检索结果-内蒙古大学图书馆

An efficient binary programming method for black-box optimization and its application in processor design

Science China(information Sciences) 2024年第12期67卷 140-156页

作者： Xiaoliang LV Qiaozhu ZHAI Jianchen HU Yuhang ZHU Jinhui LIU Xiaohong GUAN Faculty of Electronic and Information Engineering Xi'an Jiaotong University Department of Automation Tsinghua University

Optimizing the parameter settings in a large design space for the processor with limited simulation resources is a challenging task. The current black-box optimization algorithms for processor design space exploration(DSE) problems usually require a large amount of simulation resources for high-dimensional and discrete problems. Besides, the constraints handling techniques in these algorithms need to be improved. To address the issues, we propose an efficient binary integer programming(BIP) approach for the DSE of the processor with strictly guaranteed constraints. Our approach involves adopting the separability assumption to establish a surrogate objective function that is ordinal consistent, thus avoiding the complex non-linearity of the real objective function. Moreover, the design rules can be taken simply as constraints in BIP model to further reduce the design space. Thus, the efforts spent in the infeasible exploration space can be *** experimental results show that the proposed algorithm outperforms the state-of-the-art Bayesian optimization and evolutionary algorithms in terms of exploration efficiency, required simulation points and performance of the recommended points.

关键词： black-box optimization binary programming design space exploration processor design expensive simulation system

来源：评论

学校读者我要写书评

暂无评论

An Encode-and CRT-Based Scalability Scheme for Optimizing Transmission in Blockchain

引用

Computer Modeling in engineering & Sciences 2024年第5期139卷 1733-1754页

作者： Qianqi Sun Fenhua Bai Faculty of Information Engineering and Automation Kunming University of Science and TechnologyKunming650500China

Blockchain technology has witnessed a burgeoning integration into diverse realms of economic and societal ***,scalability challenges,characterized by diminished broadcast efficiency,heightened communication overhead,and escalated storage costs,have significantly constrained the broad-scale application of *** paper introduces a novel Encode-and CRT-based Scalability Scheme(ECSS),meticulously refined to enhance both block broadcasting and ***,ECSS categorizes nodes into distinct domains,thereby reducing the network diameter and augmenting transmission ***,ECSS streamlines block transmission through a compact block protocol and robust RS coding,which not only reduces the size of broadcasted blocks but also ensures transmission ***,ECSS utilizes the Chinese remainder theorem,designating the block body as the compression target and mapping it to multiple modules to achieve efficient storage,thereby alleviating the storage burdens on *** evaluate ECSS’s performance,we established an experimental platformand conducted comprehensive *** results demonstrate that ECSS attains superior network scalability and stability,reducing communication overhead by an impressive 72% and total storage costs by a substantial 63.6%.

关键词： Blockchain network coding block compression transmission optimization

来源：评论

学校读者我要写书评

暂无评论

Efficient Attention Transformer Network With Self-Similarity Feature Enhancement for Hyperspectral Image Classification

引用

IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing 2025年 18卷 11469-11486页

作者： Wang, Yuyang Shu, Zhenqiu Yu, Zhengtao Kunming University of Science and Technology Faculty of Information Engineering and Automation Kunming650500 China

Recently, transformer has gained widespread application in hyperspectral image classification (HSIC) tasks due to its powerful global modeling ability. However, the inherent high-dimensional property of hyperspectral images (HSIs) leads to a sharp increase in the number of parameters and expensive computational costs. Moreover, self-attention operations in transformer-based HSIC methods may introduce irrelevant spectral–spatial information, and thus may consequently impact the classification performance. To mitigate these issues, in this article, we introduce an efficient deep network, named efficient attention transformer network (EATN), for practice HSIC tasks. Specifically, we propose two self-similarity descriptors based on the original HSI patch to enhance spatial feature representations. The center self-similarity descriptor emphasizes pixels similar to the central pixel. In contrast, the neighborhood self-similarity descriptor explores the similarity relationship between each pixel and its neighboring pixels within the patch. Then, we embed these two self-similarity descriptors into the original patch for subsequent feature extraction and classification. Furthermore, we design two efficient feature extraction modules based on the preprocessed patches, called spectral interactive transformer module and spatial conv-attention module, to reduce the computational costs of the classification framework. Extensive experiments on four benchmark datasets show that our proposed EATN method outperforms other state-of-the-art HSI classification approaches. © 2008-2012 IEEE.

关键词： Pixels

来源：评论

学校读者我要写书评

暂无评论

Spatial-Frequency Complementary Transformer and Multi-Scale Attention Refinement for Image Inpainting

引用

Journal of Shanghai Jiaotong University (Science) 2025年 1-15页

作者： Ren, Shuangde Li, Huafeng Zhang, Yafei Faculty of Information Engineering and Automation Kunming University of Science and Technology Kunming650500 China

The ability to capture long-range contextual information is crucial for image inpainting networks when dealing with complex semantic scenes and large areas of damage. Therefore, Transformers and frequency-based learning methods have gained wide attention due to their powerful global modeling capabilities. However, these methods only learn global features within a single domain, ignoring the complementarity of modeling global dependencies in both spatial and frequency domains. To address this issue, a spatial-frequency complementary Transformer is proposed, which combines the advantages of spatial global learning for global structure generation and frequency global learning for repetitive texture inpainting. This solution enhances the network’s performance in complex semantic scenes and large areas of damage. For spatial global learning, deformable convolution v4 (DCNv4) is introduced to image inpainting for the first time. By utilizing its sparse and efficient global modeling capabilities, a more lightweight inpainting is realized, and the occurrence of blurring and artifacts in the inpainting results is minimized. Additionally, a multi-scale attention refinement module is proposed to enhance the network’s ability to inpaint local detail. Extensive experiments on three benchmark datasets demonstrate that the proposed method outperforms existing advanced inpainting models while maintaining lower parameter counts and a faster inference speed. © Shanghai Jiao Tong University 2025.

关键词： Contrastive Learning

来源：评论

学校读者我要写书评

暂无评论

Technique for Recovering Wavefront Phase Bad Points by Deep Learning

引用

Chinese Journal of Electronics 2023年第2期32卷 303-312页

作者： WU Jiali LIANG Jingyuan FEI Shaolong ZHONG Xirui Faculty of Automation & Information Engineering Xi'an University of Technology

In adaptive optics systems, the bad spot detected by the wavefront detector affects the wavefront reconstruction accuracy. A convolutional neural network(CNN) model is established to estimate the missing information on bad points, reduce the reconstruction error of the distorted wavefront. By training 10,000 groups of spot array images and the corresponding 30th order Zernike coefficient samples, learns the relationship between the light intensity image and the Zernike coefficient, and predicts the Zernike mode coefficient based on the spot array image to restore the wavefront. Following the wavefront restoration of 1,000 groups of test set samples, the root mean square(RMS) error between the predicted value and the real value was maintained at approximately 0.2 μm. Field wavefront correction experiments were carried out on three links of 600 m, 1.3 km,and 10 km. The wavefront peak-to-valley values corrected by the CNN decreased from 12.964 μm, 13.958 μm,and 31.310 μm to 0.425 μm, 3.061 μm, and 11.156 μm, respectively, and the RMS values decreased from 2.156 μm,9.158 μm, and 12.949 μm to approximately 0.166 μm, 0.852μm, and 6.963 μm, respectively. The results show that the CNN method predicts the missing wavefront information of the sub-aperture from the bad spot image, reduces the wavefront restoration error, and improves the wavefront correction performance.

关键词： Training Atmospheric modeling Optical distortion Feature extraction Adaptive optics Image restoration Convolutional neural networks

来源：评论

学校读者我要写书评

暂无评论

An Adaptive Framework Embedded with LLM for Knowledge Graph Construction

引用

IEEE Transactions on Multimedia 2025年 27卷 2912-2923页

作者： Wang, Qingwang Li, Chaohui Liu, Yi Zhu, Qiubai Song, Jian Shen, Tao Kunming University of Science and Technology Faculty of Information Engineering and Automation Kunming650500 China

Knowledge graph construction is aimed at storing and representing the knowledge of the objective world in a structured form. Existing methods for automatic construction of knowledge graphs have problems such as difficulty in understanding potential semantics and low precision. The emergence of Large Language Models (LLMs) provides an effective way for automatic knowledge graph construction. However, using LLMs as automatic knowledge graph construction engines relies on the embedding of schema layers, which brings challenges to the input length of LLMs. In this paper, we present a framework for Adaptive Construction of Knowledge Graph by leveraging the exceptional generation capabilities of LLMs and the latent relational semantic information of triples, named ACKG-LLM. Our proposed framework divides the knowledge graph construction task into three subtasks within a unified pipeline: triple extraction of open information, additional relational semantic information embedding and knowledge graph normalization based on schema-level embedding. The framework can construct knowledge graphs in different domains, making up for the defects of existing frameworks that need to retrain and fine-tune the internal model. Extensive experiments demonstrate that our proposed ACKG-LLM performs favorably against representative methods on the REBEL and WiKi-NRE datasets. The code is available at https://***/KustTeamWQW/ACKG-LLM © 1999-2012 IEEE.

关键词： Knowledge graph

来源：评论

学校读者我要写书评

暂无评论

Research on a vehicle and pedestrian detection algorithm based on improved attention and feature fusion

引用

Mathematical Biosciences and engineering 2024年第4期21卷 5782-5802页

作者： Liang, Wenjie Faculty of Information Engineering and Automation Kunming University of Science and Technology Kunming China

With the widespread integration of deep learning in intelligent transportation and various industrial sectors, target detection technology is gradually becoming one of the key research areas. Accurately detecting road vehicles and pedestrians is of great significance for the development of autonomous driving technology. Road object detection faces problems such as complex backgrounds, significant scale changes, and occlusion. To accurately identify traffic targets in complex environments, this paper proposes a road target detection algorithm based on the enhanced YOLOv5s. This algorithm introduces the weighted enhanced polarization self attention (WEPSA) self-attention mechanism, which uses spatial attention and channel attention to strengthen the important features extracted by the feature extraction network and suppress insignificant background information. In the neck network, we designed a weighted feature fusion network (CBiFPN) to enhance neck feature representation and enrich semantic information. This strategic feature fusion not only boosts the algorithm's adaptability to intricate scenes, but also contributes to its robust performance. Then, the bounding box regression loss function uses EIoU to accelerate model convergence and reduce losses. Finally, a large number of experiments have shown that the improved YOLOv5s algorithm achieves mAP@0.5 scores of 92.8% and 53.5% on the open-source datasets KITTI and Cityscapes. On the self-built dataset, the mAP@0.5 reaches 88.7%, which is 1.7%, 3.8%, and 3.3% higher than YOLOv5s, respectively, ensuring real-time performance while improving detection accuracy. In addition, compared to the latest YOLOv7 and YOLOv8, the improved YOLOv5 shows good overall performance on the open-source datasets. ©2024 the Author(s)

关键词： Object detection

来源：评论

学校读者我要写书评

暂无评论

CT-MFENet: Context Transformer and Multi-Scale Feature Extraction Network via Global-Local Features Fusion for Retinal Vessels Segmentation

引用

Journal of Shanghai Jiaotong University (Science) 2024年 1-15页

作者： Shao, Dangguo Yang, Yuanbiao Ma, Lei Yi, Sanli Faculty of Information Engineering and Automation Kunming University of Science and Technology Kunming650500 China

Segmentation of the retinal vessels in the fundus is crucial for diagnosing ocular diseases. Retinal vessel images often suffer from category imbalance and large scale variations. This ultimately results in incomplete vessel segmentation and poor continuity. In this study, we propose CT-MFENet to address the aforementioned issues. First, the use of context transformer (CT) allows for the integration of contextual feature information, which helps establish the connection between pixels and solve the problem of incomplete vessel continuity. Second, multi-scale dense residual networks are used instead of traditional CNN to address the issue of inadequate local feature extraction when the model encounters vessels at multiple scales. In the decoding stage, we introduce a local-global fusion module. It enhances the localization of vascular information and reduces the semantic gap between high- and low-level features. To address the class imbalance in retinal images, we propose a hybrid loss function that enhances the segmentation ability of the model for topological structures. We conducted experiments on the publicly available DRIVE, CHASEDB1, STARE, and IOSTAR datasets. The experimental results show that our CT-MFENet performs better than most existing methods, including the baseline U-Net. (Figure presented.) © Shanghai Jiao Tong University 2024.

关键词： Feature extraction

来源：评论

学校读者我要写书评

暂无评论

Regional Constraint Module-Based Multi-Agent Path Planning Approach for Car-like Agents

引用

Journal of Shanghai Jiaotong University (Science) 2024年 1-11页

作者： Fang, Chengyuan Mao, Jianlin Li, Dayan Wang, Ning Wang, Niya Faculty of Information Engineering and Automation Kunming University of Science and Technology Kunming650500 China

Multi-agent path finding (MAPF) is a challenging problem widely employed in automated docks and warehouse systems. However, when the above scenarios require car-like agents to perform the tasks, due to the complexity of the environment and the specificity of the shape of the agents, numerous conflicts between agents may occur in the process of path planning, which seriously affects the efficiency of the system and leads to a long runtime. To address these above problems, we propose a regional constraint module-based car-like conflict-based search (RCM-CL-CBS), which sets up the safe region to detect the conflicts between agents, maximizing the selection of paths with larger spatial resources under the same cost, and specifies the safe-exclusive region for colliding agents, reducing the probability of agents’ collisions within a certain region. We conduct experiments under four scenario types including factory and warehousing instances. Compared with the baseline algorithms, the experimental results denote that our method reduces the computational burden in terms of resolving agent conflicts, and improves the efficiency of problem-solving. In particular, in the warehouse scenario, compared to the car-like conflict-based search (CL-CBS), CL-CBS in the sequential framework (CL-CBS-SE), and improved CL-CBS (ICL-CBS), our method in the sequential framework minimizes the runtime by 89.6%, 53.4%, and 46.6%, respectively. (Figure presented.). © Shanghai Jiao Tong University 2024.

关键词： Motion planning

来源：评论

学校读者我要写书评

暂无评论

DSFNet: Deep Fusion Parallel MRI Reconstruction Based on SPIRiT and Complex Convolution

引用

Journal of Shanghai Jiaotong University (Science) 2024年 1-13页

作者： Duan, Jizhong Ren, Xinmin Chen, Shengyi Faculty of Information Engineering and Automation Kunming University of Science and Technology Kunming650500 China

Long data acquisition time is an inherent disadvantage of magnetic resonance imaging (MRI). To accelerate the data acquisition speed of MRI, undersampling is required, which results in low imaging quality. Based on an iterative self-consistent parallel imaging reconstruction (SPIRiT) and improved complex convolutional neural networks, a deep SPIRiT fusion network (DSFNet) is proposed to improve the quality of reconstructed images. The DSFNet model first uses the SPIRiT model for the reconstruction of under-sampled k-space data. Subsequently, a cascaded complex convolutional neural network with dense connections is utilized for further reconstruction of the calibrated k-space data. Besides, a data consistency layer ensures the fidelity of the reconstructed image in both k-space and image domains. The final magnitude image is fused from the two parts of the reconstruction magnitude images by the sequentialized model-based Bayesian optimization fusion module at a certain ratio. Experimental results on different knee datasets show that DSFNet can bring about a substantial improvement in the visualization quality, peak signal-to-noise ratio and structural similarity of the reconstructed images compared to the SPIRiT, Deepcomplex and DONet models. (Figure presented.) © Shanghai Jiao Tong University 2024.

关键词： Signal to noise ratio

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：