检索结果-内蒙古大学图书馆

2014 International Conference on Communication technology and Application, CTA 2014

作者： Yang, Z.H. Dong, P. Institute of Image Processing and Pattern Recognition North China University of Technology Beijing China

ISBN: (纸本)9781845649302

This paper presents a novel approach for finding more accurate feature pairs which is not only invariant to affine transformation, but also deals with images with repetitive shapes. First, the more accurate and robust homographic transformation between different views could be calculated through bi-directional matching and appropriate selection of threshold. Second, an affine geometric property, ratios of areas of corresponding triangles are affine invariant, is used to obtain more feature pairs based on the first step. Experiments demonstrate that the proposed method outperforms the state-of-art ASIFT in the number of correct feature pairs and matching accuracy among images with repetitive patterns. © 2014 WIT Press.

关键词： Affine invariance ASIFT Feature points image matching Repetitive shapes

来源：评论

学校读者我要写书评

暂无评论

Privacy-preserving public auditing scheme for shared data with supporting multi-function

引用

Journal of Communications 2015年第7期10卷 535-542页

作者： Zhang, Jianhong Zhao, Xubing College of Sciences North China University of Technology China Institute of Image Process and Pattern Recognition North China University of Technology Beijing China

The most common concerns for users in cloud storage are data integrity, confidentiality and availability, so various data integrity auditing schemes for cloud storage have been proposed in the past few years, some of which achieve privacy-preserving public auditing, data sharing and group dynamic, or support data dynamic. However, as far as we know, until now yet there doesn’t exist a practical auditing scheme which can simultaneously realize all the functions above;In addition, in all the existing schemes, Block Authentication Tag (BAT) is adopted by data owner to achieve data integrity auditing;nevertheless, it’s a arduous task to compute BATs for the resource-constrained data owner. In this paper, we propose a novel privacy-preserving public auditing scheme for shared data in the cloud, which can also support data dynamic operations and group dynamic. Our scheme has the following advantages:(1) we introduce proxy signature into the existing auditing scheme to reduce the cloud user’s computation burden;(2) by introducing a Lagrange interpolating polynomial, our scheme realizes the identity’s privacy-preserving without increasing computation cost and communication overhead, moreover it makes group dynamic simple;(3) it can realize the practical and secure dynamic operations of shared data bycombining the Merkle Hash Tree and index-switch table which is built by us;(4) to protect the data privacy and resist the active attack, the cloud storage server hides the actual proof information by inserting its private key in producing proof information process. Theoretical analysis demonstrates our scheme’s security. © 2015 Journal of Communications.

关键词： Cloud storage

来源：评论

学校读者我要写书评

暂无评论

Microarray camera image super-resolution with neural network and fusion of V-System 13

Microarray camera image super-resolution with neural network...

引用

13th International Conference on Computer Science and Education, ICCSE 2018

作者： Zou, Jiancheng Zhang, Honggen North China University of Technology Institute of Image Processing and Pattern Recognition Beijing China

ISBN: (纸本)9781538654958

With the rapid development of artificial intelligence technology, more and more intelligent devices are beginning to be manufactured. At the same time, the demand of the image resolution is getting higher. Super-resolution of the image has become a research hotspot in the field of artificial intelligence. The same time, convolution neural networks have been successful in many fields of computer vision, including image super-resolution. Single image super-resolution has received a lot of attention and made good progress recently. However, there are few neural network models for multi-image super-resolution. Array images are a completely new experience and application for super-resolution. In this paper, we use the microarray camera (3∗3) to get nine pictures in the same scene and use nine images to realize multiple images super-resolution. A super-resolution model combining software with hardware is proposed. The structure of image registration, V- system image fusion and convolution operation are constructed. The super-resolution model can extract the feature of the target image more efficiently, and the array camera has the characteristic of trading space for the time change. The model can be more accurate and efficient than single image super-resolution. In the PSNR and SSIM indexes, 1.01dB and 0.03 are increased respectively. We use ISO12233 camera resolution chart to detect the experimental pictures. In the experimental results, the resolution of the algorithm is the best. © 2018 IEEE.

关键词： Cameras

来源：评论

学校读者我要写书评

暂无评论

Pose estimation based on human detection and segmentation

引用

Science in china(Series F) 2009年第2期52卷 244-251页

作者： CHEN Qiang ZHENG EnLiang LIU YunCai Institute of image Processing and Pattern Recognition Shanghai Jiao Tong University Shanghai 200240 China

We address the problem of 3D human pose estimation in a single real scene image. Normally, 3D pose estimation from real image needs background subtraction to extract the appropriate features. We do not make such assumption, In this paper, a two-step approach is proposed, first, instead of applying background subtraction to get the segmentation of human, we combine the segmentation with human detection using an ISM-based detector. Then, silhouette feature can be extracted and 3D pose estimation is solved as a regression problem. RVMs and ridge regression method are applied to solve this problem. The results show the robustness and accuracy of our method.

关键词： human detection and segmentation 3D pose estimation regression machine learning

来源：评论

学校读者我要写书评

暂无评论

Predicting Transmembrane β-barrel Segments with Chain Learning and Sparse Coding

引用

IFAC-PapersOnLine 2015年第28期48卷 245-250页

作者： Xi, Yin Hong-Bin, Shen Institute of Image Processing and Pattern Recognition Shanghai Jiao Tong University Shanghai200240 China

We developed a novel method named MemBrain-TMB to predict the spanning segments of transmembrane Â-barrel from amino acid sequence. MemBrain-beta is a statistical machine learningbased model, which is constructed using a new chain learning algorithm with the input features are encoded by the image sparse representation approach. To deal with the diverse loop length problem, we applied a dynamic threshold method, which is particularly useful for enhancing the recognition of short loops and tight turns. MemBrain-TMB achieves a Q2 accuracy of 93% and SOV of 97% on the benchmark dataset, which is 5%~10% higher than other existing predictors. © 2015

关键词： Learning algorithms Forecasting image processing Learning systems Machine learning pattern recognition Amino acid sequence Benchmark datasets Dynamic threshold Evaluation Input features Prediction methods Sparse representation Transmembranes

来源：评论

学校读者我要写书评

暂无评论

Solving the Euclidean Steiner Minimum Tree Using Cellular Stochastic Diffusion Search Algorithm

引用

Journal of Shanghai Jiaotong university(Science) 2011年第6期16卷 734-741页

作者：张瑾赵雅靓马良 Institute of Image Processing and Pattern Recognition Henan University School of Management University of Shanghai for Science and Technology

The Euclidean Steiner minimum tree problem is a classical NP-hard combinatorial optimization *** of the intrinsic characteristic of the hard computability,this problem cannot be solved accurately by efficient algorithms up to *** to the extensive applications in real world,it is quite important to find some heuristics for *** stochastic diffusion search algorithm is a newly population-based algorithm whose operating mechanism is quite different from ordinary intelligent algorithms,so this algorithm has its own advantage in solving some optimization *** paper has carefully studied the stochastic diffusion search algorithm and designed a cellular automata stochastic diffusion search algorithm for the Euclidean Steiner minimum tree problem which has low time *** results show that the proposed algorithm can find approving results in short time even for the large scale size,while exact algorithms need to cost several hours.

关键词： Euclidean Steiner minimum tree stochastic diffusion search cellular automata

来源：评论

学校读者我要写书评

暂无评论

Target tracking in infrared imagery using a novel particle filter

引用

Chinese Optics Letters 2009年第7期7卷 576-579页

作者：王芳林刘尔琦杨杰郁生阳周越 Institute of Image Processing and Pattern Recognition Shanghai Jiao Tong University Institute of the Second Academy China Aerospace Science and Industry Corporation

To address two challenging problems in infrared target tracking, target appearance changes and unpre- dictable abrupt motions, a novel particle filtering based tracking algorithm is introduced. In this method, a novel saliency model is proposed to distinguish the salient target from background, and the eigenspace model is invoked to adapt target appearance changes. To account for the abrupt motions efficiently, a two- step sampling method is proposed to combine the two observation models. The proposed tracking method is demonstrated through two real infrared image sequences, which include the changes of luminance and size, and the drastic abrupt motions of the target.

关键词： Infrared imaging Nonlinear filtering Targets

来源：评论

学校读者我要写书评

暂无评论

Non-photorealistic rendering for streamline stylization using sequential filters

Non-photorealistic rendering for streamline stylization usin...

引用

2012 4th International Conference on Intelligent Human-Machine Systems and Cybernetics, IHMSC 2012

作者： Chen, Xiaoguang Liu, Jiang Wu, Ran Institute of Image Processing and Pattern Recognition North China University of Technology Beijing 100144 China

ISBN: (纸本)9780769547213

Non-photorealistic rendering (NPR) such as stylization has received much concern in area of computer graphics and image processing. In this paper, a simple and effective method is proposed for streamline stylization. Through a kind of sequential filters, the method performs a series of operations, including detail removal, region smoothing, structure enhancement, and edge sharpen, to transform common pictures into Impressionist oil-painting like images with streamline style. Experimental results on a gallery of images demonstrate that our method can render images in good NPR style impressively and provide encouraging visual effects. © 2012 IEEE.

关键词： Bandpass filters

来源：评论

学校读者我要写书评

暂无评论

SHAPE NORMALIZATION THROUGH VISIBLE REGION CENTER AND INVISIBLE REGION CENTER

引用

pattern recognition LETTERS 1993年第5期14卷 407-414页

作者： SUN, Y QI, FH Institute of Image Processing and Pattern Recognition Shanghai Jiao Tong University Shanghai 200030 People's Republic of China

In this paper we present a new shape normalization method that is invariant to shape translation, rotation and scaling. We define a visible area density function and an unvisible area density function for a planar shape. Using these two functions we define a visible region center and an unvisible region center of the shape, respectively. When the visible and unvisible region centers of a shape locate at different positions they can be utilized as characteristic points to normalize the shape to a standard form. The normalizing process by use of the centers is presented. Experiments are executed on five groups of shapes with distortion of translation, rotation and scaling adding quantilization noise. The results show that the method is reasonable and available.

关键词： SHAPE REPRESENTATION SHAPE NORMALIZATION SHAPE CLASSIFICATION VISIBLE REGION CENTER INVISIBLE REGION CENTER

来源：评论

学校读者我要写书评

暂无评论

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

引用

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

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：