检索结果-内蒙古大学图书馆

Convolutional deep-learning artificial neural networks

JOURNAL OF OPTICAL TECHNOLOGY 2015年第8期82卷 499-508页

作者： Lutsiv, v. P. St Petersburg State Univ Aerosp Instrumentat St Petersburg Russia

This paper discusses the history of the appearance and development of the concept of convolutional artificial neural networks, which, because they use a learning technology based on back-propagation of the error signal, have become one of the most efficient tools of automatic image classification. Along with the possibilities of modern convolutional neural networks in the area of shape classification of objects, the features of their use in analyzing the information of other hierarchical levels have also been analyzed-from the classification of textures to the structural decomposition of images, based on the formation of attention zones. Convolutional networks are considered in close association with a description of their natural analogs, found in the neural ensembles of living visual systems. (C) 2015 Optical Society of America.

关键词： image analysis image processing image processing algorithms image recognition devices Neural networks visual system

来源：评论

学校读者我要写书评

暂无评论

Parallel Implementation of FCM-Based volume Segmentation of 3D images

Parallel Implementation of FCM-Based Volume Segmentation of ...

引用

IEEE/ACS International Conference of Computer systems and Applications

作者： Shadi AlZu'bi Moahmmed A. Shehab Mahmoud Al-Ayyoub Elhadj Benkhelifa Yaser Jararweh Zaytoonah University of Jordan Jordan University of Science and Technology Staffordshire University

ISBN: (纸本)9781509043217

Parallel programming has many benefits that can help developers and researchers to improve the performance of some algorithms to become more efficient in real life. This is especially true for systems involving medical images. image segmentation for volume extraction is a famous segmentation process that takes long time to finish execution. In this paper, we consider a new version of the Fuzzy C-Means (FCM) segmentation algorithm (known as IT2FPCM) and provide a parallel implementation of it that is 12X time faster than the sequential implementation. The considered algorithm is based on Interval Type-2 FCM and combines fuzzy and possibilistic ideas in order to obtain higher accuracy. We conduct our experiments using two different machines and the results show that the improvement gains for both machines 11X and 12X, respectively.

关键词： Clustering algorithms Mathematical model image segmentation Three-dimensional displays Graphics processing units Phase change materials Linear programming

来源：评论

学校读者我要写书评

暂无评论

LEARNING A LOW-RANK SHARED DICTIONARY FOR OBJECT CLASSIFICATION

LEARNING A LOW-RANK SHARED DICTIONARY FOR OBJECT CLASSIFICAT...

引用

IEEE International Conference on image processing

作者： Tiep H. vu vishal Monga Pennsylvania State University

ISBN: (纸本)9781467399623

Despite the fact that different objects possess distinct class-specific features, they also usually share common patterns. Inspired by this observation, we propose a novel method to explicitly and simultaneously learn a set of common patterns as well as class-specific features for classification. Our dictionary learning framework is hence characterized by both a shared dictionary and particular (class-specific) dictionaries. For the shared dictionary, we enforce a low-rank constraint, i.e. claim that its spanning subspace should have low dimension and the coefficients corresponding to this dictionary should be similar. For the particular dictionaries, we impose on them the well-known constraints stated in the Fisher discrimination dictionary learning (FDDL). Further, we propose a new fast and accurate algorithm to solve the sparse coding problems in the learning step, accelerating its convergence. The said algorithm could also be applied to FDDL and its extensions. Experimental results on widely used image databases establish the advantages of our method over state-of-the-art dictionary learning methods.

关键词： Dictionaries Training Sparse matrices Encoding Learning systems Optimization Signal processing algorithms

来源：评论

学校读者我要写书评

暂无评论

Improving and Assessing Planet Sensitivity of the GPI Exoplanet Survey with a Forward Model Matched Filter

引用

ASTROPHYSICAL JOURNAL 2017年第1期842卷 14-14页

作者： Ruffio, Jean-Baptiste Macintosh, Bruce Wang, Jason J. Pueyo, Laurent Nielsen, Eric L. De Rosa, Robert J. Czekala, Ian Marley, Mark S. Arriaga, Pauline Bailey, vanessa P. Barman, Travis Bulger, Joanna Chilcote, Jeffrey Cotten, Tara Doyon, Rene Duchene, Gaspard Fitzgerald, Michael P. Follette, Katherine B. Gerard, Benjamin L. Goodsell, Stephen J. Graham, James R. Greenbaum, Alexandra Z. Hibon, Pascale Hung, Li-Wei Ingraham, Patrick Kalas, Paul Konopacky, Quinn Larkin, James E. Maire, Jerome Marchis, Franck Marois, Christian Metchev, Stanimir Millar-Blanchaer, Maxwell A. Morzinski, Katie M. Oppenheimer, Rebecca Palmer, David Patience, Jennifer Perrin, Marshall Poyneer, Lisa Rajan, Abhijith Rameau, Julien Rantakyro, Fredrik T. Savransky, Dmitry Schneider, Adam C. Sivaramakrishnan, Anand Song, Inseok Soummer, Remi Thomas, Sandrine Wallace, J. Kent Ward-Duong, Kimberly Wiktorowicz, Sloane Wolff, Schuyler Stanford Univ Kavli Inst Particle Astrophys & Cosmol Stanford CA 94305 USA Univ Calif Berkeley Dept Astron Berkeley CA 94720 USA Space Telescope Sci Inst 3700 San Martin Dr Baltimore MD 21218 USA SETI Inst Carl Sagan Ctr 189 Bernardo Ave Mountain View CA 94043 USA NASA Ames Res Ctr Mountain View CA 94035 USA Univ Calif Los Angeles Dept Phys & Astron Los Angeles CA 90095 USA Univ Arizona Lunar & Planetary Lab Tucson AZ 85721 USA NAOJ Subaru Telescope 650 North Aohoku Pl Hilo HI 96720 USA Univ Toronto Dunlap Inst Astron & Astrophys Toronto ON M5S 3H4 Canada Univ Georgia Dept Phys & Astron Athens GA 30602 USA Univ Montreal Inst Rech Exoplanetes Montreal PQ H3C 3J7 Canada Univ Grenoble Alpes CNRS IPAG F-38000 Grenoble France Univ Victoria 3800 Finnerty Rd Victoria BC V8P 5C2 Canada Natl Res Council Canada Herzberg 5071 West Saanich Rd Victoria BC V9E 2E7 Canada Gemini Observ 670 N Aohoku Pl Hilo HI 96720 USA Univ Michigan Dept Astron Ann Arbor MI 48109 USA Gemini Observ Casilla 603 La Serena Chile Large Synopt Survey Telescope 950N Cherry Ave Tucson AZ 85719 USA Univ Calif San Diego Ctr Astrophys & Space Sci La Jolla CA 92093 USA Univ Western Ontario Ctr Planetary Sci & Explorat Dept Phys & Astron London ON N6A 3K7 Canada SUNY Stony Brook Dept Phys & Astron Stony Brook NY 11794 USA CALTECH Jet Prop Lab Pasadena CA 91125 USA Univ Arizona Steward Observ Tucson AZ 85721 USA Amer Museum Nat Hist Dept Astrophys New York NY 10024 USA Lawrence Livermore Natl Lab Livermore CA 94551 USA Arizona State Univ Sch Earth & Space Explorat POB 871404 Tempe AZ 85287 USA Cornell Univ Sibley Sch Mech & Aerosp Engn Ithaca NY 14853 USA UC Santa Cruz Dept Astron 1156 High St Santa Cruz CA 95064 USA Johns Hopkins Univ Dept Phys & Astron Baltimore MD 21218 USA

We present a new matched-filter algorithm for direct detection of point sources in the immediate vicinity of bright stars. The stellar point-spread function (PSF) is first subtracted using a Karhunen-Love image processing (KLIP) algorithm with angular and spectral differential imaging (ADI and SDI). The KLIP-induced distortion of the astrophysical signal is included in the matched-filter template by computing a forward model of the PSF at every position in the image. To optimize the performance of the algorithm, we conduct extensive planet injection and recovery tests and tune the exoplanet spectra template and KLIP reduction aggressiveness to maximize the signalto- noise ratio (S/N) of the recovered planets. We show that only two spectral templates are necessary to recover any young Jovian exoplanets with minimal S/N loss. We also developed a complete pipeline for the automated detection of point-source candidates, the calculation of receiver operating characteristics (ROC), contrast curves based on. false positives, and completeness contours. We process in a uniform manner more than 330 data sets from the Gemini Planet imager Exoplanet Survey and assess GPI typical sensitivity as a function of the star and the hypothetical companion spectral type. This work allows for the first time a comparison of different detection algorithms at a survey scale accounting for both planet completeness and false-positive rate. We show that the new forward model matched filter allows the detection of 50% fainter objects than a conventional cross-correlation technique with a Gaussian PSF template for the same false-positive rate.

关键词： instrumentation: adaptive optics methods: statistical planetary systems surveys techniques: high angular resolution techniques: image processing

来源：评论

学校读者我要写书评

暂无评论

Suppressing wrinkles and enhancing human face using L0 norm and ACO algorithm

Suppressing wrinkles and enhancing human face using L0 norm ...

引用

International Conference on (ISCO) Intelligent systems and Control

作者： Rajesh Kochher Shreya Goyal Deptt.of Computer Science Engineering D.A.V.I.E. T Jalandhar (Pb.) Deptt. of Information Technology D.A.V.I.E. T Jalandhar (Pb.)

ISBN: (纸本)9781467378086

In digital image processing various filters are provided to enhance the quality of image. visibility of some parts in images are enhanced while suppressing the information in some parts of it. Fei Kou et al had found that L 0 norm is a useful mathematical formula to enhance the quality of the image by amplifying the edges present in the detail layer. However in this the problem of suppressing the desired information is not taken into account. In this letter, we are presenting L 0 norm and ant colony optimisation based filter which not only enhance the quality of the image but also suppress the unwanted wrinkles from the human faces, thus using the image enhancement algorithm from both perspective i.e. to enhance the quality of the detail layer as well as to suppress the unwanted information. Simulation results shows that the signal quality is improved and mean square error is considerably reduced using this approach. Moreover histogram equalisation of the image also get improved.

关键词： Face Filtering algorithms Information filters Optimization Mean square error methods image enhancement

来源：评论

学校读者我要写书评

暂无评论

An efficient image aesthetic analysis system using Hadoop

引用

SIGNAL processing-image COMMUNICATION 2015年 39卷 499-508页

作者： Wang, Weining Zhao, Weijian Cai, Chengjia Huang, Jiexiong Xu, Xiangmin Li, Lei S China Univ Technol Sch Elect & Informat Engn Guangzhou 510641 Guangdong Peoples R China

Assessing aesthetic appeal of images is a highly subjective task and has attracted a lot of research interests recently. Prior researchers have developed several aesthetic analysis systems on standalone computers. However, it is challenging to run the algorithms on mobile devices since the process of aesthetic analysis is quite complicated and time-consuming, especially for large amount of images. Hadoop is a popular technology for big data processing on cloud to offload computing burden from terminals. However it has NOT been used on image aesthetic yet. In this paper, we present an image aesthetic analysis system based on Hadoop framework to provide an efficiency solution and better user experience. We address several major problems: (1) adapt MapReduce for image data format and aesthetic analysis algorithms;(2) improve computing performance for large amount of small image files;(3) design a dynamic scheduling mechanism to optimize concurrent multiple users' requests;(4) design an effective commutation service between cloud and terminals. Experimental results demonstrate significant performance improvements with our system. At the same time, the system efficiency increases linearly with the expansion of the slaves in Hadoop. (C) 2015 Elsevier B.v. All rights reserved.

关键词： image aesthetic analysis Distributed processing MapReduce Hadoop

来源：评论

学校读者我要写书评

暂无评论

Glaucoma Diagnosis by Means of Optic Cup Feature Analysis in Color Fundus images

Glaucoma Diagnosis by Means of Optic Cup Feature Analysis in...

引用

European Signal processing Conference

作者： Andres Diaz Sandra Morales valery Naranjo Pablo Alcocer Aitor Lanzagorta Instituto de Investigacion e Innovacion en Bioingenieria I3B Universitat Politecnica de Valencia Grupo Tecnologias de Informatica Aplicadas a la Oftalmologia

ISBN: (纸本)9781509018918

Glaucoma is an asymptomatic eye disease and one of the major causes of irreversible blindness worldwide. For this reason, there have been significant advances in automatic screening tools for early detection. In this paper, an automatic glaucoma diagnosis algorithm based on retinal fundus image is presented. This algorithm uses anatomical characteristics such as the position of the vessels and the cup within the optic nerve. Using several color spaces and the Stochastic Watershed transformation, different characteristics of the optic nerve were analyzed in order to distinguish between a normal and a glaucomatous fundus. The proposed algorithm was evaluated on 53 images (24 normal and 29 glaucomatous images). The specificity and sensitivity obtained by the proposed algorithm are 0.81 and 0.87 using Luv color space, which means considerable performance in diagnosis systems.

关键词： Optical imaging image color analysis Optical signal processing Signal processing algorithms Biomedical optical imaging image segmentation Optical sensors

来源：评论

学校读者我要写书评

暂无评论

Robust non-local stereo matching for outdoor driving images using segment-simple-tree

引用

SIGNAL processing-image COMMUNICATION 2015年 39卷 173-184页

作者： Cuong Cao Pham vinh Quang Dinh Jeon, Jae Wook Sungkyunkwan Univ Coll Informat & Commun Engn Suwon South Korea

Non-local cost aggregation has recently emerged as a promising approach for stereo-matching and has attracted much interest over the past few years. Most non-local algorithms are reportedly better than state-of-the-art local algorithms for high-quality indoor images. However, the accuracy of non-local algorithms is still limited for outdoor images. Computing disparity maps for outdoor images in driver assistance systems is one of the most actively researched topics in the field of stereo vision. In this paper, we present a robust non-local stereo matching algorithm that improves the performance of non-local approaches for outdoor driving images. The proposed algorithm is inspired by the non-local cost aggregation method based on a minimum spanning tree, and it improves the estimation accuracy by introducing an alternate, effective segment-simple-tree that is more adequate for outdoor driving images than the minimum spanning tree. Experimental results showed that the proposed algorithm is superior to the existing local and non-local algorithms, and is comparable to semi-global matching. (C) 2015 Elsevier B.v. All rights reserved.

关键词： Stereo matching Non-local cost aggregation Tree structure Driver assistance systems

来源：评论

学校读者我要写书评

暂无评论

MULTI-SOURCE DOMAIN ADAPTATION USING C^1-SMOOTH SUBSPACES INTERPOLATION

MULTI-SOURCE DOMAIN ADAPTATION USING C^1-SMOOTH SUBSPACES IN...

引用

IEEE International Conference on image processing

作者： Jorge Batista Krzysztof Krakowski Luis Machado Pedro Martins Fatima Silva Leite Institute of Systems and Robotics

ISBN: (纸本)9781467399623

Manifold-based domain adaptation algorithms are receiving increasing attention in computer vision to model distribution shifts between source and target domain. In contrast to early works, that mainly explore intermediate subspaces along geodesics, in this work we propose to interpolate subspaces through C~1-smooth curves on the GraBmann manifold. The new methodis based on the geometric Casteljau algorithm that is used to generate smooth interpolating polynomial curves on non-euclidean spaces and can be extended to generate polynomial splines that interpolate a given set of data on the Grassmann manifold. To evaluate the usefulness of the proposed interpolating curves on vision related problems, several experiments were conducted. We show the advantage of using smooth subspaces interpolation in multi-source unsupervised domain adaptation problems and in object recognition problems across datasets.

关键词： Multi-Source domain adaptation Grassmann manifolds Casteljau algorithm Subspaces interpolation

来源：评论

学校读者我要写书评

暂无评论

A PWM-Mode Pixel-Parallel image-processing Circuit Performing Directional State-Propagation and Its Application to Subjective Contour Generation

引用

CIRCUITS systems AND SIGNAL processing 2015年第2期34卷 605-623页

作者： Kim, Youngjae Morie, Takashi Kyushu Inst Technol Grad Sch Life Sci & Syst Engn Kitakyushu Fukuoka 8080196 Japan

As an approach to constructing highly organized and intelligent image recognition systems mimicking the visual architecture in the human brain, we have designed an image processing circuit that performs a directional pixel-state propagation algorithm based on a pixel-parallel architecture using pulse-width modulation (PWM) signals. Directional pixel-state propagation can be used for subjective contour generation, which is a typical function of the visual system in the human brain to fill gaps in natural-image information. A PWM-mode circuit can provide more compact circuit configuration than digital circuits, and can achieve focal-plane analog processing by combining with image sensors without A/D converters. Such circuits are suitable for repetitive algorithms, which require offset-free arithmetic. The proof-of-concept LSI chip has been designed and fabricated using a 0.25 m mixed-signal CMOS process. There are 35 35 processing units included in the chip, the operation cycle time for propagation step is 2.2 s, and the power consumption is 363 mW at a supply voltage of 3.3 v. In experiments using this LSI chip, we have successfully generated subjective contours of Kanizsa figures.

关键词： PWM Pixel-parallel architecture Focal-plane analog processing Directional pixel-state propagation Subjective contour generation CMOS LSI

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：