检索结果-内蒙古大学图书馆

arXiv 2021年

作者： Feng, Chun-Mei Yang, Zhanyuan Fu, Huazhu Xu, Yong Yang, Jian Shao, Ling Shenzhen Key Laboratory of Visual Object Detection and Recognition Harbin Institute of Technology Shenzhen518055 China School of Automation Engineering University of Electronic Science and Technology of China 611731 China Inception Institute of Artificial Intelligence Abu Dhabi United Arab Emirates PCA Laboratory Key Lab. of Intelligent Percept. and Syst. for High-Dimensional Information of Ministry of Education Nanjiang University of Science and Technology Nanjiang210094 China Jiangsu Key Laboratory of Image and Video Understanding for Social Security School of Computer Science and Engineering Nanjing University of Science and Technology Nanjing210094 China

Magnetic resonance (MR) image acquisition is an inherently prolonged process, whose acceleration has long been the subject of research. This is commonly achieved by obtaining multiple undersampled images, simultaneously, through parallel imaging. In this paper, we propose the Dual-Octave Network (DONet), which is capable of learning multi-scale spatialfrequency features from both the real and imaginary components of MR data, for parallel fast MR image reconstruction. More specifically, our DONet consists of a series of Dual-Octave convolutions (Dual-OctConv), which are connected in a dense manner for better reuse of features. In each Dual-OctConv, the input feature maps and convolutional kernels are first split into two components (i.e., real and imaginary), and then divided into four groups according to their spatial frequencies. Then, our Dual-OctConv conducts intra-group information updating and inter-group information exchange to aggregate the contextual information across different groups. Our framework provides three appealing benefits: (i) It encourages information interaction and fusion between the real and imaginary components at various spatial frequencies to achieve richer representational capacity. (ii) The dense connections between the real and imaginary groups in each Dual-OctConv make the propagation of features more efficient by feature reuse. (iii) DONet enlarges the receptive field by learning multiple spatial-frequency features of both the real and imaginary components. Extensive experiments on two popular datasets (i.e., clinical knee and fastMRI), under different undersampling patterns and acceleration factors, demonstrate the superiority of our model in accelerated parallel MR image reconstruction. © 2021, CC BY.

关键词： Magnetic resonance imaging

来源：评论

学校读者我要写书评

暂无评论

Performance of Systematic Convolutional Low Density Generator Matrix Codes over Rayleigh Fading Channels with Impulsive Noise 3rd

Performance of Systematic Convolutional Low Density Generato...

引用

3rd International Conference on Space information Networks, SINC 2018

作者： Ji, Meiying Chen, Shengxiao Ma, Xiao School of Data and Computer Science Sun Yat-sen University Guangzhou510006 China School of Electronics and Information Technology Sun Yat-sen University Guangzhou510006 China Guangdong Key Laboratory of Information Security Technology Sun Yat-sen University Guangzhou510006 China

ISBN: (纸本)9789811359361

We investigate the systematic convolutional low density generator matrixÂ (SC-LDGM) codes over Rayleigh fading channels with symmetric alpha-stableÂ (SαS) impulsive noise. The performance is analyzed by deriving a lower bound based on an equivalent genie-aidedÂ (GA) system. Numerical simulations show that the SC-LDGM codes can achieve a significant gain compared to the convolutional codes over Rayleigh fading channels with impulsive noise. Numerical results also show that the performance of the SC-LDGM codes can be around one dB away from Shannon limits at the bit-error rateÂ (BER) of 10-5 and matches well with the GA lower bound in the low BER region. © 2019, Springer Nature Singapore Pte Ltd.

关键词： Impulse noise

来源：评论

学校读者我要写书评

暂无评论

Heavy metals prediction in coastal marine sediments using hybridized machine learning models with metaheuristic optimization algorithm

引用

Chemosphere 2024年 352卷 141329页

作者： Yaseen, Zaher Mundher Melini Wan Mohtar, Wan Hanna Homod, Raad Z. Alawi, Omer A. Abba, Sani I. Oudah, Atheer Y. Togun, Hussein Goliatt, Leonardo Ul Hassan Kazmi, Syed Shabi Tao, Hai Civil and Environmental Engineering Department King Fahd University of Petroleum and Minerals Dhahran 31261 Saudi Arabia Interdisciplinary Research Center for Membranes and Water Security King Fahd University of Petroleum & Minerals (KFUPM) Dhahran Saudi Arabia Department of Civil Engineering Faculty of Engineering and Built Environment Universiti Kebangsaan Malaysia UKM Selangor Bangi 43600 Malaysia Environmental Management Centre Institute of Climate Change Universiti Kebangsaan Malaysia Selangor UKM Bangi 43600 Malaysia Department of Oil and Gas Engineering Basrah University for Oil and Gas Basra Iraq Department of Thermofluids School of Mechanical Engineering Universiti Teknologi Malaysia UTM Skudai Johor Bahru 81310 Malaysia Department of Computer Sciences College of Education for Pure Science University of Thi-Qar Nasiriyah 64001 Iraq Information and Communication Technology Research Group Scientific Research Center Al-Ayen University Nasiriyah 64001 Iraq Department of Mechanical Engineering College of Engineering University of Baghdad Baghdad Iraq Computational and Applied Mechanics Department Federal University of Juiz de Fora 36036-900 Brazil Guangdong Provincial Key Laboratory of Marine Disaster Prediction and Prevention and Guangdong Provincial Key Laboratory of Marine Biotechnology Shantou University Shantou 515063 China School of Computer and Information Qiannan Normal University for Nationalities Guizhou Duyun 558000 China Institute of Big Data Application and Artificial Intelligence Qiannan Normal University for Nationalities Guizhou Duyun 558000 China Faculty of Data Science and Information Technology INTI International University 71800 Malaysia

This study proposes different standalone models viz: Elman neural network (ENN), Boosted Tree algorithm (BTA), and f relevance vector machine (RVM) for modeling arsenic (As (mg/kg)) and zinc (Zn (mg/kg)) in marine sediments owing to anthropogenic activities. A heuristic algorithm based on the potential of RVM and a flower pollination algorithm (RVM-FPA) was developed to improve the prediction performance. Several evaluation indicators and graphical methods coupled with visualized cumulative probability function (CDF) were used to evaluate the accuracy of the models. Akaike (AIC) and Schwarz (SCI) information criteria based on Dickey-Fuller (ADF) and Philip Perron (PP) tests were introduced to check the reliability and stationarity of the data. The prediction performance in the verification phase indicated that RVM-M2 (PBAIS = -o.0465, MAE = 0.0335) and ENN-M2 (PBAIS = 0.0043, MAE = 0.0322) emerged as the best model for As (mg/kg) and Zn (mg/kg), respectively. In contrast with the standalone approaches, the simulated hybrid RVM-FPA proved merit and the most reliable, with a 5 % and 18 % predictive increase for As (mg/kg) and Zn (mg/kg), respectively. The study's findings validated the potential for estimating complex HMs through intelligent data-driven models and heuristic optimization. The study also generated valuable insights that can inform the decision-makers and stockholders for environmental management strategies. © 2024 Elsevier Ltd

关键词： Artificial intelligence Heavy metals Sensitivity analysis Soil contamination

来源：评论

学校读者我要写书评

暂无评论

Auto-Generating Neural Networks with Reinforcement Learning for Multi-Purpose Image Forensics

Auto-Generating Neural Networks with Reinforcement Learning ...

引用

IEEE International Conference on Multimedia and Expo (ICME)

作者： Yujun Wei Yifang Chen Xiangui Kang Z. Jane Wang Liang Xiao Guangdong Key Lab of Information Security School of Data and Computer Science Sun Yat-Sen University Guangzhou China Department of ECE University of British Colombia Vancouver Canada Department of Communication Engineering Xiamen University Xiamen China

ISBN: (数字)9781728113319

ISBN: (纸本)9781728113326

Designing a forensic convolutional neural network (CNN) is usually based on some ad-hoc intuition and domain knowledge. Many methods to automate neural network design have been proposed for computer vision tasks, but they may not be directly applied to image forensic problems, which tend to detect weak traces signals left by image operations rather than strong image content signals. In this paper, we propose an approach to learn an optimal forensic CNN structure with reinforcement learning for detecting multiple image tampering operations. A learning agent is introduced to select CNN layers sequentially in a limited state-action space using Q-learning with an $\epsilon$-greedy strategy and experience replay. The experiments demonstrate that the auto-generated network performs better than other classic image forensic methods and shows more robustness against JPEG compression. To our knowledge, this is the first attempt to design forensic deep neural networks automatically with reinforcement learning.

关键词： Image forensics Learning (artificial intelligence) Task analysis Neural networks Feature extraction Transform coding

来源：评论

学校读者我要写书评

暂无评论

Interactive Two-Stream Decoder for Accurate and Fast Saliency Detection

Interactive Two-Stream Decoder for Accurate and Fast Salienc...

引用

Conference on computer Vision and Pattern Recognition (CVPR)

作者： Huajun Zhou Xiaohua Xie Jian-Huang Lai Zixuan Chen Lingxiao Yang School of Data and Computer Science Sun Yat-sen University China Guangdong Province Key Laboratory of Information Security Technology China Key Laboratory of Machine Intelligence and Advanced Computing Ministry of Education China Sun Yat-sen University Guangzhou China

ISBN: (数字)9781728171685

ISBN: (纸本)9781728171692

Recently, contour information largely improves the performance of saliency detection. However, the discussion on the correlation between saliency and contour remains scarce. In this paper, we first analyze such correlation and then propose an interactive two-stream decoder to explore multiple cues, including saliency, contour and their correlation. Specifically, our decoder consists of two branches, a saliency branch and a contour branch. Each branch is assigned to learn distinctive features for predicting the corresponding map. Meanwhile, the intermediate connections are forced to learn the correlation by interactively transmitting the features from each branch to the other one. In addition, we develop an adaptive contour loss to automatically discriminate hard examples during learning process. Extensive experiments on six benchmarks well demonstrate that our network achieves competitive performance with a fast speed around 50 FPS. Moreover, our VGG-based model only contains 17.08 million parameters, which is significantly smaller than other VGG-based approaches. Code has been made availab.e at: https://***/moothes/ITSD-pytorch.

关键词： Correlation Saliency detection Task analysis Decoding Linear programming Silicon Benchmark testing

来源：评论

学校读者我要写书评

暂无评论

Person re-identification by contour sketch under moderate clothing change

arXiv

引用

arXiv 2020年

作者： Yang, Qize Wu, Ancong Zheng, Wei-Shi School of Data and Computer Science Sun Yat-sen University Guangzhou510275 China School of Electronics and Information Technology Sun Yat-sen University Guangzhou510275 China Guangdong Province Key Laboratory of Information Security China School of Data and Computer Science Sun Yatsen University Guangzhou510275 China Peng Cheng Laboratory Shenzhen518005 China Ministry of Education China

Person re-identification (re-id), the process of matching pedestrian images across different camera views, is an important task in visual surveillance. Substantial development of re-id has recently been observed, and the majority of existing models are largely dependent on color appearance and assume that pedestrians do not change their clothes across camera views. This limitation, however, can be an issue for re-id when tracking a person at different places and at different time if that person (e.g., a criminal suspect) changes his/her clothes, causing most existing methods to fail, since they are heavily relying on color appearance and thus they are inclined to match a person to another person wearing similar clothes. In this work, we call the person re-id under clothing change the "cross-clothes person re-id". In particular, we consider the case when a person only changes his clothes moderately as a first attempt at solving this problem based on visible light images;that is we assume that a person wears clothes of a similar thickness, and thus the shape of a person would not change significantly when the weather does not change substantially within a short period of time. We perform cross-clothes person re-id based on a contour sketch of person image to take advantage of the shape of the human body instead of color information for extracting features that are robust to moderate clothing change. To select/sample more reliable and discriminative curve patterns on a body contour sketch, we introduce a learning-based spatial polar transformation (SPT) layer in the deep neural network to transform contour sketch images for extracting reliable and discriminant convolutional neural network (CNN) features in a polar coordinate space. An angle-specific extractor (ASE) is applied in the following layers to extract more fine-grained discriminant angle-specific features. By varying the sampling range of the SPT, we develop a multistream network for aggregating multi-granular

关键词： Color

来源：评论

学校读者我要写书评

暂无评论

Depthwise Separable Convolutional Neural Network for Image Forensics 34

Depthwise Separable Convolutional Neural Network for Image F...

引用

34th IEEE International Conference on Visual Communications and Image Processing, VCIP 2019

作者： Chen, Yifang Peng, Feng Kang, Xiangui Jane Wang, Z. Sun Yat-sen University Guangdong Key Lab of Information Security School of Data and Computer Science China ECE Dept University of British Colombia VancouverBCV6T 1Z4 Canada

ISBN: (纸本)9781728137230

General-purpose forensics on small image patches appears to be feasible and important, but in fact poses a challenge due to insufficient statistics. Furthermore, there is a need to develop a forensic approach that can automatically learn effective and robust features related to image forensics with high parameter efficiency. In this paper, we propose a depthwise separable convolutional neural network (CNN) for the simultaneous detection of eleven types of image manipulations in image patches. Different from the previous CNNs based on standard convolution, depthwise separable convolution is introduced in the proposed CNN to adaptively extract forensics-related features from image patches with better parameter efficiency. When compared with four state-of-The-Art methods, experiments demonstrate that the proposed CNN architecture can achieve better performance, e.g., the improvement in terms of accuracy in the detection of 32 × 32 images is up to 7.33%. It also achieves significantly better overall performance for different databases and better robustness against JPEG compression. © 2019 IEEE.

关键词： Convolution

来源：评论

学校读者我要写书评

暂无评论

Is AI Robust Enough for Scientific Research?

arXiv

引用

arXiv 2024年

作者： Zhang, Jun-Jie Song, Jiahao Wang, Xiu-Cheng Li, Fu-Peng Liu, Zehan Chen, Jian-Nan Dang, Haoning Wang, Shiyao Zhang, Yiyan Xu, Jianhui Shi, Chunxiang Wang, Fei Pang, Long-Gang Cheng, Nan Zhang, Weiwei Zhang, Duo Meng, Deyu Northwest Institute of Nuclear Technology No. 28 Pingyu Road Shaanxi Xi’an710024 China School of Telecommunications Engineering Xidian University No. 2 South Taibai Road Shaanxi Xi’an710071 China State Key Laboratory of ISN No. 2 South Taibai Road Shaanxi Xi’an710071 China Institute of Particle Physics Central China Normal University No. 152 Luoyu Road Hubei Wuhan30079 China School of Computer Science and Technology Xi’an Jiaotong University No. 28 Xianning West Road Shaanxi Xi’an710049 China Ministry of Education Key Lab of Intelligent Networks and Network Security Xi’an Jiaotong University No. 28 Xianning West Road Shaanxi Xi’an710049 China Guangzhou Institute of Geography Academy of Sciences No. 100 Xianlie Road Guangdong Guangzhou510070 China National Meteorological Information Center Beijing100044 China MDX Research Center for Element Strategy Institute of Integrated Research Institute of Science Tokyo Midori-ku Yokohama226-8503 Japan School of Physics and Information Technology Shaanxi Normal University No. 620 West Chang’an Avenue Shaanxi Xi’an710119 China School of Aeronautics Northwestern Polytechnical University No. 127 West Youyi Road Shaanxi Xi’an710072 China AI for Science Institute Beijing100080 China DP Technology Beijing100080 China Academy for Advanced Interdisciplinary Studies Peking University Beijing100871 China School of Mathematics and Statistics Xi’an Jiaotong University No. 28 Xianning West Road Shaanxi Xi’an710049 China

We uncover a phenomenon largely overlooked by the scientific community utilizing AI: neural networks exhibit high susceptibility to minute perturbations, resulting in significant deviations in their outputs. Through an analysis of five diverse application areas—weather forecasting, chemical energy and force calculations, fluid dynamics, quantum chromodynamics, and wireless communication—we demonstrate that this vulnerability is a broad and general characteristic of AI systems. This revelation exposes a hidden risk in relying on neural networks for essential scientific computations, calling further studies on their reliability and security. © 2024, CC0.

关键词： Weather forecasting

来源：评论

学校读者我要写书评

暂无评论

ZSTAD: Zero-Shot Temporal Activity Detection

ZSTAD: Zero-Shot Temporal Activity Detection

引用

Conference on computer Vision and Pattern Recognition (CVPR)

作者： Lingling Zhang Xiaojun Chang Jun Liu Minnan Luo Sen Wang Zongyuan Ge Alexander Hauptmann School of Computer Science and Technology Xi'an Jiaotong University Xian China Ministry of Education Key Lab For Intelligent Networks and Network Security Xian China Faculty of Information Technology Monash University Australia National Engineering Lab for Big Data Analytics Xi'an Jiaotong University Xian China School of Information Technology and Electrical Engineering The University of Queensland Australia School of Computer Science Carnegie Mellon University USA

ISBN: (数字)9781728171685

ISBN: (纸本)9781728171692

An integral part of video analysis and surveillance is temporal activity detection, which means to simultaneously recognize and localize activities in long untrimmed videos. Currently, the most effective methods of temporal activity detection are based on deep learning, and they typically perform very well with large scale annotated videos for training. However, these methods are limited in real applications due to the unavailab.e videos about certain activity classes and the time-consuming data annotation. To solve this challenging problem, we propose a novel task setting called zero-shot temporal activity detection (ZSTAD), where activities that have never been seen in training can still be detected. We design an end-to-end deep network based on R-C3D as the architecture for this solution. The proposed network is optimized with an innovative loss function that considers the embeddings of activity lab.ls and their super-classes while learning the common semantics of seen and unseen activities. Experiments on both the THUMOS'14 and the Charades datasets show promising performance in terms of detecting unseen activities.

关键词： Semantics Proposals Training Task analysis Feature extraction Microsoft Windows Three-dimensional displays

来源：评论

学校读者我要写书评

暂无评论

ZSTAD: Zero-shot temporal activity detection

arXiv

引用

arXiv 2020年

作者： Zhang, Lingling Chang, Xiaojun Liu, Jun Luo, Minnan Wang, Sen Ge, Zongyuan Hauptmann, Alexander School of Computer Science and Technology Xian Jiaotong University Xian China Ministry of Education Key Lab For Intelligent Networks and Network Security Xian China Faculty of Information Technology Monash University Australia National Engineering Lab for Big Data Analytics Xian Jiaotong University Xian China School of Information Technology and Electrical Engineering University of Queensland Australia School of Computer Science Carnegie Mellon University United States

An integral part of video analysis and surveillance is temporal activity detection, which means to simultaneously recognize and localize activities in long untrimmed videos. Currently, the most effective methods of temporal activity detection are based on deep learning, and they typically perform very well with large scale annotated videos for training. However, these methods are limited in real applications due to the unavailab.e videos about certain activity classes and the time-consuming data annotation. To solve this challenging problem, we propose a novel task setting called zero-shot temporal activity detection (ZSTAD), where activities that have never been seen in training can still be detected. We design an end-to-end deep network based on R-C3D as the architecture for this solution. The proposed network is optimized with an innovative loss function that considers the embeddings of activity lab.ls and their super-classes while learning the common semantics of seen and unseen activities. Experiments on both the THUMOS14 and the Charades datasets show promising performance in terms of detecting unseen activities. Copyright © 2020, The Authors. All rights reserved.

关键词： Deep learning

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：