检索结果-内蒙古大学图书馆

Spatial Correlation Module for Classification of Multi-Label Ocular Diseases Using Color Fundus Images

Computers, Materials & Continua 2023年第7期76卷 133-150页

作者： Ali Haider Khan Hassaan Malik Wajeeha Khalil Sayyid Kamran Hussain Tayyaba Anees Muzammil Hussain Department of Software Engineering Faculty of Computer SciencesLahore Garrison UniversityLahorePakistan Department of Computer Science School of Systems and TechnologyUniversity of Management and TechnologyLahore54000Pakistan Department of Computer Science&Information Technology University of Engineering and Technology(UET)PeshawarPakistan Department of Computer Science TIMES InstituteMultan60000Pakistan Department of Software Engineering School of Systems and TechnologyUniversity of Management and TechnologyLahore54000Pakistan

To prevent irreversible damage to one’s eyesight,ocular diseases(ODs)need to be recognized and treated *** fundus imaging(CFI)is a screening technology that is both effective and *** to CFIs,the early stages of the disease are characterized by a paucity of observable symptoms,which necessitates the prompt creation of automated and robust diagnostic *** traditional research focuses on image-level diagnostics that attend to the left and right eyes in isolation without making use of pertinent correlation data between the two sets of *** addition,they usually only target one or a few different kinds of eye diseases at the same *** this study,we design a patient-level multi-label OD(PLML_ODs)classification model that is based on a spatial correlation network(SCNet).This model takes into consideration the relevance of patient-level diagnosis combining bilateral eyes and multi-label ODs ***_ODs is made up of three parts:a backbone convolutional neural network(CNN)for feature extraction i.e.,DenseNet-169,a SCNet for feature correlation,and a classifier for the development of classification *** DenseNet-169 is responsible for retrieving two separate sets of attributes,one from each of the left and right *** then,the SCNet will record the correlations between the two feature sets on a pixel-by-pixel *** the attributes have been analyzed,they are integrated to provide a representation at the patient *** the whole process of ODs categorization,the patient-level representation will be *** efficacy of the PLML_ODs is examined using a soft margin loss on a dataset that is readily accessible to the public,and the results reveal that the classification performance is significantly improved when compared to several baseline approaches.

关键词： Ocular disease multi-label spatial correlation CNN eye disease

来源：评论

学校读者我要写书评

暂无评论

An Efficient Dialogue Policy Agent with Model-Based Causal Reinforcement Learning 31

An Efficient Dialogue Policy Agent with Model-Based Causal R...

引用

31st International Conference on Computational Linguistics, COLING 2025

作者： Xu, Kai Wang, Zhenyu Zhao, Yangyang Fang, Bopeng School of Software Engineering South China University of Technology Guangdong China Department of Computer and Communication Engineering Changsha University of Science and Technology Changsha China School of Computer Science and Engineering University of Electronic Science and Technology of China Chengdu China

ISBN: (纸本)9798891761964

Dialogue policy trains an agent to select dialogue actions frequently implemented via deep reinforcement learning (DRL). The model-based reinforcement methods built a world model to generate simulated data to alleviate the sample inefficiency. However, traditional world model methods merely consider one-step dialogues, leading to an inaccurate environmental simulation. Furthermore, different users may have different intention preferences, while most existing studies lack consideration of the intention-preferences causal relationship. This paper proposes a novel framework for dialogue policy learning named MCA, implemented through model-based reinforcement learning with automatically constructed causal chains. The MCA model utilizes an autoregressive Transformer to model dialogue trajectories, enabling a more accurate simulation of the environment. Additionally, it constructs a causal chains module that outputs latent preference distributions for intention-action pairs, thereby elucidating the relationship between user intentions and agent actions. The experimental results show that MCA can achieve state-of-the-art performances on three dialogue datasets over the compared dialogue agents, highlighting its effectiveness and robustness. © 2025 Association for Computational Linguistics.

关键词： Deep reinforcement learning

来源：评论

学校读者我要写书评

暂无评论

Non-Hermitian polarization quantized by time-reversal symmetry

引用

science China(Physics,Mechanics & Astronomy) 2024年第12期67卷 141-147页

作者： Jing Cheng Jinbing Hu Zhigao Hu Lin Chen Minghui Lu Songlin Zhuang Technical Center for Multifunctional Magneto-Optical Spectroscopy(Shanghai) Engineering Research Center of Nanophotonics&Advanced Instrument(Ministry of Education)Department of Physics and Electronic ScienceEast of China Normal UniversityShanghai200241China College of Optical-Electrical Information and Computer Engineering University of Shanghai for Science and TechnologyShanghai200093China Wuhan National Laboratory for Optoelectronics and School of Optical and Electronic Information Huazhong University of Science and TechnologyWuhan430074China National Laboratory of Solid State Microstructures Science and Engineering Nanjing UniversityNanjing210093China

It is well known that in one-dimensional(1D) crystalline insulators,the electric polarization is a manifestation of Berry phase,which can not be quantized by time-reversal symmetry(TRS) as in Hermitian physics TRS does not induce any topological phase in one *** this paper we report that even though associated with complex eigenenergies a 1D non-Hermitian insulator obeying only TRS is capable of presenting quantized bulk *** underlying physical reason is unveiled:TRS guarantees the complex energies to come in pair(E,E*),and the corresponding decaying and amplifying wave functions also come in pair and have the same variation rate,hence,giving rise to a stable wannier *** electron transport is performed by means of charge pumping process,which verifies the physical mechanism *** last,we discuss the possible experimental implementation of the proposed model by means of twisted-π gauge flux.

关键词： non-Hermitian polarization time-reversal symmetry complex hopping

来源：评论

学校读者我要写书评

暂无评论

Collision avoidance path planning method combining improved DWA and A* algorithms in complex sea environments

Collision avoidance path planning method combining improved ...

引用

2024 China Automation Congress, CAC 2024

作者： Liu, Liqiang Sun, Min Zhang, Zhenxing Zhao, Enjiao School of Information and Electronic Engineering Zhejiang University of Science and Technology Hangzhou China School of Intelligent Science and Engineering Harbin Engineering University Harbin China

ISBN: (纸本)9798350368604

This paper proposes a collision avoidance path planning method that combines the improved DWA algorithm with the A* algorithm to address navigation challenges in complex sea environments. First, a simplified grid map is constructed based on electronic chart data and the start and end points of the path planning, while the physical information of the ship is determined. The improved DWA algorithm is then used for collision avoidance path planning. When the ship reaches a safe distance from an obstacle, the improved A* algorithm is triggered to plan a path, and redundant points in the returned path are extracted to obtain key waypoints. These waypoints are used as temporary target points for the DWA algorithm, and the improved DWA algorithm is further applied for collision avoidance path planning. The improvements in the DWA algorithm mainly focus on the trajectory evaluation function, while the enhancements in the A* algorithm focus on the heuristic function. Finally, by combining the proposed algorithm with rich multi-source marine environment data and target fusion data, the ship's intelligent collision avoidance decision-making system is able to generate the optimal collision avoidance path in complex water environments. Simulation results demonstrate that the proposed method can effectively plan a safe and efficient collision avoidance path. © 2024 IEEE.

关键词： A* algorithm collision avoidance decision-making complex sea environments Dynamic Window Approach (DWA) path planning

来源：评论

学校读者我要写书评

暂无评论

Partial VGG16: reformed network for classification under partial occlusion 3

Partial VGG16: reformed network for classification under par...

引用

3rd International Conference on Neural Networks, information and Communication engineering, NNICE 2023

作者： Sun, Bohang Chen, Bo Deng, Yuandan School of Information and Software Engineering University of Electronic Science and Technology of China Chengdu China Yi Bin Park of Uestc University of Electronic Science and Technology of China Yibin China

ISBN: (纸本)9798350335972

Standard deep Convolutional Neural Networks (CNNs), even those with encoder-decoder transformers, often struggle to generalize well when faced with partial occlusion in images. To address this issue, we were inspired by the success of inpainting models that are particularly sensitive to the missed areas in an image. Therefore, we propose to replace the conventional convolutional layers in CNNs with partial convolution blocks, which are commonly used in image inpainting and other related tasks. This new structure allows CNNs to focus only on the valid parts of an image by ignoring the masked parts. To evaluate this approach, we conducted experiments on a model enhanced with partial convolution blocks, by artificially removing rectangular areas from the original images. The results indicate that this model is more sensitive to the non-occluded parts of the object and achieves higher precision than the standard VGG16 network, with a 5% (±2%) increase in top-1 accuracy, has reached to 94.11% in top-2 accuracy. © 2023 IEEE.

关键词： Convolution

来源：评论

学校读者我要写书评

暂无评论

An Explainable Intellectual Property Protection Method for Deep Neural Networks Based on Intrinsic Features

IEEE Transactions on Artificial Intelligence

引用

IEEE Transactions on Artificial Intelligence 2024年第9期5卷 4649-4659页

作者： Xue, Mingfu Wang, Xin Wu, Yinghao Ni, Shifeng Zhang, Leo Yu Zhang, Yushu Liu, Weiqiang East China Normal University School of Communication and Electronic Engineering Shanghai200241 China Nanjing University of Aeronautics and Astronautics College of Computer Science and Technology Nanjing211106 China Griffith University School of Information and Communication Technology QLD4215 Australia Nanjing University of Aeronautics and Astronautics College of Electronic and Information Engineering Nanjing211106 China

Intellectual property (IP) protection for deep neural networks (DNNs) has raised serious concerns in recent years. Most existing works embed watermarks in the DNN model for IP protection, which need to modify the model and do not consider/mention interpretability. In this article, for the first time, we propose an interpretable IP protection method for DNN based on explainable artificial intelligence. Compared with existing works, the proposed method does not modify the DNN model, and the decision of the ownership verification is interpretable. We extract the intrinsic features of the DNN model by using deep Taylor decomposition. Since the intrinsic feature is composed of unique interpretation of the model's decision, the intrinsic feature can be regarded as fingerprint of the model. If the fingerprint of a suspected model is the same as the original model, the suspected model is considered as a pirated model. Experimental results demonstrate that the fingerprints can be successfully used to verify the ownership of the model and the test accuracy of the model is not affected. Furthermore, the proposed method is robust to fine-tuning attack, pruning attack, watermark overwriting attack, and adaptive attack. © 2020 IEEE.

关键词： Deep neural networks

来源：评论

学校读者我要写书评

暂无评论

SnP entangled by carbon nanotube networks as anode for pseudocapacitive half/full battery

引用

Journal of Materiomics 2023年第2期9卷 362-369页

作者： Qian Zhao Dan Zhao Lan Feng Jian Yu Yi Liu Shouwu Guo School of Materials Science and Engineering Shaanxi Key Laboratory of Green Preparation and Functionalization for Inorganic MaterialsShaanxi University of Science and TechnologyXi'an710021China Department of Electronic Engineering School of Electronic Information and Electrical EngineeringShanghai Jiao Tong UniversityShanghai200240China

Due to the lower operating voltage and higher theoretical specific capacity,tin phosphide is considered a class of materials with prospects as an anode material for lithium-ion batteries(LIBs).Among them,tin monophosphide has attracted people's attention due to its unique layered ***,because of the challenging synthesis method and metastable nature,the application of SnP is *** this work,tin phosphide/carbon nanotubes(SnP/CNTs)are prepared by controlling the nucleation and adjusting the ratio of phosphorus/carbon using carbon nanotube as ***-MOF is used as a template to make the morphology of SnP more evenly,and carbon nanotubes can also be used as a conductive network to increase the speed of electron *** an anode material for LIBs,SnP/CNTs reveals superior rate performances(reversible capability of 610 mA·h·g^(-1)at 2000 mA·g^(-1)).The full-cell was assembled and tested,after 50 cycles at 0.1 C,the capacity can maintain 292 mA·h·g^(-1),and its capacity retention rate can reach 80.5%.After 230 cycles,its capacity can maintain at around 223 mA·h·g*** addition,SnP/CNTs materials exhibit 89%pseudocapacitance contribution upon cycling,which indicates the robust Lit storage and satisfactory fast-charging ***,SnP/CNTs suggests a promising anode material for energy storage system.

关键词： SnP Lithium-ion storage Carbon nanotubes Sn4P3 Pseudocapacitive

来源：评论

学校读者我要写书评

暂无评论

Context-based Image Inpainting with Improved Generative Adversarial Networks 3

Context-based Image Inpainting with Improved Generative Adve...

引用

3rd International Conference on Neural Networks, information and Communication engineering, NNICE 2023

作者： Qian, Minqian Chen, Bo Chen, Weiqin School of Information and Software Engineering University of Electronic Science and Technology of China Chengdu China Yi Bin Park of Uestc University of Electronic Science and Technology of China Yibin China

ISBN: (纸本)9798350335972

Recent deep learning based approaches especially with generative adversarial networks and their variants have shown promising results for the challenging task of inpainting large missing regions in an image. These methods can generate visually plausible image structures and textures, but often create distorted structures or blurry textures inconsistent with surrounding areas. For this problem, we proposed a novel approach which not only can repair the defective image with global and local consistency, but also is extremely lightweight. This model is an improved version of Generative Adversarial Network, which has a global and local context discriminator to guarantee the global and local consistency of generated images, utilizes Wasserstein loss to assist the training of its generator and replaces a part of standard convolution with some dilated convolution to expand the receptive field. After experimentation, we found our approach can generate satisfactory image content on the restoration problem of specific scene images such as human faces and street views. © 2023 IEEE.

关键词： Generative adversarial networks

来源：评论

学校读者我要写书评

暂无评论

Ocean wave parameters retrieved directly from compact polarimetric SAR data

引用

Acta Oceanologica Sinica 2022年第4期41卷 129-137页

作者： Yu Liu Yijun He Biao Zhang School of Electronic and Information Engineering Nanjing University of Information Science&TechnologyNanjing 210044China School of Marine Science Nanjing University of Information Science&TechnologyNanjing 210044China

We aim to directly invert wave parameters by using the data of a compact polarimetric synthetic aperture radar(CP SAR)and validate the effectiveness of ocean wave parameter retrieval from the circular transmit/linear receive mode andπ/4 compact polarimetric *** data from the RADARSAT-2 fully polarimetric SAR on the C-band were used to obtain the compact polarimetric SAR images,and a polarimetric SAR wave retrieval algorithm was used to verify the sea surface wave *** the data and algorithm,there is no need to estimate complex hydrodynamic modulation transfer functions,even at large radar incidence ***,the radar backscattering cross-sections and backscattering cross-section of the radar linearly polarized with any polarization orientation angle were calculated in the two compact polarimetric SAR ***,the wave slopes along the azimuth direction and the range direction were calculated directly using CP SAR ***,we obtained the slope spectrum of the wave from the estimated wave slopes along azimuth and range *** wave parameters extracted from the synthetic wave slope spectrum were compared with those obtained from buoy observations of the National Data Buoy Center,verifying a suitable agreement.

关键词： RADARSAT-2 CP SAR wave parameters wave slope spectrum

来源：评论

学校读者我要写书评

暂无评论

Multi-Person Estimation Method Combining Interactive and Contextual information 5

Multi-Person Estimation Method Combining Interactive and Con...

引用

5th International Seminar on Artificial Intelligence, Networking and information technology, AINIT 2024

作者： Chen, Zhuxiang Li, Shujie Jia, Wei School of Computer Science and Information Engineering Hefei University of Technology Hefei China School of Software Hefei University of Technology Hefei China

ISBN: (纸本)9798350385557

Multi-person pose estimation based on monocular cameras is one of the hot research topics in computer vision. Current monocular multi-person 3D pose estimation methods often treat individuals as independent entities for estimation and are based solely on single-frame assessments. This methodology has notable limitations, particularly its insufficient utilization of the rich interactive information interlinking individuals, as well as the contextual continuity inherent across sequential frames. To address these challenges, our study introduces a novel method for monocular multi-person 3D pose estimation. Firstly, it utilizes the attention mechanism of Transformer to learn the contextual and interactive information of both local and global coordinates. Subsequently, the methd employs low-order convolution operators of a root trajectory network to further assimilate global coordinate information. Finally, an optimization algorithm is used to adaptively filter frames with significant occlusions. Our method effectively integrates the interactive information among multiple individuals and the contextual information across multiple frames. Experimental results on the public dataset MuPoTS-3D demonstrate that compared to the state-of-the-art methods, our approach shows a 4.8% improvement in average joint accuracy and a 19. 4% reduction in average root joint error. © 2024 IEEE.

关键词： Computer vision

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：