

作者： Long, Fei Xiong, Haoren Hu, Hang Qian, Cheng Sang, Jun School of Big Data & Software Engineering Chongqing University Chongqing401331 China

In recent years, contrastive learning has gained traction in computer vision, especially for medical image analysis. Traditional supervised methods for diabetic retinopathy diagnosis require large annotated datasets, which are costly and time-consuming. This paper introduces a novel detection method using a contrastive learning framework with image-text pairs, aiming to decrease reliance on extensive annotated data while enhancing diagnostic accuracy. The proposed model includes a text encoder based on a multilayer perceptron and an image encoder that employs FastVIT with multi-scale convolution, self-attention, and spectral attention modules to capture spatial and contextual information. A binomial loss function maximizes the similarity between image and text feature vectors. In the contrastive learning phase, the model generates feature vectors from preprocessed data, utilizing the binomial loss to align them. Following this, a fully connected network fine-tunes the model during classification, ultimately outputting the diagnosis for fundus images. Experimental validation on the Kaggle APTOS dataset revealed an accuracy of 92.38% in detection tasks and 75.20% in severity grading. The model also achieved in image retrieval with a Top-5 accuracy of 72.7%, underscoring the effectiveness of the contrastive learning approach. © 2025 Elsevier B.V.

关键词： Deep learning

来源：评论

学校读者我要写书评

暂无评论

Aligning Instance-Semantic Sparse Representation towards Unsupervised Object Segmentation and Shape Abstraction with Repeatable Primitives

引用

IEEE Transactions on Visualization and Computer Graphics 2025年第5期31卷 2579-2588页

作者： Li, Jiaxin Wang, Hongxing Tan, Jiawei Ou, Zhilong Yuan, Junsong Chongqing University School of Big Data and Software Engineering Chongqing401331 China State University of New York Department of Computer Science and Engineering BuffaloNY14260 United States

Understanding 3D object shapes necessitates shape representation by object parts abstracted from results of instance and semantic segmentation. Promising shape representations enable computers to interpret a shape with meaningful parts and identify their repeatability. However, supervised shape representations depend on costly annotation efforts, while current unsupervised methods work under strong semantic priors and involve multi-stage training, thereby limiting their generalization and deployment in shape reasoning and understanding. Driven by the tendency of high-dimensional semantically similar features to lie in or near low-dimensional subspaces, we introduce a one-stage, fully unsupervised framework towards semantic-aware shape representation. This framework produces joint instance segmentation, semantic segmentation, and shape abstraction through sparse representation and feature alignment of object parts in a high-dimensional space. For sparse representation, we devise a sparse latent membership pursuit method that models each object part feature as a sparse convex combination of point features at either the semantic or instance level, promoting part features in the same subspace to exhibit similar semantics. For feature alignment, we customize an attention-based strategy in the feature space to align instance- and semantic-level object part features and reconstruct the input shape using both of them, ensuring geometric reusability and semantic consistency of object parts. To firm up semantic disambiguation, we construct cascade unfrozen learning on geometric parameters of object parts. Experiments conducted on benchmark datasets confirm that our approach results in instance- and semantic-level joint segmentation and shape abstraction with repeatable primitives, providing coherent semantic interpretations of 3D object shapes across categories in a one-stage, fully unsupervised manner, without relying on annotations or heuristic semantic priors. Code will be

关键词： Semantic Segmentation

来源：评论

学校读者我要写书评

暂无评论

Successive Interference Cancellation-Enabled Timely Status Update in Linear Multi-hop Wireless Networks

引用

IEEE Transactions on Mobile Computing 2025年第6期24卷 5298-5311页

作者： Han, Xinhui Pan, Haoyuan Wang, Zhaorui Li, Jianqiang Shenzhen University College of Computer Science and Software Engineering Shenzhen China Future Network of Intelligence Institute The School of Science and Engineering Shenzhen China Shenzhen University National Engineering Laboratory for Big Data System Computing Technology The College of Computer Science and Software Engineering Shenzhen China

We investigate the timely status update in linear multi-hop wireless networks, where a source tries to deliver status update packets to a destination through a sequence of half-duplex relays. Timeliness is measured by the age of information (AoI) metric. Maintaining a low AoI at the destination typically necessitates frequent transmission of update packets from the source. However, high packet transmission frequency in multi-hop scenarios can result in mutual wireless interference at intermediate relays. Specifically, when an intermediate relay receives wireless signals of a new packet from its previous node, simultaneous transmission of an old packet by its subsequent node to the next hop may cause wireless signals to interfere at the intermediate relay, conventionally leading to packet collision. A key motivation to solve this issue is that the intermediate relay has previously received the old packet (which can thus be forwarded to the subsequent node for further relaying). Hence, successive interference cancellation (SIC) can be employed to mitigate interference of the old packet and recover the new packet. This paper designs an SIC-enabled packet relaying scheme tailored to low AoI. Initially focusing on a three-hop network, we subsequently extend our approach to general multi-hop networks. We model the multi-hop relaying scheme using a Markov chain to derive the theoretical average AoI. Theoretical and simulation results indicate that the SIC-enabled packet relaying scheme significantly reduces the average AoI compared to the non-SIC approaches, owing to an increased packet transmission frequency at the source and the effectiveness of SIC techniques at the relays. © 2002-2012 IEEE.

关键词： Markov chains

来源：评论

学校读者我要写书评

暂无评论

A sharding blockchain-based UAV system for search and rescue missions

引用

Frontiers of Computer Science 2025年第3期19卷 103-118页

作者： Xihan ZHANG Jiashuo ZHANG Jianbo GAO Libin XIA Zhi GUAN Hao HU Zhong CHEN School of Computer Science Peking UniversityBeijing 100871China Peking University Chongqing Research Institute of Big Data Chongqing 401329China National Engineering Research Center for Software Engineering Peking UniversityBeijing 100871China State Key Lab for Novel Software Technology Nanjing UniversityNanjing 210023China

Sharding is a promising technique to tackle the critical weakness of scalability in blockchain-based unmanned aerial vehicle(UAV)search and rescue(SAR)*** breaking up the blockchain network into smaller partitions called shards that run independently and in parallel,shardingbased UAV systems can support a large number of search and rescue UAVs with improved scalability,thereby enhancing the rescue ***,the lack of adaptability and interoperability still hinder the application of sharded blockchain in UAV SAR *** refers to making adjustments to the blockchain towards real-time surrounding situations,while interoperability refers to making cross-shard interactions at the mission *** address the above challenges,we propose a blockchain UAV system for SAR missions based on dynamic sharding *** from the benefits in scalability brought by sharding,our system improves adaptability by dynamically creating configurable and mission-exclusive shards,and improves interoperability by supporting calls between smart contracts that are deployed on different *** implement a prototype of our system based on Quorum,give an analysis of the improved adaptability and interoperability,and conduct experiments to evaluate the *** results show our system can achieve the above goals and overcome the weakness of blockchain-based UAV systems in SAR scenarios.

关键词： blockchain sharding unmanned aerial vehicle search and rescue blockchain interoperability

来源：评论

学校读者我要写书评

暂无评论

Retrieval-Augmented Fine-Tuning for Improving Retrieve-and-Edit Based Assertion Generation

引用

IEEE Transactions on software engineering 2025年第05期51卷 1591-1614页

作者： Li, Hongyan Sun, Weifeng Yan, Meng Xu, Ling Li, Qiang Zhang, Xiaohong Zhang, Hongyu Key Laboratory of Dependable Service Computing in Cyber Physical Society Chongqing University Ministry of Education Chongqing400044 China School of Big Data and Software Engineering Chongqing University Chongqing400044 China

Unit Testing is crucial in software development and maintenance, aiming to verify that the implemented functionality is consistent with the expected functionality. A unit test is composed of two parts: a test prefix, which drives the unit under test to a particular state, and a test assertion, which determines what the expected behavior is under that state. To reduce the effort of conducting unit tests manually, Yu et al. proposed an integrated approach (integration for short), combining information retrieval with a deep learning-based approach to generate assertions for test prefixes, and obtained promising results. In our previous work, we found that the overall performance of integration is mainly due to its success in retrieving assertions. Moreover, integration is limited to specific types of edit operations and struggles to understand the semantic differences between the retrieved focal-test (focal-test includes a test prefix and a unit under test) and the input focal-test. Based on these insights, we then proposed a retrieve-and-edit approach named EditAS to learn the assertion edit patterns to improve the effectiveness of assertion generation in our prior study. Despite being promising, we find that the effectiveness of EditAS can be further improved. Our analysis shows that: ① The editing ability of EditAS still has ample room for improvement. Its performance degrades as the edit distance between the retrieval assertion and ground truth increases. Specifically, the average accuracy of EditAS is 12.38% when the edit distance is greater than 5. ② EditAS lacks a fine-grained semantic understanding of both the retrieved focal-test and the input focal-test themselves, which leads to many inaccurate token modifications. In particular, an average of 25.57% of the incorrectly generated assertions that need to be modified are not modified, and an average of 6.45% of the assertions that match the ground truth are still modified. Thanks to pre-trained models employing

关键词： software design

来源：评论

学校读者我要写书评

暂无评论

A study of data-driven fault diagnosis and early warning systems for power batteries in new energy vehicles 24

A study of data-driven fault diagnosis and early warning sys...

引用

8th International Conference on Electronic Information Technology and Computer engineering, EITCE 2024

作者： Li, Qiang Ma, Xinqiang Yan, Meng School of Big Data and Software Engineering Chongqing University Chongqing China School of Mathematics and Artificial Intelligence Chongqing University of Arts and Sciences Chongqing China

ISBN: (纸本)9798400710094

This study addresses the prevalent issues with new energy vehicle batteries, including failure and other complications. It focuses on lithium-ion batteries in pure electric vehicles and proposes a diagnostic approach based on SDO and NARX. The study also examines the methodology employed in fault judgement and early warning. The findings offer novel insights into the diagnosis of faults in new energy vehicles and provide technical guidance for ensuring the safe operation of these vehicles. © 2024 Copyright held by the owner/author(s).

关键词： Fault diagnosis and early warning

来源：评论

学校读者我要写书评

暂无评论

A survey on cross-user federated recommendation

引用

Science China(Information Sciences) 2025年第4期68卷 7-32页

作者： Enyue YANG Yudi XIONG Wei YUAN Weike PAN Qiang YANG Zhong MING College of Computer Science and Software Engineering Shenzhen University School of Electrical Engineering and Computer Science The University of Queensland WeBank AI Lab WeBank Department of Computer Science and Engineering Hong Kong University of Science and Technology College of Big Data and Internet Shenzhen Technology University Guangdong Laboratory of Artificial Intelligence and Digital Economy (SZ)

Recommender systems are effective in mitigating information overload, yet the centralized storage of user data raises significant privacy concerns. Cross-user federated recommendation(CUFR) provides a promising distributed paradigm to address these concerns by enabling privacy-preserving recommendations directly on user devices. In this survey, we review and categorize current progress in CUFR, focusing on four key aspects: privacy, security, accuracy, and efficiency. Firstly,we conduct an in-depth privacy analysis, discuss various cases of privacy leakage, and then review recent methods for privacy protection. Secondly, we analyze security concerns and review recent methods for untargeted and targeted *** untargeted attack methods, we categorize them into data poisoning attack methods and parameter poisoning attack methods. For targeted attack methods, we categorize them into user-based methods and item-based methods. Thirdly,we provide an overview of the federated variants of some representative methods, and then review the recent methods for improving accuracy from two categories: data heterogeneity and high-order information. Fourthly, we review recent methods for improving training efficiency from two categories: client sampling and model compression. Finally, we conclude this survey and explore some potential future research topics in CUFR.

关键词： cross-user federated recommendation federated recommendation federated learning recommender systems user privacy

来源：评论

学校读者我要写书评

暂无评论

NCD: Normal-Guided Chamfer Distance Loss for Watertight Mesh Reconstruction from Unoriented Point Clouds

引用

Computer Graphics Forum 2025年第0期

作者： Li, Jiaxin Tan, Jiawei Ou, Zhilong Wang, Hongxing School of Big Data and Software Engineering Chongqing University Chongqing China

As a widely used loss function in learnable watertight mesh reconstruction from unoriented point clouds, Chamfer Distance (CD) efficiently quantifies the alignment between the sampled point cloud from the reconstructed mesh and its corresponding input point cloud. Occasionally, to enhance reconstruction fidelity, CD incorporates a normal consistency term, albeit at the cost of efficiency. In this context, normal estimation for unoriented point clouds requires computationally intensive matrix decomposition or specialized pre-trained models, whereas deriving normals for mesh-sampled points can be readily achieved using the cross product of mesh vertices. However, the reconstruction models employing CD and its variants typically rely solely on the spatial coordinates of the points, which omits normal information in favor of efficiency and deployability. To tackle this challenge, we propose a novel loss function for watertight mesh reconstruction from unoriented point clouds, termed Normal-guided Chamfer Distance (NCD). Building upon CD, NCD introduces a normal-steered weighting mechanism based on the angle between the normal at each mesh-sampled point and the vector to its corresponding input point, offering several advantages: (i) it leverages readily available mesh-sampled point normals to weight coordinate-based Euclidean distances, thus extending the capability of CD;(ii) it eliminates the need for normal estimation from input unoriented point clouds;(iii) it incurs a negligible increase in computational complexity compared to CD. We employ NCD as the training loss for point-to-mesh reconstruction with multiple models and initial watertight meshes on benchmark datasets, demonstrating its superiority over state-of-the-art CD variants. © 2025 Eurographics - The European Association for Computer Graphics and John Wiley & Sons Ltd.

关键词： Mesh generation

来源：评论

学校读者我要写书评

暂无评论

Secret Sharing Based Key Agreement Protocol for Body Area Networks 18th

Secret Sharing Based Key Agreement Protocol for Body Area N...

引用

18th International Conference on Wireless Artificial Intelligent Computing Systems and Applications, WASA 2024

作者： Sheng, Weihong Cai, Bin Hu, Chunqiang Li, Ruinian School of Big Data and Software Engineering Chongqing University Chongqing China Computer Science Department Bowling Green State University Bowling Green United States

ISBN: (纸本)9783031714634

In order to provide more comprehensive medical services and personalized health monitoring according to individual needs, Body Area Networks (BANs) have been extensively studied by many researchers. As BANs involve the transmission of personal private data, the security of the communication is of utmost importance. Unfortunately, existing encryption techniques cannot be directly applied with the limited power or computation ability of the sensors in BANs. An alternative direction is to use physiological signals for key agreement. However, many of the current physiological signal-based key agreement schemes either have high overheads or are vulnerable to certain security issues. In this paper, we proposed a novel secret sharing and Bloom filter based key agreement scheme that balances overhead and security for BANs. Also, we use an approximate hash table to free ourselves from the problem of dependence on the ordering of features, which has been neglected by previous researchers. We validate the security of our scheme with based on real datasets and compare overheads with other solutions. © The Author(s), under exclusive license to Springer Nature Switzerland AG 2025.

关键词： Electronic health record

来源：评论

学校读者我要写书评

暂无评论

Context Aware Deep Learning-Based Fault Localization for Hardware Design Code

引用

IEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems 2025年

作者： Hu, Jian Liu, Zhenlei Chongqing University School of Big Data and Software Engineering Chongqing400044 China

Fault localization is a technique that plays a valuable role in reducing the burdensome task of hardware design code, such as Verilog or VHDL. The recent advancements in deep learning have demonstrated significant progress, showcasing the promising potential of various neural network architectures in comprehending and extracting meaningful insights from data. Importantly, this potential offers a new perspective that could potentially benefit fault localization techniques. In this paper, we proposed Carpel: a context aware deep learning-based fault localization method for hardware design code. First, a dynamic execution coverage matrix and an error vector are generated by simulating the hardware code with various stimuli. Subsequently, a failing context is established to narrow the search space for fault localization. Next, three customized deep neural network models are developed and trained using the context-aware coverage matrix and error vector. Finally, a virtual test set is employed to these models to assess the suspiciousness of each statement. Experimental results indicate that our method outperforms state-of-the-art dynamic fault localization techniques, namely Detraque, Tarsel, and Cirfix. Specifically, our approach identifies 37.03%, 22.22%, and 33.33% more faults than Detraque, Tarsel, and Cirfix, respectively, using the Top-1 metric. © 1982-2012 IEEE.

关键词： Hardware-software codesign

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：