检索结果-内蒙古大学图书馆

28th International Conference on Database Systems for Advanced Applications, DASFAA 2023

作者： Yuan, Zhiyang Xiao, Yingyuan Yang, Peilin Hao, Qingbo Wang, Hongya Engineering Research Center of Learning-Based Intelligent System Ministry of Education Tianjin University of Technology Tianjin300384 China Tianjin Key Laboratory of Intelligence Computing and Novel Software Technology Tianjin University of Technology Tianjin300384 China College of Computer Science and Engineering Donghua University Shanghai China

ISBN: (纸本)9783031306716

CTR prediction plays an important role in increasing company revenue and user experience, and many efforts start with historical behavior to uncover user interest. There are two main problems with previous works: (1) When most previous works mined interests from users’ historical behaviors, they only focus on implicit or explicit interests. (2) When most previous works mined user interests through the relationship between target users and similar users, the representation of target users was not rich and accurate enough, resulting in less accurate item recommendations. Therefore the Deep User and Item Inter-Matching Network (DUIIN) is proposed in this paper to solve the above problems. First, we design Item-to-Item Network (IIN), using two different sub-networks Evolving Interest Network (EIN) and Feature Interaction Network (FIN) to mine the hidden interests and explicit interests shown by users’ historical behaviors, respectively. Then the User-to-User Network (UUN) is designed to mine user interests through the relationship between target users and similar users after representing the target users more accurately and richly. The experimental results show that the DUIIN model proposed in this paper performs better than other state-of-the-art models. © 2023, The Author(s), under exclusive license to Springer Nature Switzerland AG.

关键词： Forecasting

来源：评论

学校读者我要写书评

暂无评论

HIGH RESOLUTION IMAGE QUALITY DATABASE

arXiv

引用

arXiv 2024年

作者： Huang, Huang Wan, Qiang Korhonen, Jari Shenzhen University School of Computer Science and Software Engineering Shenzhen China University of Aberdeen School of Natural and Computing Sciences Aberdeen United Kingdom

With technology for digital photography and high resolution displays rapidly evolving and gaining popularity, there is a growing demand for blind image quality assessment (BIQA) models for high resolution images. Unfortunately, the publicly available large scale image quality databases used for training BIQA models contain mostly low or general resolution images. Since image resizing affects image quality, we assume that the accuracy of BIQA models trained on low resolution images would not be optimal for high resolution images. Therefore, we created a new high resolution image quality database (HRIQ), consisting of 1120 images with resolution of 2880 × 2160 pixels. We conducted a subjective study to collect the subjective quality ratings for HRIQ in a controlled laboratory setting, resulting in accurate MOS at high resolution. To demonstrate the importance of a high resolution image quality database for training BIQA models to predict mean opinion scores (MOS) of high resolution images accurately, we trained and tested several traditional and deep learning based BIQA methods on different resolution versions of our database. The database is publicly available in https://***/jarikorhonen/hriq. © 2024, CC BY.

关键词： Image quality

来源：评论

学校读者我要写书评

暂无评论

Robust Vietnam’s Motorcycle License Plate Detection and Recognition Using Deep Learning Model 10th

Robust Vietnam’s Motorcycle License Plate Detection and Re...

引用

10th International Conference on Future Data and Security engineering, FDSE 2023

作者： Le, Duc Hoa Mazumder, Debarshi Quach, Luyl-Da Banerjee, Shreya Nguyen, Vinh Dinh Bosch Global Software Technologies Company Limited Ho Chi Minh city Viet Nam School of Computing and Information Technology Eastern International University Thu Dau Mot Viet Nam Department of Software Engineering FPT University Can Tho City Can Tho Viet Nam Department of Computer Science and Engineering Manipal Institute of Technology Bengaluru Manipal Academy of Higher Education Manipal India

ISBN: (纸本)9789819982950

Nowadays, driving fast is an emergency state in many countries in the world and Vietnam in particular, and the consequence is really cruel when an accident occurs. Therefore, it is very important to control the speed of vehicles. Motorcycle license plate recognition (MLPR) is one of many methods. If any motorcycle drives above the speed limit on the road that has a speed camera, a ticket will be sent to your living address. This research proposes a method that can automatically detect license plates (LP) and extract their data. The highest mAP achieved after three hundred epochs is 93% with Yolov8. There are three stages to extracting the LP’s data, the first is detecting the motorcycle, the second is detecting the LP in the motorcycle’s bounding box, and the third is LP recognition using Yolov8 also. © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd 2023.

关键词： Digital storage

来源：评论

学校读者我要写书评

暂无评论

SyNet:Medical image anomaly detection with noise synthesis network

SyNet:Medical image anomaly detection with noise synthesis n...

引用

IEEE International Conference on Big Data Security on Cloud (BigDataSecurity), High Performance and Smart computing (HPSC) and Intelligent Data and Security (IDS)

作者： Jingnian Liu Wei Liang Yuxiang Chen Kai Jin School of Computer Science and Engineering Hunan Key Laboratory for Service Computing and Novel Software Technology Hunan University of Science and Technology Xiangtang China

ISBN: (数字)9798350389524

ISBN: (纸本)9798350389531

Medical image anomaly detection refers to machine learning techniques to analyze and identify lesions and abnormalities in them. However, in medical images, anomaly samples are usually sparse, which can lead to supervised learning not being able to obtain sufficient training. While existing unsupervised learning is mainly based on reconstruction and generation as well as feature embedding, these methods do not take into account the problem of data type bias during domain migration and often face the problem of loose judgment boundaries. To this end, this paper proposes a novel unsupervised learning method, SyNet, based on noisy anomaly synthesis. First, the problem at domain migration is well solved by aggregating the range of features within the feature map, fusing the multiscale features at the intermediate level, and adding feature adapters. Then, anomaly samples are synthesized by adding noise in the feature space, which is more efficient and stable than generating samples in the image space. Finally, normal and abnormal data are distinguished by training the discriminator. In this paper, experiments are conducted on ISIC and brain MRI datasets, and SyNet achieves nearly 20% improvement in AUC, ACC and other metrics compared with the current mainstream methods, and also has good inference efficiency.

关键词： Training Magnetic resonance imaging Noise Supervised learning Security Noise measurement Lesions

来源：评论

学校读者我要写书评

暂无评论

IP2FL: Interpretation-Based Privacy-Preserving Federated Learning for Industrial Cyber-Physical Systems

IEEE Transactions on Industrial Cyber-Physical Systems

引用

IEEE Transactions on Industrial Cyber-Physical Systems 2024年 2卷 321-330页

作者： Namakshenas, Danyal Yazdinejad, Abbas Dehghantanha, Ali Parizi, Reza M. Srivastava, Gautam University of Guelph Cyber Science Lab Canada Cyber Foundry GuelphONN1G2W1 Canada Kennesaw State University Decentralized Science Lab College of Computing and Software Engineering MariettaGA30060 United States Brandon University Department of Math and Computer Science ManitobaMBR7A6A9 Canada China Medical University Research Centre for Interneural Computing Taichung40402 Taiwan Lebanese American University Department of Computer Science and Mathematics Beirut1102 Lebanon

The expansion of Industrial Cyber-Physical Systems (ICPS) has introduced new challenges in security and privacy, highlighting a research gap in effective anomaly detection while preserving data confidentiality. In the ICPS landscape, where vast amounts of sensitive industrial data are exchanged, ensuring privacy is not just a regulatory compliance issue but a critical shield against industrial espionage and cyber threats. Existing solutions often compromise data privacy for enhanced security, leaving a significant void in protecting sensitive information within ICPS networks. Addressing this, our research presents the IP2FL model, an Interpretation-based Privacy-Preserving Federated Learning approach tailored for ICPS. This model combines Additive Homomorphic Encryption (AHE) for privacy with advanced feature selection methods and Shapley Values (SV) for enhanced explainability. The proposed solution mitigates privacy concerns in federated learning, where traditional methods fall short due to computational constraints and lack of interpretability. By integrating AHE, the IP2FL model minimizes computational overhead and ensures data privacy. Our dual feature selection approach optimizes system performance while incorporating SV to provide critical insights into model decisions, advancing the field towards more transparent and understandable AI systems in ICPS. The validation of our model using ICPS-specific datasets demonstrates its effectiveness and potential for practical applications. © 2023 IEEE.

关键词： Embedded systems

来源：评论

学校读者我要写书评

暂无评论

Advancing Retinal Image Segmentation: A Denoising Diffusion Probabilistic Model Perspective

Advancing Retinal Image Segmentation: A Denoising Diffusion ...

引用

IEEE Conference on Multimedia Information Processing and Retrieval (MIPR)

作者： Alnur Alimanov Md Baharul Islam Department of Computer Engineering Bahcesehir University Istanbul Turkey Department of Computing & Software Engineering Florida Gulf Coast University Fort Myers USA Bahcesehir University Istanbul Turkey

ISBN: (数字)9798350351422

ISBN: (纸本)9798350351439

Retinal images and vessel trees play a crucial role in aiding ophthalmologists to identify and diagnose various illnesses related to the eyes, blood vessels, and brain. However, manual retinal image segmentation is a laborious and highly skilled procedure, posing challenges in terms of both difficulty and time consumption. This study proposes a novel approach to retinal image segmentation, leveraging the Denoising Diffusion Probabilistic Model (DDPM) for precise performance. To our best knowledge, DDPM is being applied in this domain for the first time. Our approach incorporates a novel constraint to prevent DDPM from generating vessel structures that not present in the original retinal images during the segmentation process. Additionally, our model is not limited to the original DDPM size of 64 × 64 pixels. Instead, we train it to effectively segment images sized 256 × 256 pixels. This is a significant advancement since the original DDPM works exclusively with 64 × 64 image sizes and is primarily designed for generating random image samples. In our work, we address both limitations with a novel, efficient approach for accurate retinal image segmentation. A comprehensive evaluation of our methodology includes both quantitative and qualitative assessments. Our proposed method demonstrates competitive performance compared to state-of-the-art techniques, as indicated by both qualitative and quantitative scores. The source code of our method can be accessed at https://***/AAleka/DDPM-segmentation.

关键词： Training Image segmentation Adaptation models Source coding Noise reduction Superresolution Retina Diffusion models Biomedical imaging Testing

来源：评论

学校读者我要写书评

暂无评论

A Comprehensive Overview of Object Detection Based on Deep Learning

A Comprehensive Overview of Object Detection Based on Deep L...

引用

Intelligent Data and Security (IDS), IEEE International Conference on

作者： Gaojing Yuan Linshu Chen Jiahong Cai Chaoyi Yang Jinnian Liu Hunan Key Laboratory for Service Computing and Novel Software Technology School of Computer Science and Engineering Hunan University of Science and Technology Xiangtan China

ISBN: (数字)9798350389487

ISBN: (纸本)9798350389494

Target detection is an important task in the field of computer vision and its application has been extended to the medical field. In the medical field, target detection techniques can be used for diagnosis, treatment and research. For example, by performing target detection on medical images, lesion regions or foci can be automatically identified and localized to help doctors perform accurate diagnosis and treatment. In traditional machine learning methods, commonly used features include morphological features, texture features, and histogram features. And in deep learning methods, the commonly used structures include Convolutional Neural Networks (CNN) and Recurrent Neural Networks (RNN), etc. In addition, there are many special problems and challenges for medical images, such as image quality, variation between modalities, and lack of data. Therefore, target detection in the medical field requires special techniques and algorithms to address these issues. By integrating expertise in the medical field, combining traditional and deep learning techniques, and developing improved and optimized algorithms for the special problems of medical images, the accuracy and efficiency of medical target detection can be improved, thus contributing to the advancement of medical research and clinical practice.

关键词： Deep learning Image quality Recurrent neural networks Accuracy Object detection Medical services Convolutional neural networks

来源：评论

学校读者我要写书评

暂无评论

Question-Answer Cross Language Image Matching for Weakly Supervised Semantic Segmentation

arXiv

引用

arXiv 2024年

作者： Deng, Songhe Zhuo, Wei Xie, Jinheng Shen, Linlin Computer Vision Institute School of Computer Science and Software Engineering Shenzhen University China National Engineering Laboratory for Big Data System Computing Technology Shenzhen University China

Class Activation Map (CAM) has emerged as a popular tool for weakly supervised semantic segmentation (WSSS), allowing the localization of object regions in an image using only image-level labels. However, existing CAM methods suffer from under-activation of target object regions and false-activation of background regions due to the fact that a lack of detailed supervision can hinder the model’s ability to understand the image as a whole. In this paper, we propose a novel Question-Answer Cross-Language-Image Matching framework for WSSS (QA-CLIMS), leveraging the vision-language foundation model to maximize the text-based understanding of images and guide the generation of activation maps. First, a series of carefully designed questions are posed to the VQA (Visual Question Answering) model with Question-Answer Prompt engineering (QAPE) to generate a corpus of both foreground target objects and backgrounds that are adaptive to query images. We then employ contrastive learning in a Region Image Text Contrastive (RITC) network to compare the obtained foreground and background regions with the generated corpus. Our approach exploits the rich textual information from the open vocabulary as additional supervision, enabling the model to generate high-quality CAMs with a more complete object region and reduce false-activation of background regions. We conduct extensive analysis to validate the proposed method and show that our approach performs state-of-the-art on both PASCAL VOC 2012 and MS COCO datasets. Code is available at: https://***/CVI-SZU/QA-CLIMS Copyright © 2024, The Authors. All rights reserved.

关键词： Chemical activation

来源：评论

学校读者我要写书评

暂无评论

Cross-Lingual Word Alignment for ASEAN Languages with Contrastive Learning

Cross-Lingual Word Alignment for ASEAN Languages with Contra...

引用

International Conference on Asian Language Processing (IALP)

作者： Jingshen Zhang Xinying Qiu Teng Shen Wenyu Wang Kailin Zhang Wenhe Feng School of Information Science and Technolgy College of Computer and Software Engineering Hohai University Nanjing China Laboratory of Language Engineering and Computing Guangdong University of Foreign Studies Guangzhou China

ISBN: (数字)9798331540852

ISBN: (纸本)9798331540869

Cross-lingual word alignment plays a crucial role in various natural language processing tasks, particularly for low-resource languages. Recent study proposes a BiLSTM-based encoder-decoder model that outperforms pre-trained language models in low-resource settings. However, their model only considers the similarity of word embedding spaces and does not explicitly model the differences between word embeddings. To address this limitation, we propose incorporating contrastive learning into the BiLSTM-based encoder-decoder framework. Our approach introduces a multi-view negative sampling strategy to learn the differences between word pairs in the shared cross-lingual embedding space. We evaluate our model on five bilingual aligned datasets spanning four ASEAN languages: Lao, Vietnamese, Thai, and Indonesian. Experimental results demonstrate that integrating contrastive learning consistently improves word alignment accuracy across all datasets, confirming the effectiveness of the proposed method in low-resource scenarios. We will release our data set and code to support future research on ASEAN or more low-resource word alignment.

关键词： Codes Accuracy Contrastive learning Natural language processing

来源：评论

学校读者我要写书评

暂无评论

GAN vs. LSTM: Network State Prediction 17

GAN vs. LSTM: Network State Prediction

引用

17th International Conference on Ubiquitous Information Management and Communication, IMCOM 2023

作者： Byun, Gyurin Raza, Syed M. Yang, Huigyu Kim, Moonseong Choo, Hyunseung Sungkyunkwan University Dept. of Superintelligence Engineering Suwon Korea Republic of College of Computing and Informatics Sungkyunkwan University Suwon Korea Republic of Seoul Theological University Dept. of It Convergence Software Bucheon Korea Republic of Sungkyunkwan University Dept. of Electrical and Computer Engineering Suwon Korea Republic of

ISBN: (纸本)9781665453486

Network Traffic prediction is the prerequisite for proactive traffic management, where a longer duration and high accuracy of prediction ensures a more effective solution. This paper exploits Generative Adversarial Network (GAN) architecture to propose a model that utilizes spatiotemporal features in network traffic data for extended prediction of future traffic. The generator in GAN consists of a Convolutional Long Short-Term Memory (ConvLSTM) model for predicting network traffic, while the discriminator uses Convolutional Neural Network (CNN) model that provides feedback to the generator for training. The experiments based on network traces show that the proposed model reduces the error by 12% compared to the baseline model while predicting next 48 mins of network traffic. © 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

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：