检索结果-内蒙古大学图书馆

Random Fourier features via fast surrogate leverage weighted sampling

学校读者我要写书评

暂无评论

arXiv 2019年

作者： Liu, Fanghui Huang, Xiaolin Chen, Yudong Yang, Jie Suykens, Johan A.K. Department of Electrical Engineering [ESAT-STADIUS KU Leuven Belgium Institute of Image Processing and Pattern Recognition Shanghai Jiao Tong University China Institute of Medical Robotics Shanghai Jiao Tong University China School of Operations Research and Information Engineering Cornell University United States

In this paper, we propose a fast surrogate leverage weighted sampling strategy to generate refined random Fourier features for kernel approximation. Compared to the current state-of-the-art method that uses the leverage weighted scheme (Li et al. 2019), our new strategy is simpler and more effective. It uses kernel alignment to guide the sampling process and it can avoid the matrix inversion operator when we compute the leverage function. Given n observations and s random features, our strategy can reduce the time complexity for sampling from O(ns2 + s3) to O(ns2), while achieving comparable (or even slightly better) prediction performance when applied to kernel ridge regression (KRR). In addition, we provide theoretical guarantees on the generalization performance of our approach, and in particular characterize the number of random features required to achieve statistical guarantees in KRR. Experiments on several benchmark datasets demonstrate that our algorithm achieves comparable prediction performance and takes less time cost when compared to (Li et al. 2019). Copyright © 2019, The Authors. All rights reserved.

关键词： Benchmarking

A novel fast object tracking method based on hierarchical block matching

学校读者我要写书评

暂无评论

A novel fast object tracking method based on hierarchical bl...

2017 IEEE International Conference on Applied System Innovation, ICASI 2017

作者： Yang, Zhi-Hui Liu, Jie-Fei Institute of Image Processing and Pattern Recognition Beijing100144 China

ISBN: (纸本)9781509048977

Research on object tracking has been an active field because of its fundamental roles in surveillance and monitoring. In this paper, a new adaptive algorithm for fast target tracking based on hierarchical block matching is proposed. The improvement of the algorithm lies in three aspects. First, the hierarchical pyramid method is applied to enhance the search efficiency of the local features. In addition, at different pyramid level, an adaptive searching strategy is adopted according to the number of candidate features. Second, since image blocks make full use of local characters of features, the structure similarity method, associated with the relation coefficient in statistics, can be exploited to design a new matching criteria, which provides more precise matching results, especially in the complex scenes such as the change of illumination and partial occlusion. Lastly, traditional algorithms often track the object through adjacent frames, whereas our algorithm will search two frames with a variable interval on the basis of matching results. The intervals are obtained by using the bisection method and thus the searching time can be greatly saved. Simulation results show that the proposed block matching method has better performances in the searching time and accuracy in contrast to the classic Mean-Shift target tracking algorithm and Particle Filter target tracking algorithm under the complex circumstances such as illumination variation and partial occlusion. © 2017 IEEE.

关键词： Adaptive algorithms

Adversarial attack type I: Cheat classifiers by significant changes

学校读者我要写书评

暂无评论

arXiv 2018年

作者： Tang, Sanli Huang, Xiaolin Chen, Mingjian Sun, Chengjin Yang, Jie Institute of Image Processing and Pattern Recognition Institute of Medical Robotics Shanghai Jiao Tong University Shanghai China

Despite the great success of deep neural networks, the adversarial attack can cheat some well-trained classifiers by small permutations. In this paper, we propose another type of adversarial attack that can cheat classifiers by significant changes. For example, we can significantly change a face but well-trained neural networks still recognize the adversarial and the original example as the same person. Statistically, the existing adversarial attack increases Type II error and the proposed one aims at Type I error, which are hence named as Type II and Type I adversarial attack, respectively. The two types of attack are equally important but are essentially different, which are intuitively explained and numerically evaluated. To implement the proposed attack, a supervised variation autoencoder is designed and then the classifier is attacked by updating the latent variables using gradient information. Besides, with pretrained generative models, Type I attack on latent spaces is investigated as well. Experimental results show that our method is practical and effective to generate Type I adversarial examples on large-scale image datasets. Most of these generated examples can pass detectors designed for defending Type II attack and the strengthening strategy is only efficient with a specific type attack, both implying that the underlying reasons for Type I and Type II attack are different. Copyright © 2018, The Authors. All rights reserved.

关键词： Deep neural networks

Microarray camera image segmentation with Faster-RCNN

学校读者我要写书评

暂无评论

Microarray camera image segmentation with Faster-RCNN

International Conference on Applied System Innovation (ICASI)

作者： Jiancheng Zou Rui Song Institute of Image Processing and Pattern Recognition North China University of Technology Shijingshan District Beijing China

A novel method of image segmentation based on Faster-RCNN and microarray camera (3 × 3) is proposed in this paper, we use the microarray camera to obtain nine images in the same scene and use nine array images to achieve image segmentation. Through the microarray camera, we can obtain images quickly. The result of microarray camera and high-speed camera is similar, the shooting interval is short and the disparity information between array images is small. We use the Faster-RCNN algorithm and the GrabCut algorithm to segment target for the array images. Because of the size of microarray camera is small, it can be widely used on portable devices in the further. In comparison to Faster-RCNN network, the experimental results show that the proposed method is more efficient.

关键词： Cameras Lenses image segmentation Calibration Convolution Proposals Feature extraction

EEG-Based Brain-Computer Interfaces Are Vulnerable to Backdoor Attacks

学校读者我要写书评

暂无评论

arXiv 2020年

作者： Meng, Lubin Huang, Jian Zeng, Zhigang Jiang, Xue Yu, Shan Jung, Tzyy-Ping Lin, Chin-Teng Chavarriaga, Ricardo Wu, Dongrui Ministry of Education Key Laboratory of Image Processing and Intelligent Control School of Artificial Intelligence and Automation Huazhong University of Science and Technology Wuhan China Brainnetome Center National Laboratory of Pattern Recognition Institute of Automation Chinese Academy of Sciences Beijing China La Jolla CA United States Center for Advanced Neurological Engineering Institute of Engineering in Medicine Ucsd La Jolla CA United States Centre of Artificial Intelligence Faculty of Engineering and Information Technology University of Technology Sydney Australia Zhaw DataLab Zürich University of Applied Sciences Winterthur8401 Switzerland

Research and development of electroencephalogram (EEG) based brain-computer interfaces (BCIs) have advanced rapidly, partly due to deeper understanding of the brain and wide adoption of sophisticated machine learning approaches for decoding the EEG signals. However, recent studies have shown that machine learning algorithms are vulnerable to adversarial attacks. This article proposes to use narrow period pulse for poisoning attack of EEG-based BCIs, which is implementable in practice and has never been considered before. One can create dangerous backdoors in the machine learning model by injecting poisoning samples into the training set. Test samples with the backdoor key will then be classified into the target class specified by the attacker. What most distinguishes our approach from previous ones is that the backdoor key does not need to be synchronized with the EEG trials, making it very easy to implement. The effectiveness and robustness of the backdoor attack approach is demonstrated, highlighting a critical security concern for EEG-based BCIs and calling for urgent attention to address it. Copyright © 2020, The Authors. All rights reserved.

关键词： Electroencephalography

Deep model-based feature extraction for predicting protein subcellular localizations from bio-images

学校读者我要写书评

暂无评论

Frontiers of Computer Science 2017年第2期11卷 243-252页

作者： Wei SHAO Yi DING Hong-Bin SHEN Daoqiang ZHANG School of Computer Science and Technology Nanjing University of Aeronautics and Astronautics Nanjing 211106 China Institute of Image Processing and Pattern Recognition Shanghai Jiao Tong University Shanghai 200240 China

Protein subcellular localization prediction is im- portant for studying the function of proteins. Recently, as significant progress has been witnessed in the field of mi- croscopic imaging, automatically determining the subcellular localization of proteins from bio-images is becoming a new research hotspot. One of the central themes in this field is to determine what features are suitable for describing the pro- tein images. Existing feature extraction methods are usually hand-crafted designed, by which only one layer of features will be extracted, which may not be sufficient to represent the complex protein images. To this end, we propose a deep model based descriptor （DMD） to extract the high-level fea- tures from protein images. Specifically, in order to make the extracted features more generic, we firstly trained a convolu- tion neural network （i.e., AlexNe0 by using a natural image set with millions of labels, and then used the partial parame- ter transfer strategy to fine-tnne the parameters from natural images to protein images. After that, we applied the Lasso model to select the most distinguishing features from the last fully connected layer of the CNN （Convolution Neural Net- work）, and used these selected features for final classifica- tions. Experimental results on a protein image dataset vali- date the efficacy of our method.

关键词： partial parameter transfer subcellular location classification feature extraction deep model convolution neural network

Indefinite kernel spectral learning 30

学校读者我要写书评

暂无评论

Indefinite kernel spectral learning

30th Benelux Conference on Artificial Intelligence, BNAIC 2018

作者： Mehrkanoon, Siamak Huang, Xiaolin Suykens, Johan A.K. Dept. of Data Science and Knowledge Engineering Maastricht University Netherlands Institute of Image Processing and Pattern Recognition Shanghai Jiao Tong University Shanghai200240 China KU Leuven ESAT-STADIUS Kasteelpark Arenberg 10 LeuvenB-3001 Belgium

This paper introduces the indefinite learning in the framework of least squares support vector machines (LS-SVM). Here the analysis of the Multi-Class Semi-Supervised Kernel Spectral Clustering (MSS-KSC) model with indefinite kernels is provided. In indefinite MSS-KSC one finds the solution by solving a linear system of equations in the dual. In addition the use of the propped indefinite model for large scale data using Nyström approximation technique as well as unsupervised learning using Kernel Spectral Clustering are also discussed in the published full paper1. Experimental results on several real-life datasets are given to illustrate the efficiency of the proposed indefinite kernel spectral learning. © 2018 University of Groningen. All rights reserved.

关键词： Clustering algorithms

Author Correction: A comprehensive multi-domain dataset for mitotic figure detection

学校读者我要写书评

暂无评论

Scientific data 2024年第1期11卷 717页

作者： Marc Aubreville Frauke Wilm Nikolas Stathonikos Katharina Breininger Taryn A Donovan Samir Jabari Mitko Veta Jonathan Ganz Jonas Ammeling Paul J van Diest Robert Klopfleisch Christof A Bertram Technische Hochschule Ingolstadt Ingolstadt Germany. marc.aubreville@thi.de. Pattern Recognition Lab Friedrich-Alexander-Universität Erlangen-Nürnberg Erlangen Germany. Department Artificial Intelligence in Biomedical Engineering Friedrich-Alexander-Universität Erlangen-Nürnberg Erlangen Germany. Department of Pathology University Medical Center Utrecht Utrecht The Netherlands. Schwarzman Animal Medical Center New York USA. Department of Neuropathology Universitätsklinikum Erlangen Friedrich-Alexander-Universität Erlangen-Nürnberg Erlangen Germany. Medical Image Analysis Group Eindhoven University of Technology Eindhoven the Netherlands. Technische Hochschule Ingolstadt Ingolstadt Germany. Institute of Veterinary Pathology Freie Universität Berlin Berlin Germany. Institute of Pathology University of Veterinary Medicine Vienna Vienna Austria.

A regularization approach for instance-based superset label learning

学校读者我要写书评

暂无评论

arXiv 2019年

作者： Gong, Chen Liu, Tongliang Tang, Yuanyan Yang, Jian Yang, Jie Tao, Dacheng School of Computer Science and Engineering Nanjing University of Science and Technology Institute of Image Processing and Pattern Recognition Shanghai Jiao Tong University School of Software Faculty of Engineering and Information Technology University of Technology Sydney UltimoNSW2007 Australia Faculty of Science and Technology University of Macau Macau999078 China College of Computer Science Chongqing University Chongqing400000 China School of Computer Science and Engineering Nanjing University of Science and Technology Nanjing210094 China Institute of Image Processing and Pattern Recognition Shanghai Jiao Tong University Shanghai200240 China School of Information Technologies Faculty of Engineering and Information Technologies University of Sydney J12/318 Cleveland St DarlingtonNSW2008 Australia

Different from the traditional supervised learning in which each training example has only one explicit label, Superset Label Learning (SLL) refers to the problem that a training example can be associated with a set of candidate labels, and only one of them is correct. Existing SLL methods are either regularization-based or instance-based, and the latter of which has achieved state-of-the-art performance. This is because the latest instance-based methods contain an explicit disambiguation operation that accurately picks up the groundtruth label of each training example from its ambiguous candidate labels. However, such disambiguation operation does not fully consider the mutually exclusive relationship among different candidate labels, so the disambiguated labels are usually generated in a nondiscriminative way, which is unfavorable for the instance-based methods to obtain satisfactory performance. To address this defect, we develop a novel Regularization approach for Instancebased Superset Label (RegISL) learning so that our instancebased method also inherits the good discriminative ability possessed by the regularization scheme. Specifically, we employ a graph to represent the training set, and require the examples that are adjacent on the graph to obtain similar labels. More importantly, a discrimination term is proposed to enlarge the gap of values between possible labels and unlikely labels for every training example. As a result, the intrinsic constraints among different candidate labels are deployed, and the disambiguated labels generated by RegISL are more discriminative and accurate than those output by existing instance-based algorithms. The experimental results on various tasks convincingly demonstrate the superiority of our RegISL to other typical SLL methods in terms of both training accuracy and test accuracy. Copyright © 2019, The Authors. All rights reserved.

关键词： Machine learning