检索结果-内蒙古大学图书馆

Source Camera Identification Algorithm Based on Multi-Scale Feature Fusion

computers, Materials & Continua 2024年第8期80卷 3047-3065页

作者： Jianfeng Lu Caijin Li Xiangye Huang Chen Cui Mahmoud Emam School of Computer Science and Technology Hangzhou Dianzi UniversityHangzhou310018China Shangyu Institute of Science and Engineering Hangzhou Dianzi UniversityShaoxing312300China Key Laboratory of Public Security Information Application Based on Big-Data Architecture Ministry of Public SecurityZhejiang Police CollegeHangzhou310000China Faculty of Artificial Intelligence Menoufia UniversityShebin El-Koom32511Egypt

The widespread availability of digital multimedia data has led to a new challenge in digital *** source camera identification algorithms usually rely on various traces in the capturing ***,these traces have become increasingly difficult to extract due to wide availability of various image processing *** Neural Networks(CNN)-based algorithms have demonstrated good discriminative capabilities for different brands and even different models of camera ***,their performances is not ideal in case of distinguishing between individual devices of the same model,because cameras of the same model typically use the same optical lens,image sensor,and image processing algorithms,that result in minimal overall *** this paper,we propose a camera forensics algorithm based on multi-scale feature fusion to address these *** proposed algorithm extracts different local features from feature maps of different scales and then fuses them to obtain a comprehensive feature *** representation is then fed into a subsequent camera fingerprint classification *** upon the Swin-T network,we utilize Transformer Blocks and Graph Convolutional Network(GCN)modules to fuse multi-scale features from different stages of the backbone ***,we conduct experiments on established datasets to demonstrate the feasibility and effectiveness of the proposed approach.

关键词： Source camera identification camera forensics convolutional neural network feature fusion transformer block graph convolutional network

来源：评论

学校读者我要写书评

暂无评论

Knowledge-Enhancement Module for RGB-T Semantic Segmentation in Remote Sensing

Knowledge-Enhancement Module for RGB-T Semantic Segmentation...

引用

IEEE International Symposium on Geoscience and Remote Sensing (IGARSS)

作者： Qingwang Wang Haochen Song Junlin Ouyang Yebo Gu Jian Song Tao Shen Kunming University of Science and Technology Kunming China Yunnan Key Laboratory of Computer Technologies Application Kunming China

ISBN: (数字)9798350360325

ISBN: (纸本)9798350360332

In accomplishing the task of semantic segmentation of RGB-T remote sensing images, there is a great challenge due to severe occlusion, long-tailed data distribution, and insufficient pixel representation of certain objects. The effective use of high-level semantic contexts among various ground object categories is crucial, yet presents considerable diffi-culties. Traditional RGB-T remote sensing image semantic segmentation methods often fail to capture and utilize complex relationships and interdependencies among these categories, leading to reduced semantic segmentation accuracy. This paper proposes a novel, knowledge-enhancement module, empowering the model to utilize human-like commonsense knowledge. Specifically, we firstly employ the self-attention and cross-attention mechanisms to fuse RGB and Thermal features. Subsequently, we collect the weights of the classification layer to get a high-level semantic pool based on all the categories. Alongside this, a prior knowledge graph is developed to enable information propagation among all categories. We applied this knowledge-enhancement module to enhance the Mask R-CNN, named KEMask R-CNN. Experiments results on the RS RGB-T dataset demonstrate the progressiveness of the proposed method.

关键词： Measurement Knowledge engineering Deep learning Accuracy Fuses Semantic segmentation Semantics

来源：评论

学校读者我要写书评

暂无评论

Differential Feature-Enhanced Fusion Network for Hyperspectral Image Classification

Differential Feature-Enhanced Fusion Network for Hyperspectr...

引用

IEEE International Symposium on Geoscience and Remote Sensing (IGARSS)

作者： Qingwang Wang Jiangbo Huang Pengcheng Jin Yebo Gu Tao Shen Kunming University of Science and Technology Kunming China Yunnan Key Laboratory of Computer Technologies Application Kunming China

ISBN: (数字)9798350360325

ISBN: (纸本)9798350360332

Recently, some hybrid networks, combining graph convolutional network (GCN) and convolutional neural network (CNN) into a unified framework, have drawn increasing attention in hyperspectral image (HSI) classification. Compared with the single CNN or GCN architecture, hybrid networks can simultaneously perform feature learning on pixel-level and superpixel-level regions and generate complementary spectral-spatial features. However, existing methods primarily employ simple fusion strategies such as linear combination or concatenation, resulting in the extracted complementary features not being fully exploited and utilized. In this work, we propose a differential feature-enhanced fusion network (DFEFN) for HSI classification. Specifically, DFEFN consists of two different convolutional network architectures, (i.e. GCN and CNN), and a differential feature enhancement fusion (DFEF) module. The features extracted by CNN and GCN can be enhanced and fused through the DFEF module. Experiments results on two benchmark HSI datasets demonstrate that DFEFN achieves better classification performance compared with state-of-the-art methods.

关键词： Representation learning Graph convolutional networks Neural networks Geoscience and remote sensing Benchmark testing Feature extraction Hybrid power systems

来源：评论

学校读者我要写书评

暂无评论

Leveraging Interactive Paths for Sequential Recommendation 28th

Leveraging Interactive Paths for Sequential Recommendation

引用

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

作者： Li, Aoran Zang, Yalei Wang, Yani Li, Bohan College of Computer Science and Technology Nanjing University of Aeronautics and Astronautics Nanjing211106 China Key Laboratory of Safety-Critical Software Ministry of Industry and Information Technology Beijing China National Engineering Laboratory for Integrated Aero-Space-Ground Ocean Big Data Application Technology Xi’an China

ISBN: (纸本)9783031306716

Sequential recommendation systems dynamically predict the users’ next behaviors from chronological historical records to provide more accurate recommendations. However, in many real-world scenarios, recommenders find it difficult to make recommendations for cold-start users who have limited interactions with items. To address this issue, existing methods utilize auxiliary information to capture user preferences, which have achieved performance improvement to some extent. But with the gradual awareness of personal privacy, external information becomes difficult to obtain. Therefore, we develop a novel model called Path-based sequential Recommender for cold-start users (PathRec). PathRec exploits the rich sequential dependencies within the interactive paths to give reasonable recommendations for cold-start users without additional external information. Specifically, (i) PathRec constructs the global graph to learn the representations of the items. (ii) Then, PathRec elaborately selects the sampled paths to derive the underlying dependencies of the cold-start sequence. (iii) Finally, PathRec provides rich and complementary information to user-item interactions by strongly coupling the paths to give suitable recommendations. Extensive experiments on three datasets demonstrate the effectiveness of PathRec. In particular, our model can adapt to the cold-start users and has potential interpretability for recommendations. © 2023, The Author(s), under exclusive license to Springer Nature Switzerland AG.

关键词： Recommender systems

来源：评论

学校读者我要写书评

暂无评论

SLSNet: Weakly-Supervised Skin Lesion Segmentation Network with Self-attentions 20th

SLSNet: Weakly-Supervised Skin Lesion Segmentation Network w...

引用

20th Pacific Rim International Conference on Artificial Intelligence, PRICAI 2023

作者： Pei, Songwen Huang, Junjie School of Optical-Electrical and Computer Engineering University of Shanghai for Science and Technology Shanghai200093 China State Key Laboratory of Computer Architecture Institute of Computing Technology Chinese Academy of Sciences Beijing100190 China Engineering Research Center of Software/Hardware Co-design Technology and Application Ministry of Education East China Normal University Shanghai200062 China

ISBN: (纸本)9789819970247

computer-aided skin lesion segmentation with high precision is crucial to diagnose skin cancers in the early stage. However, the lack of pixel-level labels makes the skin lesion segmentation tasks challenging. To tackle this problem, a new weakly-supervised skin lesion segmentation network with self-attentions named SLSNet is proposed. SLSNet contains two modules and uses image-level labels as supervision information. One module named Intra-image Self-attention Seed Expansion (ISE) digs intra-image self-attentions with an expansion loss and a confidence loss to expand seed areas. The other module named Inter-image Affinity-based Noise Suppression (IAS) suppresses the noise pixels in attention maps via inter-image correlations. Extensive experiments conducted on ISIC-2017 dataset show that SLSNet achieves relatively high performance while reducing human labeling efforts. © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd 2024.

关键词： Expansion

来源：评论

学校读者我要写书评

暂无评论

Towards the Identifiability and Explainability for Personalized Learner Modeling: An Inductive Paradigm 24

Towards the Identifiability and Explainability for Personali...

引用

33rd ACM Web Conference, WWW 2024

作者： Li, Jiatong Liu, Qi Wang, Fei Liu, Jiayu Huang, Zhenya Yao, Fangzhou Zhu, Linbo Su, Yu University of Science and Technology of China Anhui Province Key Laboratory of Big Data Analysis and Application State Key Laboratory of Cognitive Intelligence Hefei China University of Science and Technology of China School of Computer Science and Technology Hefei Comprehensive National Science Center Institute of Artificial Intelligence Hefei China Hefei Normal University Hefei Comprehensive National Science Center Institute of Artificial Intelligence Hefei China

ISBN: (纸本)9798400701719

Personalized learner modeling using cognitive diagnosis (CD), which aims to model learners' cognitive states by diagnosing learner traits from behavioral data, is a fundamental yet significant task in many web learning services. Existing cognitive diagnosis models (CDMs) follow theproficiency-response paradigm that views learner traits and question parameters as trainable embeddings and learns them through learner performance prediction. However, we notice that this paradigm leads to the inevitable non-identifiability and explainability overfitting problem, which is harmful to the quantification of learners' cognitive states and the quality of web learning services. To address these problems, we propose an identifiable cognitive diagnosis framework (ID-CDF) based on a novelresponse-proficiency-response paradigm inspired by encoder-decoder models. Specifically, we first devise the diagnostic module of ID-CDF, which leverages inductive learning to eliminate randomness in optimization to guarantee identifiability and captures the monotonicity between overall response data distribution and cognitive states to prevent explainability overfitting. Next, we propose a flexible predictive module for ID-CDF to ensure diagnosis preciseness. We further present an implementation of ID-CDF, i.e., ID-CDM, to illustrate its usability. Extensive experiments on four real-world datasets with different characteristics demonstrate that ID-CDF can effectively address the problems without loss of diagnosis preciseness. Our code is available at https://***/CSLiJT/ID-CDF. © 2024 ACM.

关键词： Learning systems

来源：评论

学校读者我要写书评

暂无评论

An alert-situation text data augmentation method based on MLM

引用

High technology Letters 2024年第4期30卷 389-396页

作者： DING Weijie MAO Tingyun CHEN Lili ZHOU Mingwei YUAN Ying HU Wentao Department of Computer and Information Security Zhejiang Police CollegeHangzhou 310053P.R.China Key Laboratory of Public Security Information Application Based on Big-Data Architecture Ministry of Public SecurityHangzhou 310053P.R.China Zhejiang Dahua Technology Co. LtdHangzhou 310053P.R.China

The performance of deep learning models is heavily reliant on the quality and quantity of train-ing *** training data will lead to ***,in the task of alert-situation text classification,it is usually difficult to obtain a large amount of training *** paper proposes a text data augmentation method based on masked language model(MLM),aiming to enhance the generalization capability of deep learning models by expanding the training *** method em-ploys a Mask strategy to randomly conceal words in the text,effectively leveraging contextual infor-mation to predict and replace masked words based on MLM,thereby generating new training *** Mask strategies of character level,word level and N-gram are designed,and the performance of each Mask strategy under different Mask ratios is analyzed and *** experimental results show that the performance of the word-level Mask strategy is better than the traditional data augmen-tation method.

关键词： deep learning text data augmentation masked language model(MLM) alert-sit-uation text classification

来源：评论

学校读者我要写书评

暂无评论

CoCGAN: Contrastive Learning for Adversarial Category Text Generation 29

CoCGAN: Contrastive Learning for Adversarial Category Text G...

引用

29th International Conference on Computational Linguistics, COLING 2022

作者： Sheng, Xin Xu, Linli Xu, Yinlong Bao, Changcun Chen, Huang Ren, Bo Anhui Province Key Laboratory of Big Data Analysis and Application School of Computer Science and Technology University of Science and Technology of China China School of Computer Science and Technology University of Science and Technology of China China State Key Laboratory of Cognitive Intelligence China Tencent Youtu Lab China

The task of generating texts of different categories has attracted more and more attention in the area of natural language generation recently. Meanwhile, generative adversarial net (GAN) has demonstrated its effectiveness on text generation, and is further applied to category text generation in later works. Different from existing methods, which mainly consider the pairwise relations between the text embedding and the corresponding fixed one-hot class label (data-to-class relations), this paper proposes a novel Contrastive Category Generative Adversarial Net (CoCGAN) to incorporate contrastive learning into adversarial category text generation, considering more flexible data-to-class relations as well as relations between the multiple text embeddings in the same batch (data-to-data relations). The discriminator of CoCGAN discriminates the authenticity of given samples and optimizes a contrastive learning objective to capture both more flexible data-to-class relations and data-to-data relations among training samples. Accordingly, the generator tries to produce more realistic samples which can confuse the discriminator. Experimental results on both synthetic and real category text generation datasets demonstrate that CoCGAN can achieve significant improvements over the baseline category text generation models. © 2022 Proceedings - International Conference on Computational Linguistics, COLING. All rights reserved.

关键词： Embeddings

来源：评论

学校读者我要写书评

暂无评论

Research on intrusion detection algorithm based on dilated convolution and bidirectional gating cyclic unit

Research on intrusion detection algorithm based on dilated c...

引用

2022 International Conference on Algorithms, Microchips and Network applications

作者： Liu, Xingyuan Miao, Xianghua Faculty of Information Engineering and Automation Kunming University of Science and Technology Kunming650500 China Computer Technology Application Key Laboratory of Yunnan Province Kunming650500 China

ISBN: (纸本)9781510653290

Intrusion detection based on machine learning plays an important role in the security protection of the network environment. In order to improve the accuracy of network intrusion detection, intrusion detection algorithm based on dilated convolution and a two-way gating unit is proposed. Firstly, dilated convolution (DCNN) is used to increase the receptive field of information and extract spatial features. Secondly, the bidirectional gated cyclic unit (bigru) model is used to capture the temporal features of data, and the attention mechanism is introduced to increase the calculation of feature importance. Compared with the classical machine learning classifier, this model has a high detection rate. The multi-classification experiments on the famous UNSW-NB15 data set show that the model has leading performance, and its classification accuracy can reach 97.14%. © COPYRIGHT SPIE. Downloading of the abstract is permitted for personal use only.

关键词： Intrusion detection

来源：评论

学校读者我要写书评

暂无评论

Learning discriminative representation with global and fine-grained features for cross-view gait recognition

引用

CAAI Transactions on Intelligence technology 2022年第2期7卷 187-199页

作者： Jing Xiao Huan Yang Kun Xie Jia Zhu Ji Zhang School of Computer Science South China Normal UniversityGuangzhouGuangdongChina Key Laboratory of Intelligent Education Technology and Application of Zhejiang Province Zhejiang Normal UniversityJinhuaZhejiangChina School of Sciences University of Southern QueenslandToowoomba QldAustralia

In this study,we examine the cross-view gait recognition *** existing methods establish global feature representation based on the whole human body ***,they ignore some important details of different parts of the human *** the latest literature,positioning partial regions to learn fine-grained features has been verified to be effective in human *** they only consider coarse fine-grained fea-tures and ignore the relationship between neighboring *** the above insights together,we propose a novel model called GaitGP,which learns both important details through fine-grained features and the relationship between neighboring regions through global *** GaitGP model mainly consists of the following two ***,we propose a Channel-Attention Feature Extractor(CAFE)to extract the global features,which aggregates the channel-level attention to enhance the spatial information in a novel convolutional ***,we present the Global and Partial Feature Combiner(GPFC)to learn different fine-grained features,and combine them with the global fea-tures extracted by the CAFE to obtain the relevant information between neighboring *** results on the CASIA gait recognition dataset B(CASIA-B),The OU-ISIR gait database,multi-view large population dataset,and The OU-ISIR gait database gait datasets show that our method is superior to the state-of-the-art cross-view gait recognition methods.

关键词： gait database grain

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：