检索结果-内蒙古大学图书馆

Hyperspectral and Panchromatic Images Fusion Based on the Dual Conditional Diffusion Models

IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING 2023年 61卷 1页

作者： Li, Shuangliang Li, Siwei Zhang, Lihao Wuhan Univ Sch Remote Sensing & Informat Engn Hubei Key Lab Quantitat Remote Sensing Land & Atm Wuhan 430079 Peoples R China Wuhan Univ Hubei Luojia Lab Wuhan 430079 Peoples R China Beijing Normal Univ Fac Geog Sci Beijing 100088 Peoples R China

The fusion between the low-resolution hyperspectral image (LRHSI) and the panchromatic (PAN) image could obtain the high-resolution hyperspectral image (HRHSI). Recently, deep learning (DL)-based fusion methods have been explored widely due to their powerful feature learning ability. However, most DL-based methods that use the one-step fusion manner can suffer from the blurring effect. In addition, they have not fully utilized the spatial and spectral feature information of two input images, which hinders the improvement of the resulting image quality. Therefore, to fully mitigate the blurring effect and utilize two input images, we propose a dual conditional diffusion models-based fusion network (DCDMF) to obtain the fused HRHSI. The conditional diffusion model (CDM) can generate the high-quality image with realistic details in an iterative denoising manner (in the inference sampling stage) other than the one-step fusion manner, which could mitigate the blurring effect greatly. To improve the spatial and spectral fidelity of the fused HRHSI, we propose the dual spatial and spectral CDM (DCDM) (two noise prediction networks with different conditional input) to, respectively, extract the image feature from the LRHSI and PAN images with different image characteristics and reconstruct the corresponding HRHSI feature and fuse them. In addition, considering the high-dimensional property of the hyperspectral image (HSI), we pretrain an auto-encoder to encode the HSI into the low-dimensional latent space with more discriminate features to reduce the computational cost. Based on the autoencoder, we also perform the image generation process in the residual latent space to focus on learning the residual latent spatial details. Extensive experimental results on three datasets show the superiority of our method over several state-of-the-art (SOTA) methods. (The Ziyuan (ZY) dataset and codes could be available at https://***/rs-lsl/DCDMF)

关键词： Feature extraction Noise reduction Spatial resolution Probabilistic logic Task analysis Superresolution Hyperspectral imaging auto-encoder diffusion model hyperspectral image (HSI) fusion residual space spectral and spatial feature

来源：评论

学校读者我要写书评

暂无评论

Unsupervised Learning of Disentangled Representation via auto-Encoding: A Survey

引用

SENSORS 2023年第4期23卷 2362-2362页

作者： Eddahmani, Ikram Pham, Chi-Hieu Napoleon, Thibault Badoc, Isabelle Fouefack, Jean-Rassaire El-Bouz, Marwa Yncrea Ouest LSL Team LbISEN F-29200 Brest France Generix Grp F-75012 Paris France Univ Brest LaTIM INSERM UMR1101 F-29200 Brest France Yncrea Ouest VISION AD Team LbISEN F-29200 Brest France

In recent years, the rapid development of deep learning approaches has paved the way to explore the underlying factors that explain the data. In particular, several methods have been proposed to learn to identify and disentangle these underlying explanatory factors in order to improve the learning process and model generalization. However, extracting this representation with little or no supervision remains a key challenge in machine learning. In this paper, we provide a theoretical outlook on recent advances in the field of unsupervised representation learning with a focus on auto-encoding-based approaches and on the most well-known supervised disentanglement metrics. We cover the current state-of-the-art methods for learning disentangled representation in an unsupervised manner while pointing out the connection between each method and its added value on disentanglement. Further, we discuss how to quantify disentanglement and present an in-depth analysis of associated metrics. We conclude by carrying out a comparative evaluation of these metrics according to three criteria, (i) modularity, (ii) compactness and (iii) informativeness. Finally, we show that only the Mutual Information Gap score (MIG) meets all three criteria.

关键词： representation learning disentanglement auto-encoder generative models neural networks metrics

来源：评论

学校读者我要写书评

暂无评论

SummCoder: An unsupervised framework for extractive text summarization based on deep auto-encoders

引用

EXPERT SYSTEMS WITH APPLICATIONS 2019年第0期129卷 200-215页

作者： Joshi, Akanksha Fidalgo, E. Alegre, E. Fernandez-Robles, Laura Univ Leon Leon Spain INCIBE Spanish Natl Inst Cybersecur Leon Spain CDAC Pune Maharashtra India

In this paper, we propose SummCoder, a novel methodology for generic extractive text summarization of single documents. The approach generates a summary according to three sentence selection metrics formulated by us: sentence content relevance, sentence novelty, and sentence position relevance. The sentence content relevance is measured using a deep auto-encoder network, and the novelty metric is derived by exploiting the similarity among sentences represented as embeddings in a distributed semantic space. The sentence position relevance metric is a hand-designed feature, which assigns more weight to the first few sentences through a dynamic weight calculation function regulated by the document length. Furthermore, a sentence ranking and a selection technique are developed to generate the document summary by ranking the sentences according to the final score obtained through the fusion of the three sentences selection metrics. We also introduce a new summarization benchmark, Tor Illegal Documents Summarization (TIDSumm) dataset, especially to assist Law Enforcement Agencies (LEAs), that contains two sets of ground truth summaries, manually created, for 100 web documents extracted from onion websites in Tor (The Onion Router) network. Empirical results show that, on DUC 2002, on Blog Summarization, and on TIDSumm datasets, our text summarization approach obtains comparable or better performance than the state-of-the-art methods for different ROUGE metrics. (C) 2019 Elsevier Ltd. All rights reserved.

关键词： Extractive text summarization auto-encoder Deep learning Sentence embedding TOR darknet Extractive summarization

来源：评论

学校读者我要写书评

暂无评论

Learning Multilevel auto-encoders for DDoS Attack Detection in Smart Grid Network

引用

IEEE ACCESS 2019年 7卷 108647-108659页

作者： Ali, Shan Li, Yuancheng North China Elect Power Univ Sch Control & Comp Engn Beijing 102206 Peoples R China

Bidirectional communication infrastructure of smart systems, such as smart grids, are vulnerable to network attacks like distributed denial of services (DDoS) and can be a major concern in the present competitive market. In DDoS attack, multiple compromised nodes in a communication network flood connection requests, bogus data packets or incoming messages to targets like database servers, resulting in denial of services for legitimate users. Recently, machine learning based techniques have been explored by researchers to secure the network from DDoS attacks. Under different attack scenarios on a system, measurements can be observed either in an online manner or batch mode and can be used to build predictive learning systems. In this work, we propose an efficient DDoS attack detection technique based on multilevel auto-encoder based feature learning. We learn multiple levels of shallow and deep auto-encoders in an unsupervised manner which are then used to encode the training and test data for feature generation. A final unified detection model is then learned by combining the multilevel features using and efficient multiple kernel learning (MKL) algorithm. We perform experiments on two benchmark DDoS attack databases and their subsets and compare the results with six recent methods. Results show that the proposed method outperforms the compared methods in terms of prediction accuracy.

关键词： auto-encoder cyber security DDoS attack detection multiple kernel learning smart grid

来源：评论

学校读者我要写书评

暂无评论

A survey on deep learning-based image forgery detection

引用

PATTERN RECOGNITION 2023年第1期144卷

作者： Mehrjardi, Fatemeh Zare Latif, Ali Mohammad Zarchi, Mohsen Sardari Sheikhpour, Razieh Yazd Univ Comp Engn Dept Yazd Iran Meybod Univ Comp Engn Dept Meybod Yazd Iran Ardakan Univ Fac Engn Dept Comp Engn POB 184 Ardakan Iran

Image is known as one of the communication tools between humans. With the development and availability of digital devices such as cameras and cell phones, taking images has become easy anywhere. Images are used in many medical, forensic medicine, and judiciary applications. Sometimes images are used as evidence, so the authenticity and reliability of digital images are increasingly important. Some people manipulate images by adding or deleting parts of an image, which makes the image invalid. Therefore, image forgery detection and localization are important. The development of image editing tools has made this issue an important problem in the field of computer vision. In recent years, many different algorithms have been proposed to detect forgery in the image and pixel levels. All these algorithms are categorized into two main methods: traditional and deeplearning methods. The deep learning method is one of the important branches of artificial intelligence science. This method has become one of the most popular methods in most computer vision problems due to the automatic identification and prediction process and robustness against geometric transformations and postprocessing operations. In this study, a comprehensive review of image forgery types, benchmark datasets, evaluation metrics in forgery detection, traditional forgery detection methods, discovering the weaknesses and limitations of traditional methods, forgery detection with deep learning methods, and the performance of this method is presented. According to the expansion of deep-learning methods and their successful performance in most computer vision problems, our main focus in this study is forgery detection based on deep-learning methods. This survey can be helpful for a researcher to obtain a deep background in the forgery detection field.

关键词： Forgery detection Deep learning Inpainting Copy move Splicing Tampered image CNN RNN R-CNN auto-encoder

来源：评论

学校读者我要写书评

暂无评论

Feature Disentangling autoencoder for Anomaly Detection of Reactor Core Temperature with Feature Increment Strategy

引用

PROCESSES 2023年第5期11卷 1486-1486页

作者： Li, Heng Li, Xianmin Mao, Wanchao Chang, Junyu Chen, Xu Zhao, Chunhui Wang, Wenhai China Nucl Power Engn Co Ltd State Key Lab Nucl Power Safety Technol & Equipmen Shenzhen 518172 Guangdong Peoples R China Zhejiang Univ Coll Control Sci & Engn Hangzhou 310027 Peoples R China

Anomaly detection for core temperature has great significance in maintaining the safety of nuclear power plants. However, traditional auto-encoder-based anomaly detection methods might extract the latent space features with redundancy, which may lead to missing and false alarms. To address this problem, the idea of feature disentangling is introduced under the auto-encoder framework in this paper. First, a feature disentangling auto-encoder (DAE) is proposed where a latent space disentangling loss is designed to disentangle the features. We further propose an incrementally feature disentangling auto-encoder (IDAE), which is the improved version of DAE. In the IDAE model, an incremental feature generation strategy is developed, which enables the model to evaluate the disentangling degree to adaptively determine the feature dimension. Furthermore, an iterative training framework is designed, which focuses on the parameter training of the newly incremented feature, overcoming the difficulty of model training. Finally, we illustrate the effectiveness and superiority of the proposed method on a real nuclear reactor core temperature dataset. IDAE achieves average false alarm rates of 4.745% and 6.315%, respectively, using two monitoring statistics, and achieves average missing alarm rates of 6.4% and 2.9%, respectively, using two monitoring statistics, outperforming the other methods.

关键词： feature disentangling feature increment auto-encoder reactor core temperature anomaly detection

来源：评论

学校读者我要写书评

暂无评论

Anti-pruning multi-watermarking for ownership proof of steganographic autoencoders

引用

JOURNAL OF INFORMATION SECURITY AND APPLICATIONS 2023年 76卷

作者： Li, Li Bai, Yu Chang, Ching-Chun Fan, Yunyuan Gu, Wei Emam, Mahmoud Hangzhou Dianzi Univ Sch Comp Sci & Technol Hangzhou 310018 Peoples R China Univ Warwick Dept Comp Sci Coventry CV4 7AL England Anhui Univ Sch Comp Sci & Technol Hefei 230039 Peoples R China Menoufia Univ Fac Artificial Intelligence Shibin Al Kawm 32511 Egypt

Model watermarking Model watermarking is a method for embedding watermark information into a neural network model. It proves the ownership of the model without affecting its performance. Since there are plenty of attacks against model pruning, it becomes more significant to design anti-pruning model watermarking algo-rithms. In this paper, multiple watermark embedding is performed to protect the model copyright for the image steganography auto-encoder model "Hiding Data with Deep Networks" (HiDDeN). Firstly, the appropriate model weights are selected by employing three classical model pruning algorithms of model weights. Secondly, the model watermark is spread by using Discrete Cosine Transform (DCT)-based image watermarking algorithm, which improves the noise and pruning resistance of the model watermark. Finally, the model watermark is embedded to the 4th and 5th decimal places of the selected model weights. The experimental results demonstrate that the proposed algorithm has a good robustness against model pruning without affecting the watermark extraction performance of the auto-encoder network model. Even with the embedded model watermark, the decoder's watermark extraction accuracy is still higher than 0.9993. and the autoencoder is still valuable when the model weights are pruned by 40%. Furthermore, the proposed algorithm has a certain degree of improve-ments in watermarking capacity.

关键词： Anti-pruning auto-encoder Multiple watermarking HiDDeN model DCT

来源：评论

学校读者我要写书评

暂无评论

Rehabilitation exercise score prediction through integration of spatial feature with graph structural learning

引用

ENGINEERING APPLICATIONS OF ARTIFICIAL INTELLIGENCE 2025年 155卷

作者： Raihan, Md. Johir Mustafa, Md. Golam Alammary, Ali Saleh Nahid, Abdullah-Al Khulna Univ Elect & Commun Engn Discipline Khulna 9208 Bangladesh Saudi Elect Univ Coll Comp & Informat Riyadh 11673 Saudi Arabia

automated exercise assessment is crucial for streamlining rehabilitation processes, especially for individuals with limited access to therapists or facing high expenses. This research addresses these challenges by proposing an innovative approach that combines the strengths of two dimensional Convolutional Neural Networks (CNN) and Graph Neural Networks (GNN). Traditional models, such as Long Short-Term Memory (LSTM)-CNN architectures, often struggle to effectively capture both spatial and temporal dependencies within dynamic exercise data. By integrating CNN, which excel at extracting spatial features from image data, with GNN, which enhance the model's ability to capture intricate spatial relationships, we offer a robust solution to these limitations. The primary contribution of this research lies in integrating the structural patterns of graph networks with the visual capabilities of CNN, specifically designed for predicting exercise scores. This proposed Convolutional-Graphical Neural Network (C-GNN) architecture demonstrates promising results, achieving a mean absolute deviation of 0.132 on the KInematic Assessment of MOvement and Clinical Scores for Remote Monitoring of Physical REhabilitation (KIMORE) dataset and 0.029 on the University of Idaho-Physical Rehabilitation Movements Data (UI-PRMD) dataset. These findings underscore the C-GNN's effectiveness in accurately assessing exercise performance across diverse rehabilitation contexts. These results highlight the C-GNN architecture's potential for improving the accessibility and efficiency of rehabilitation processes, particularly in resource-constrained settings. In future work, we aim to improve model interpretability by incorporating techniques like SHapley Additive exPlanations (SHAP) or Local Interpretable Model-Agnostic Explanations (LIME), which can provide deeper insights into the decision-making process.

关键词： Graph neural network Convolutional neural network auto-encoder Rehabilitation Exercise

来源：评论

学校读者我要写书评

暂无评论

Seismic damage identification of high arch dams based on an unsupervised deep learning approach

引用

SOIL DYNAMICS AND EARTHQUAKE ENGINEERING 2023年第1期168卷

作者： Cao, Xiangyu Chen, Liang Chen, Jianyun Li, Jing Lu, Wenyan Liu, Haixiang Ke, Minyong Tang, Yunqing Nanjing Hydraul Res Inst Nanjing 210029 Peoples R China Dalian Univ Technol Fac Infrastructure Engn Sch Hydraul Engn Dalian 116024 Peoples R China Dalian Univ Technol State Key Lab Coastal & Offshore Engn Dalian 116024 Peoples R China

In actual engineering scenarios of arch dams, the incompleteness and nonstationarity of dynamic monitoring signals limit the accurate cognition of the health state. The effectiveness and robustness of damage characteristics in complex environments restrict the practical application of damage diagnosis theory. In this study, guided by the direct extraction of damage sensitivity features from the acceleration response signals of the arch dam, a seismic damage identification approach of high arch dams based on unsupervised learning is developed. Aiming at the problems of low measurement accuracy and poor identification robustness in the existing artificially designed damage-sensitive features, a denoising contractual sparse deep auto-encoder (DCS-DAE) model is proposed by exploring the mapping relationship between monitoring data and the structural state. This model integrates the advantages of denoising auto-encoder, compressive auto-encoder, and sparse auto-encoder. On this basis, based on the principle of reconstruction error and small probability, combined with box-plot and WKNN algorithm, a damage identification framework based on DCS-DAE is constructed. The effectiveness and noise resistance of the proposed method are verified by an extremely high arch dam. The results demonstrate that the damage identification framework based on multi-objective DCS-DAE constructed in this paper only requires the vibration information of the structure in the intact scenario, which provides a solution with higher stability and robustness for the seismic damage identification of high arch dams under strong noise pollution.

关键词： Unsupervised learning Arch dam Damage identification auto-encoder Strong noise pollution

来源：评论

学校读者我要写书评

暂无评论

Face Recognition using mmWave RADAR imaging 7

Face Recognition using mmWave RADAR imaging

引用

7th IEEE International Symposium on Smart Electronic Systems (IEEE-iSES)

作者： Challa, Muralidhar Reddy Kumar, Abhinav Cenkeramaddi, Linga Reddy Indian Inst Technol Dept Artificial Intelligence Hyderabad 502284 India Univ Agder Dept ICI ACPS Res Grp N-4879 Grimstad Norway

ISBN: (纸本)9781728187532

The current work presents a novel approach to signal processing and face recognition based on 60 GHz mmWave RADAR imaging. Machine learning algorithms such as Convolutional auto-encoder and Random Forest algorithm are employed to implement the face recognition scheme. The work presents an approach towards computing and processing higher dimensional RADAR imaging information through extreme feature extraction followed by simple Random Forest, thus enabling a computationally inexpensive algorithm for a mobile friendly implementation.

关键词： mmWave RADAR face recognition Convolutional auto-encoder Random-forest higher dimensional data

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：