检索结果-内蒙古大学图书馆

Learning to Learn Transferable Generative Attack for Person Re-Identification

IEEE Transactions on Image processing 2025年 PP卷 PP页

作者： Bian, Yuan Liu, Min Wang, Xueping Ma, Yunfeng Wang, Yaonan Hunan University National Engineering Research Center of Robot Visual Perception and Control Technology College of Electrical and Information Engineering Hunan Changsha China Hunan Normal University Hunan Provincial Key Laboratory of Intelligent Computing and Language Information Processing College of Information Science and Engineering Hunan Changsha China

Deep learning-based person re-identification (reid) models are widely employed in surveillance systems and inevitably inherit the vulnerability of deep networks to adversarial attacks. Existing attacks merely consider cross-dataset and cross-model transferability, ignoring the cross-test capability to perturb models trained in different domains. To powerfully examine the robustness of real-world re-id models, the Meta Transferable Generative Attack (MTGA) method is proposed, which adopts meta-learning optimization to promote the generative attacker producing highly transferable adversarial examples by learning comprehensively simulated transfer-based crossmodel&dataset&test black-box meta attack tasks. Specifically, cross-model&dataset black-box attack tasks are first mimicked by selecting different re-id models and datasets for meta-train and meta-test attack processes. As different models may focus on different feature regions, the Perturbation Random Erasing module is further devised to prevent the attacker from learning to only corrupt model-specific features. To boost the attacker learning to possess cross-test transferability, the Normalization Mix strategy is introduced to imitate diverse feature embedding spaces by mixing multi-domain statistics of target models. Extensive experiments show the superiority of MTGA, especially in cross-model&dataset and cross-model&dataset&test attacks, our MTGA outperforms the SOTA methods by 20.0% and 11.3% on mean mAP drop rate, respectively. © 1992-2012 IEEE.

关键词： Generative adversarial networks

来源：评论

学校读者我要写书评

暂无评论

Improving Global Precipitation in Numerical Weather Prediction Systems based on Deep Learning Techniques

Improving Global Precipitation in Numerical Weather Predicti...

引用

IEEE International Conference on High Performance Computing and Communications (HPCC)

作者： Haixiang Zhao Guojing Zhang Ming Du Xiaoying Wang Department of Computer Technology and Applications Qinghai University Xining China Geospatial Information Technology and Application Qinghai Provincial Key Laboratory Xining China

Accurate weather forecasts are crucial for many aspects of people's lives. Traditional short-term precipitation forecasting methods mainly rely on numerical weather predictions based on meteorological dynamics modelling. However, numerical weather predictions based on high-resolution physical models require hours of computation on supercomputers. Deep learning networks possess strong feature extraction capabilities, directly mapping images, meteorological data, and other information to high-dimensional feature spaces. Neurons extract abstract features, enabling continuous training and learning for predictions. Therefore, deep learning provides a new solution for short-term precipitation forecasting. In the past, deep learning has been used for precipitation forecasting by employing radar echo extension methods; however, these methods are limited to short-term forecasts and do not utilize other meteorological data, resulting in inaccurate predictions. This study proposes a model containing an attention mechanism, a convolutional extended short-term neural network (Att-PredRNN), based on 5.625° (32×64 grid points) resolution ECMWF reanalysis data and three years of global meteorological and daily precipitation data. The network uses multiple meteorological fields as input factors to predict spatial precipitation distribution for 24-hour precipitation forecasting. The model significantly improves the capture of spatial and temporal correlations through the attention matrices focusing on different parts of the input sequence. Experimental results show that the proposed Att-PredRNN model outperforms other comparison models in terms of RMSE, CSI, POD, and FAR overall and dramatically enhances the accuracy of short-term precipitation forecasts. Our code and data are publicly available.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Efficient survivable mapping algorithm for logical topology in IP-over-WDM optical networks against node failure

引用

Journal of Supercomputing 2023年第5期79卷 5037-5063页

作者： Cheng, Dun-Wei Chang, Jo-Yi Lin, Chen-Yen Lin, Limei Huang, Yanze Thulasiraman, Krishnaiyan Hsieh, Sun-Yuan Department of Computer Science and Information Engineering National Yunlin University of Science and Technology 123 University Road Yunlin64002 Taiwan Department of Computer Science and Information Engineering National Cheng Kung University No. 1 University Road Tainan701 Taiwan College of Mathematics and Informatics and the Key Laboratory of Network Security and Cryptology Fujian Normal University Fujian Fuzhou350117 China Fujian Provincial Key Laboratory of Big Data Mining and Applications School of Mathematics and Physics Fujian University of Technology Fujian Fuzhou350118 China Department of Computer Science University of Oklahoma 660 Parrington Oval NormanOK73019 United States Department of Computer Science and Information Engineering National Chi Nan University No. 1 Daxue Rd Nantou County Puli Township54561 Taiwan

The survivable mapping problem in an IP-over-WDM network with a logical graph and physical graph is the problem of finding a mapping of a logical network in a physical layer so that any failure in physical topology does not break the logical topology’s connection. Determining whether a survivable mapping against failure exists is an NP-complete problem, and therefore many heuristic algorithms have been proposed in the literature. In this paper, a heuristic mapping design strategy is proposed to enable the lightpaths to more efficiently endure a node failure. We individually consider the logical topology by building a Hamiltonian cycle and then deal with the mapping problem for disjoint physical paths. Experimental results demonstrate that the proposed algorithm can provide efficient survivable mapping in IP-over-WDM networks. The simulation shows that the survivability of previous methods (SNF and SMART-Node) is around 60–80%, but our proposed algorithm can build survivable mapping in all kinds of situations. © 2022, The Author(s), under exclusive licence to Springer Science+Business Media, LLC, part of Springer Nature.

关键词： Heuristic algorithms

来源：评论

学校读者我要写书评

暂无评论

A Genuine-Equilibrium Monte Carlo Sampling-Based Effective Algorithm for Fully-Connected Ising Models

A Genuine-Equilibrium Monte Carlo Sampling-Based Effective A...

引用

IEEE International Conference on Integrated Circuits, Technologies and Applications (ICTA)

作者： Xiangrui Wang Dong Jiang Yang Zhang Hongyuan Kang Enyi Yao School of Microelectronics South China University of Technology Guangzhou China Guangdong Provincial Key Laboratory of Optoelectronic Information Processing Chips and Systems Guangzhou China

Ising machines have emerged as a potential technique for hardware acceleration in solving combinatorial optimization problems. However, an effective and hardware-implementation-friendly algorithm is still required to further enhance the fully-connected Ising machine's performance. Parallel momentum-tempering (PMT) algorithm is proposed in this paper, incorporating the full parallel updating capability of momentum annealing (MA) with the global exploration in the solution space provided by parallel tempering (PT). PMT exhibits an excellent performance, and a PMT with 8 replicas surpasses MA on the WK2000 max-cut benchmark, achieving a considerable 80%(1k/6k) reduction in required iteration steps while obtaining a mean cut value of 33,160. Besides, PMT is well-matched for hardware implementation versus MA, benefiting from invariant parameters in each replica.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Using Rough Sets to Improve the High-dimensional Data Anomaly Detection Method Based on Extended Isolation Forest 26

Using Rough Sets to Improve the High-dimensional Data Anomal...

引用

26th International Conference on computer Supported Cooperative Work in Design, CSCWD 2023

作者： Liu, Hanlin Zhou, Jiantao Li, Hua Inner Mongolia University College of Computer Science Hohhot China Natl. Loc. Jt. Eng. Research Center of Intelligent Information Processing Technology for Mongolian Engineering Research Center of Ecological Big Data Ministry of Education Inner Mongolia Engineering Laboratory for Cloud Computing and Service Software Inner Mongolia Key Laboratory of Social Computing and Data Processing Inner Mongolia Key Laboratory of Discipline Inspection and Supervision Big Data Inner Mongolia Engineering Laboratory for Big Data Analysis Technology Hohhot China

ISBN: (纸本)9798350331684

Anomaly detection refers to the identification of data objects that deviate from the general data distribution. One of the important challenges in anomaly detection is handling high-dimensional data, especially when it contains a large number of redundant attributes. Although traditional Isolation Forest (iForest) and Extended Isolation Forest (EIF) have performed well in anomaly detection, the detection accuracy is still extremely influenced by redundant attributes. In this paper, we propose a two-tier architecture named RS-EIF, to alleviate the negative impact of redundant attributes. Specifically, the RS-EIF combines the rough set theory and EIF algorithm and provides an effective basis for the EIF-based anomaly detection method by removing redundant attributes with rough set theory. Experimental results show that the RS-EIF algorithm performs better than the iForest and EIF algorithms in terms of anomaly detection rate, and the results of the combination of rough set theory and EIF are more prominent compared to the other combination methods. © 2023 IEEE.

关键词： Anomaly detection

来源：评论

学校读者我要写书评

暂无评论

A Detection Method Toward Yellow Dragon Disease Spreads Using Spark and Deep Learning 3

A Detection Method Toward Yellow Dragon Disease Spreads Usin...

引用

3rd International Conference on High Performance Computing and Communication Engineering, HPCCE 2023

作者： Xie, Wu Su, Zhenzhao Zhang, Huimin Kang, Ping Qin, Kun Fan, Yong Zhao, Quanyou Guangxi Key Laboratory of Image and Graphic Intelligent Processing Guilin University of Electronic Technology Guilin541214 China Key Laboratory of Education Blockchain and Intelligent Technology Ministry of Education Guangxi Normal University Guilin541004 China Lingchuan Science and Technology Information Institute Guilin541299 China Guilin Hivision Technology Co. Ltd. Guilin541004 China School of Mechanical and Electrical Engineering Guilin University of Electronic Technology Guilin541214 China School of Computer Science and Engineering Guilin University of Aerospace Technology Guilin541004 China

ISBN: (数字)9781510674653

ISBN: (纸本)9781510674646

The detection methods of yellow dragon disease spread via wood lice transmission networks like social networks are very important for diverse citrus trees and farmers. Although current methods have some detection accuracy or low cost, the detection processes are relatively troublesome and the detection cycles are long, making it be a difficult problem to apply them in large-scale orange farms as practical scenarios to detect citrus yellow dragon disease in a timely manner. A new method toward detecting citrus yellow dragon disease spread utilizing Spark and deep learning is proposed for this problem. By obtaining citrus field video stream data through high-definition cameras, and transferring the stream data to the Spark cluster through Kafka like intelligent agents, it is practicable to use the structured streaming component under the Spark framework via big data ecosphere to process video or image stream data transmitted via monitoring. We construct a citrus yellow dragon disease detection model via YOLOv7, and use self-made citrus yellow dragon disease images as training and testing data sets. The preliminary experimental results show that the new methods achieved an accuracy of 83.14%. To reduce the occurrence of missed and false detections, the shallow detection heads are added to the feature fusion networks for improvement, extracting and fusing shallow network information to try to improve the detection effects of yellow dragon disease. Then replace the convolution operation in the ELAN (Effective Long-range Aggregation Network) module with deep separable convolution to reduce the number of model parameters. The preliminary experimental results show that compared to the original YOLOv7 model, our improved citrus yellow dragon disease detection model with YOLOv7 has an accuracy improvement of 2.43%, maintaining a higher detection accuracy with lower time than before. © 2024 SPIE.

关键词： Deep learning

来源：评论

学校读者我要写书评

暂无评论

Retinal Vessels Segmentation Based on Multi-scale Hybrid Convolutional Network 4th

Retinal Vessels Segmentation Based on Multi-scale Hybrid Con...

引用

4th International Conference on Smart Vehicular technology, Transportation, Communication and Applications, VTCA 2021

作者： Li, Rui Li, Zuoyong Cao, Xinrong Teng, Shenghua College of Electronic and Information Engineering Shandong University of Science and Technology Qingdao266590 China Fujian Provincial Key Laboratory of Information Processing and Intelligent Control Minjiang University Fuzhou350121 China Fuzhou Digital Healthcare Industry Technology Innovation Center Minjiang University Fuzhou350121 China

ISBN: (纸本)9789811640384

Retinal fundus image can reveal the information on the early symptoms of diabetes, hypertension, hyperlipidemia and other diseases. Accurate segmentation of retinal vessels can assist the detection and diagnosis of the related diseases. Due to the intricate characteristic information of retinal vessels images, traditional segmentation methods lead to inaccurate segmentation for small vessels and pathological segmentation errors. In this paper, we propose a new multi-scale hybrid convolution U-Net. Firstly, we take the hybrid convolution module by combining dilated convolution and standard convolution as the core structure for feature extraction to obtain more abundant semantic feature information, while expanding the receptive field. Then, we add a multi-scale fusion module to the network encoding and decoding connection part, which fuses the feature information of different layers to reduce the loss of information and enhance the representation ability of the network. We evaluate the performance of the proposed method on two public retinal datasets (DRIVE and CHASE_DB1). The results of quantitative and qualitative experiments show that the proposed model can improve good accuracy in retinal vessels segmentation. © 2022, The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd.

关键词： Convolution

来源：评论

学校读者我要写书评

暂无评论

HairDiffusion: Vivid Multi-Colored Hair Editing via Latent Diffusion

arXiv

引用

arXiv 2024年

作者： Zeng, Yu Zhang, Yang Liu, Jiachen Shen, Linlin Deng, Kaijun He, Weizhao Wang, Jinbao Computer Vision Institute School of Computer Science & Software Engineering Shenzhen University China Shenzhen Institute of Artificial Intelligence and Robotics for Society China National Engineering Laboratory for Big Data System Computing Technology Shenzhen University China Guangdong Provincial Key Laboratory of Intelligent Information Processing China

Hair editing is a critical image synthesis task that aims to edit hair color and hairstyle using text descriptions or reference images, while preserving irrelevant attributes (e.g., identity, background, cloth). Many existing methods are based on StyleGAN to address this task. However, due to the limited spatial distribution of StyleGAN, it struggles with multiple hair color editing and facial preservation. Considering the advancements in diffusion models, we utilize Latent Diffusion Models (LDMs) for hairstyle editing. Our approach introduces Multi-stage Hairstyle Blend (MHB), effectively separating control of hair color and hairstyle in diffusion latent space. Additionally, we train a warping module to align the hair color with the target region. To further enhance multi-color hairstyle editing, we fine-tuned a CLIP model using a multi-color hairstyle dataset. Our method not only tackles the complexity of multi-color hairstyles but also addresses the challenge of preserving original colors during diffusion editing. Extensive experiments showcase the superiority of our method in editing multi-color hairstyles while preserving facial attributes given textual descriptions and reference images. © 2024, CC BY.

关键词： Color image processing

来源：评论

学校读者我要写书评

暂无评论

Few-Shot Object Detection Based on Generalized Features

Few-Shot Object Detection Based on Generalized Features

引用

Artificial Intelligence and Intelligent information processing (AIIIP), International Conference on

作者： Qiuqin Chen Xiao Ke Fujian Provincial Key Laboratory of Networking Computing and Intelligent Information Processing College of Computer and Data Science Fuzhou University Fuzhou China Key Laboratory of Spatial Data Mining & Information Sharing Ministry of Education Fuzhou Fuzhou China

Few-shot object detection aims to rapidly detect novel classes of objects using a minimal number of annotated instances. Compared to methods such as meta-learning, few-shot object detection based on transfer learning can achieve better performance. However, when training on base-class datasets, the model gradually improves its detection performance on base classes but does not enhance its generalization. A base-class model lacking generalization tends to result in lower detection performance for novel classes during fine-tuning. To address the aforementioned issues, this paper tackles the problem from two aspects: firstly, we introduce the Confusion-Dropblock Module (CDM) to perturb the model's features, thus enhancing feature generalization. Secondly, we incorporate Deformable Convolution Modules to mitigate the impact of object shapes and sizes on detection performance, thereby improving overall detection performance. Experimental results on the COCO dataset demonstrate the effectiveness of the proposed algorithm, outperforming the compared methods.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Research on Emotional Analysis of Tibetan Short Text Based on Fusion Sentiment Lexicon 1

引用

1st International Artificial Intelligence Conference, IAIC 2023

作者： Ji, Wan Ma Dao Rang, An Jian Cai Cuo, Jan Yang Jia, Gong Bao Cuo, De Ji Gong, Zan La Qin, Yang School of Computer Science Qinghai University for Minzu Xining810007 China State Key Laboratory of Intelligent Information Processing and Application of Tibetan Language Jointly Constructed By the Ministry of Provincial Affairs and the Ministry of Education of China Xining China Qinghai Key Laboratory of Tibetan Information Processing and Machine Translation Item No. 2021-Z-001 Xining China

ISBN: (数字)9789819712779

ISBN: (纸本)9789819712762

Short-text sentiment analysis plays an integral role in predicting sentiment polarity. The current Tibetan short-text sentiment analysis model applies deep neural networks to learn some local grammatical structure information but ignores known sentiment lexicon information in modeling. Aiming at the above problems, this paper proposes a method for sentiment analysis of Tibetan short-text that incorporates sentiment lexicon information. The method firstly uses the word features extracted by Word2vec and Glove to fuse with the sentiment dictionary information, and inputs them into CNN & BiLSTM networks to extract text context features;secondly, adjusts the weights of the sentiment features through the self-attention mechanism;and finally, uses Softmax to infer the sentiment categories. The experimental results show that the accuracy of this method can reach 93.18% on the self-built sentiment analysis corpus. In addition, compared with the methods of extracting syllable features and word features, the accuracy of this paper's method can be improved by 4.92 and 3.43% respectively, which proves the effectiveness of the fused sentiment lexicon. © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2024.

关键词： Sentiment analysis

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：