检索结果-内蒙古大学图书馆

Proceedings of the 35th International conference on Neural Information processing Systems

作者： Qiujiang Jin Aryan Mokhtari ECE Department The University of Texas at Austin Austin TX

ISBN: (纸本)9781713845393

In this paper, we study the application of quasi-Newton methods for solving empirical risk minimization (ERM) problems defined over a large dataset. Traditional deterministic and stochastic quasi-Newton methods can be executed to solve such problems; however, it is known that their global convergence rate may not be better than first-order methods, and their local superlinear convergence only appears towards the end of the learning process. In this paper, we use an adaptive sample size scheme that exploits the superlinear convergence of quasi-Newton methods globally and throughout the entire learning process. The main idea of the proposed adaptive sample size algorithms is to start with a small subset of data points and solve their corresponding ERM problem within its statistical accuracy, and then enlarge the sample size geometrically and use the optimal solution of the problem corresponding to the smaller set as an initial point for solving the subsequent ERM problem with more samples. We show that if the initial sample size is sufficiently large and we use quasi-Newton methods to solve each subproblem, the subproblems can be solved superlinearly fast (after at most three iterations), as we guarantee that the iterates always stay within a neighborhood that quasi-Newton methods converge superlinearly. Numerical experiments on various datasets confirm our theoretical results and demonstrate the computational advantages of our method.

关键词：

来源：评论

学校读者我要写书评

暂无评论

NeurIPS 2020观察与分析

引用

中国图象图形学报 2021年第2期26卷 229-244页

作者：林宙辰王奕森北京大学信息科学技术学院机器感知与智能教育部重点实验室北京100871

神经信息处理系统大会(conference on Neural Information processing Systems,Neur IPS)是机器学习领域的顶级会议,在中国计算机学会(China Computer Federation,CCF)推荐国际学术会议中被评为人工智能领域的A类会议,一直广受关注。Neur IPS 2020收到了创纪录的9467篇投稿,最终录用1898篇论文。收录的论文涵盖了人工智能的各种主题,包括深度学习及其应用、强化学习与规划、纯理论研究、概率方法、优化及机器学习与社会等。本文回顾了Neur IPS 2020的亮点及论文录用情况,详细解读了特邀报告、最佳论文、口头报告及部分海报论文,希望能帮助读者快速了解Neur IPS 2020的盛况。

关键词：人工智能机器学习深度学习强化学习理论优化学术会议 NeurIPS 2020

来源：评论

学校读者我要写书评

暂无评论

SAR image Specle Reduction based on a Generative Adversarial Network

SAR Image Specle Reduction based on a Generative Adversarial...

引用

2020 International Joint conference on Neural Networks, IJCNN 2020

作者： Liu, Ruijiao Li, Yangyang Jiao, Licheng Xidian University School of Artificial Intelligence Xian China

ISBN: (纸本)9781728169262

Synthetic aperture radar (SAR) image despeckling is recognized as the basis for SAR image processing and interpretation. Over the past decades, many impressed speckle reduction methods have been developed and achieved good performance under certain circumstances. However, how to suppress speckle noise in a homogeneous region while more effectively protecting details and avoid distortion of data features caused by homomorphic transformation is still an urgent problem. In this paper, a novel speckle reduction algorithm based on generative adversarial network (GAN) is proposed, which contains a generator and a discriminator. For the generator that is used directly for subsequent noise reduction, a total variation (TV) loss function is added. Meanwhile, we directly learn the mapping between the input image and the ground truth rather than the logarithmic transformation. Indeed, the improved lightweight discriminative network will also provide learning guidance for the generator. Experiments on simulatedSAR images and real SAR images demonstrate the improvement in visual and statistical performance comparing to the state-of-the-art despeckling algorithms. © 2020 IEEE.

关键词： Generative adversarial networks

来源：评论

学校读者我要写书评

暂无评论

Inspection of a machined metal surface adopting scanned contrast image toward unmanned lapping process 20

Inspection of a machined metal surface adopting scanned cont...

引用

20th International conference of the European Society for Precision Engineering and Nanotechnology, EUSPEN 2020

作者： Ravimal, Dinuka Shim, Jaeyoon Kim, Hanul Lee, Sun-Kyu School of Mechanical Engineering Gwangju Institute of Science and Technology Gwangju Korea Republic of

ISBN: (纸本)9780995775176

This paper presents a noble approach for the efficient examination and classification of surface textures on medium- and large-sized mold products, such as used for automobiles, TVs, and refrigerators. Although there are many types of precise surface inspection and measurement methods, most are difficult to apply at industrial sites or by finishing robots due to problems such as speed, setup limitations, and robustness. An accurate and objective method of machined surface inspection has been needed. This study proposes techniques based on image processing that aims to automate surface inspection during an unmanned lapping process that is mainly employed to eliminate milling tool marks. First, both the distribution of the reflected light and the intensity of the captured near-field contrast image generated right adjacent to the reflected specular, are used to determine the machined surface state as well as the presence of tool marks as the line light source scans clockwise or counter-clockwise. The most significant advantage of near-field contrast image, as compared with other vision techniques, is the capturing of highly sensitive contrast image data. It makes possible to distinguish the machined surface state by using the shape of the reflected light and the intensity of the image according to the tool mark, and the tool mark can also be detected. Second, the photometric stereo technique is adopted to detect surface scratches through the normal map that recovers the surface, which shows the entire surface at one time and can detect scratches. The proposed techniques show localized machined patterns and classify them with high accuracy through the statistical process. Copyright © Proceedings of the 20th International conference of the European Society for Precision Engineering and Nanotechnology, EUSPEN 2020. All rights reserved.

关键词： Inspection

来源：评论

学校读者我要写书评

暂无评论

A kNN Based Voyage’s Containers’ Entering Time Distribution Prediction System

A kNN Based Voyage’s Containers’ Entering Time Distributio...

引用

Symposium of image, Signal processing, and Artificial Vision (STSIVA)

作者： Shitong Shen Jian Cao Yinyue Yang Yameng Guo Shanghai Jiao Tong University Shanghai China Shangdong Container Terminal Branch Shanghai International Port (Group) Co.Ltd Shanghai China

ISBN: (纸本)9781665426565

Compared with the air transportation and land transportation, water transportation has many advantages such as larger loading capacity, lower unit transportation cost, lower construction investment and so on. What’s more, water transportation has played an important role in the economical development of China, especially in the aspect of international trade. Therefore, the improvement in the efficiency of water transportation will be of great significance. In this paper, we designed a system to predict the containers’ entering time distribution of a given voyage at a specific port by using machine learning algorithms and statistical methods. Using Shanghai Yangshan Port phase IV automated terminal’s data, we perform some experiments, and the result shows that our system can provide valid predictions.

关键词： International trade Machine learning algorithms Costs statistical analysis Loading Land transportation Containers

来源：评论

学校读者我要写书评

暂无评论

Facial Complexion Recognition of Traditional Chinese Medicine Based on Computer Vision 5

Facial Complexion Recognition of Traditional Chinese Medicin...

引用

5th International conference on Computational Intelligence and Applications (ICCIA)

作者： Lin, Yi Wang, Bin Nanjing Univ Finance & Econ Sch Informat Engn Nanjing Peoples R China

ISBN: (纸本)9781728160429

This paper makes an attempt to develop an automated facial complexion recognition method for objective and quantitative facial diagnosis. In TCM diagnosis, some regions of the face like Ting, Jia and Mingtang, can provide the most valuable information, so we use deep learning technique to determine the 68 landmarks of face and use their location to segment the regions of interest (ROI). The statistical characteristics of color histograms in multiple color space and texture features, lip color features are then introduced to describe the facial complexion. Finally, several machine learning methods including KNN, SVM and BPNN are used for classification. To verify the validity of our method, we collected a dataset of 575 face images from professional TCM medical institutions. The results show that the process of ROIs' segmentation can improve the accuracy efficiently, higher than unsegmented image. The proposed method by fusing all three features achieves an accuracy of 91.03% which is higher than the existing methods and proves the effectiveness of the proposed method for facial complexion recognition. We confirm that extracting the complexion features particularly from the regions of interest of the face image achieves higher classification accuracy than characterizing the overall complexion directly from the unsegmented images. We show that the facial color features provide the most important clues for complexion classification among all the used features, which is consistent with the TCM diagnosis. Finally, we prove that the facial texture feature and lip color feature can be used as complementary clues and fused with the facial color features for further improving the complexion classification accuracy.

关键词： picture processing inspection of TCM complexion recognition feature fusion

来源：评论

学校读者我要写书评

暂无评论

Optimization methods for deep neural networks classifying OCT images to detect dental caries 26

Optimization methods for deep neural networks classifying OC...

引用

Symposium on Lasers in Dentistry XXVI held at SPIE BiOS conference

作者： Salehi, Hassan S. Barchini, Majd Mandian, Mina Calif State Univ Chico Dept Elect & Comp Engn Chico CA 95929 USA SUNY Stony Brook Sch Dent Med Stony Brook NY 11794 USA

ISBN: (纸本)9781510631977;9781510631984

Dental caries are common chronic infectious oral diseases affecting most teenagers and adults worldwide. Optical coherence tomography (OCT) has been studied extensively for the detection of early carious lesions. Deep learning techniques are a rapidly emerging new area of biomedical research and have yielded impressive results in diagnosis and prediction in the field of oral radiology. Deep learning models particularly deep convolutional neural networks (CNN) can be employed along with OCT imaging system to more accurately identify early dental caries. In this work, after OCT data acquisition, data augmentation was performed to obtain a large amount of training data in order to effectively learn, where collection of such training data is often expensive and laborious. For the backpropagation process, seven optimization methods, namely Adadelta, AdaGrad, Adam, AdaMax, Nadam, RMSProp, and stochastic Gradient Descent (SGD) were utilized to improve the accuracy of a CNN classifier for diagnosing dental caries. In this study, 75% of the data were utilized for training and 25% for testing. The diagnostic accuracy, sensitivity, specificity, positive predictive value, negative predictive value, and receiver operating characteristic (ROC) curve were calculated for detection and diagnostic performance of the deep CNN algorithm. This study highlighted the performance of various optimization methods for deep CNN models with OCT images to detect dental caries.

关键词： optimization methods image processing machine learning deep learning convolutional neural networks optical coherence tomography dental caries detection

来源：评论

学校读者我要写书评

暂无评论

Identification of the quality of premium and non-premium rice based on physical characteristics using artificial neural networks and digital image processing

引用

AIP conference Proceedings 2024年第1期3065卷

作者： Endah Sekar Palupi Riries Rulaningtyas Toetik Rahayuningsih Ahmad Yudianto Department of Forensic Science Postgraduate School Department of Health Faculty of Vocational Studies Universitas Airlangga Surabaya Indonesia Biomedical Engineering Study Program Department of Physics Universitas Airlangga Surabaya Indonesia Criminal Law Department of Law Faculty of Law Universitas Airlangga Surabaya Indonesia Department of Forensic Medicine and Medicolegal Faculty of Medicine Universitas Airlangga Surabaya Indonesia

Counterfeiting of quality rice was rife in Indonesia. This research was conducted to develop technology to identify differences in premium and non-premium rice quality based on pre-existing digital images. Artificial neural networks and digital image processing methods to identify premium and medium (non-premium) rice quality were applied in this research. statistical analysis of this study used the SPSS program. This research is observation-type research. This research design uses an artificial neural network with uses 3 layers, namely the results of shape feature extraction on the metric, eccentricity, area, and perimeter parameters as input or input layers, hidden or hidden layers, and premium rice and non-premium (medium) rice as output or output layers. This research uses 52 images as training and 20 images as testing. The obtained image was taken at a distance of 25 cm. This research showed that the results of training using artificial neural networks (ANN) on 52 images obtained an accuracy of 92%. The test results using 20 images obtained 95% accuracy, 63.33% sensitivity, and 10% specificity. Based on statistical analysis using the Mann-Whitney test, it obtained the asymph value. Sig (2-tailed) < 0.05 indicates the difference between premium and non-premium rice using metric, eccentricity, perimeter, and area parameters.

关键词：

来源：评论

学校读者我要写书评

暂无评论

New S-Box transformation based on chaotic system for image encryption 3

New S-Box transformation based on chaotic system for image e...

引用

3rd International conference on Engineering Technology and its Applications, IICETA 2020

作者： Mohaimen Hassan, J. Alaa Kadhim, F. Computer Science Department University of Technology Baghdad Iraq

ISBN: (纸本)9781728182315

Chaos theory one of the complex approaches with encryption methods is known as a perfect candidate. This is due to using cryptography applications to its properties like unpredictability and sensitive initial states. S-box based on Logistics Map has been suggested. in this paper establish a new method of composition, saving time in encryption, allowing a high-efficiency non-linearity of random sequences of the created values. showed a new diffusion system and suggested utilizing the dispersion principle, based on the Lorenz Map, disengage the color points. (R, G, B). Initiate this proposal by shuffling all pixels in tow steps for each color channel to swap the pixels horizontally, vertically, or diagonally, last step marked with an XOR. The image encryption method reasonably requires a few milliseconds, and the considerable period for the familiar image to be decrypted fairly close as the images. block encoder is reversible to the developed algorithm. The unique experiments merely demonstrate that the information security capabilities will be both safe more efficient, the current S-Box analysis with statistical criteria, including balanced criteria (BC), completeness criteria (CC), avalanche criteria (AC), strict avalanche criteria (SAC), and invertibility, the analysis of performance tests demonstrate the strong cryptographic properties of the S-box discussed in this new S-Box. The consistent result of our critical analysis of picture quality evaluation, differential attacks (NPSR), (UACI), entropy information, and pixels correlation show the peculiar quality and political strength of encryption processing. © 2020 IEEE.

关键词： Cryptography

来源：评论

学校读者我要写书评

暂无评论

Chebyshev-Harmonic-Fourier-Moments and Deep CNNs for Detecting Forged Handwriting

Chebyshev-Harmonic-Fourier-Moments and Deep CNNs for Detecti...

引用

International conference on Pattern Recognition

作者： Lokesh Nandanwar Palaiahnakote Shivakumara Sayani Kundu Umapada Pal Tong Lu Daniel Lopresti Faculty of Computer Science and Information Technology University of Malaya Kuala Lumpur Malaysia Computer Vision and Pattern Recognition Unit Indian Statistical Institute Kolkata India National Key Lab for Novel Software Technology Nanjing University Nanjing China Computer Science & Engineering Lehigh University Bethlehem PA USA

Recently developed sophisticated image processing techniques and tools have made easier the creation of high-quality forgeries of handwritten documents including financial and property records. To detect such forgeries of handwritten documents, this paper presents a new method by exploring the combination of Chebyshev-Harmonic-Fourier-Moments (CHFM) and deep Convolutional Neural Networks (D-CNNs). Unlike existing methods work based on abrupt changes due to distortion created by forgery operation, the proposed method works based on inconsistencies and irregular changes created by forgery operations. Inspired by the special properties of CHFM, such as its reconstruction ability by removing redundant information, the proposed method explores CHFM to obtain reconstructed images for the color components of the Original, Forged Noisy and Blurred classes. Motivated by the strong discriminative power of deep CNNs, for the reconstructed images of respective color components, the proposed method used deep CNNs for forged handwriting detection. Experimental results on our dataset and benchmark datasets (namely, ACPR 2019, ICPR 2018 FCD and IMEI datasets) show that the proposed method outperforms existing methods in terms of classification rate.

关键词： image color analysis Chebyshev approximation Tools Benchmark testing Distortion Forgery Pattern recognition

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：