检索结果-内蒙古大学图书馆

Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC)

作者： Zhicheng Wang Jian Liu Jinwen Tian Zeping Xie State Education Commission Key Laboratory for Image Processing and Intelligent Control Institute for Pattern Recognition and Artificial Intelligence Huazhong Univ. of Sci. & Tech. Wuhan 430074 China Key Lab. for Image Process. & Intelligent Control State Educ. Comm. School of Life Science & Technology Huazhong University of Science and Technology Wuhan China

This paper introduces the new qualitative and quantitative methods, which can diagnose breast tumors. Qualitative methods include blood vessel display inside and outside of pathological changes part of breast, display of equivalent pixel curves at the part of pathological changes and display of breast tumor image edge. Accordingly, three feature extraction operators are proposed, i.e. the combination operators of anisotropic gradient and smoothing operator, an improved Sobel operator and an edge sharpening operator. Furthermore, quantitative diagnostic approaches are discussed based on blood and oxygen contents according to abundant clinical data and pathological mechanism of breast tumors. The results of clinic show that the methods of combining qualitative and quantitative diagnose are effective for breast tumor images, especially for early and potential breast cancer

关键词： Breast tumors Displays Pathology Blood vessels Biomedical imaging Pixel Feature extraction Anisotropic magnetoresistance Smoothing methods Breast cancer

来源：评论

学校读者我要写书评

暂无评论

A Parallel-based Lifting Algorithm and VLSI Architecture for DWT

引用

Journal of Electronics(China) 2006年第2期23卷 244-248页

作者： Xiong Chengyi Tian Jinwen Liu Jian Gao Zhirong The State Key Lab of Education Commission for Image Processing and Intelligent Control Institute of Pattern Recognition ＆ Artificial Intelligence Huazhong Univ. of Science ＆ Tech. Wuhan 430074 China College of Electronic Info. Eng. South- Center Univ. for Nationalities Wuhan 430074 China Dept of Computer Science Wuhan Univ. of Science ＆ Eng. Wuhan 430074 China

A novel Parallel-Based Lifting Algorithm （PBLA） for Discrete Wavelet Transform （DWT）, exploiting the parallelism of arithmetic operations in all lifting steps, is proposed in this paper. It leads to reduce the critical path latency of computation, and to reduce the complexity of hardware implementation as well. The detailed derivation on the proposed algorithm, as well as the resulting Very Large Scale Integration （VLSI） architecture, is introduced, taking the 9/7 DWT as an example but without loss of generality. In comparison with the Conventional Lifting Algorithm Based Implementation （Clab.）, the critical path latency of the proposed architecture is reduced by more than half from （4Tm ＋ 8Ta）to Tm ＋ 4Ta, and is competitive to that of Convolution-Based Implementation （CBI）, but the new implementation will save significantly in hardware. The experimental results demonstrate that the proposed architecture has good performance in both increasing working frequency and reducing area.

关键词： Discrete Wavelet Transform （DFT） Lifting scheme Parallel Very Large Scale Integration（VLSI）

来源：评论

学校读者我要写书评

暂无评论

Natural image classification in non linear network model

Natural image classification in non linear network model

引用

SPIE - The International Society for Optical Engineering

作者： Azhar, Hanif Iftekharuddin, Khan Kozma, Robert Admala, Abhinav Intelligent Systems and Image Processing Lab. Electrical and Computer Engineering Department University of Memphis TN 38152 United States Computer Science Department University of Memphis TN 38152 United States

We study the non-linear behavior of the KIII model for natural image classification. The KIII model is designed to be a dynamic computational model that simulates the sensory cortex. The KIII model has been explored for rudimentary pattern recognition and classification in noisy environment [1-3]. We extend the study of KIII models in understanding whether self-organized neural populations can be exploited into perceptual and memory producing systems such as in natural image classification. Our goal is to obtain a quantitative index on how well the KIII model behaves when it is assigned the task to identify and distinguish one class of natural image from the other based on color and texture features. For twenty training data, twenty validation data and eighty test data set for four image classes, we obtain 80% correct classification using the KIII. We compare a standard non linear neural network tools such as back propagation for the classification of the same set of natural images and obtain 65% correct classification. We conclude that dynamic neural computational models such as KIII may be suitable candidates for natural image classification.

关键词： image segmentation

来源：评论

学校读者我要写书评

暂无评论

A generic shape matching with anchoring of knowledge primitives of object ontology

引用

2nd International Conference on image Analysis and recognition, ICIAR 2005

作者： Han, Dongil You, Bum-Jae Kim, Yong Se Suh, Il Hong Vision and Image Processing Lab. Sejong University Seoul Korea Republic of Intelligent Robotics Research Center Korea Institute of Science and Technology Seoul Korea Republic of Creative Design and Intelligent Tutoring Systems Research Center Sungkyunkwan University Suwon Korea Republic of Intelligence and Interaction Lab. Hanyang University Seoul Korea Republic of

ISBN: (纸本)3540290699

We have developed a generic ontology of objects, and a knowledge base of everyday physical objects. Objects are represented as assemblies of functional features and their spatial relations. Generic shape information of objects and features is stored using a partial boundary representation. Form-function reasoning is applied to deduce geometric shape elements from a feature's functions. We have also developed a generic geometric shape based object recognition method which uses many local features. The proposed recognition method considers the concept of ontology for representation of generic functions of objects. And the use of a general shape-function reasoning with context understanding enhances the performance of object recognition. © Springer-Verlag Berlin Heidelberg 2005.

关键词： Object recognition

来源：评论

学校读者我要写书评

暂无评论

A Multi-homed VPN Architecture Based on Extended SOCKS+TLS Protocols

引用

Journal of Shanghai Jiaotong university(Science) 2005年第3期10卷 250-255页

作者：张宝良胡汉平陈翔 State Education Ministry Key Lab. of Image Information & Intelligent Control In st. of Pattern Recognition & Artificial Intelligence Huazhong Univ. of Science & Technology Wuhan 430074 China

A multi-homed VPN architecture based on extended SOCKSv5 and TLS was proposed. The architecture employs a dynamic connection mechanism for multiple proxies in the end system,i n which the security-demanded transmission connections can switch smoothly among the multiple proxies by maint aining a coherent connection *** mechanism is transparent to application programs and can support th e building of *** the cooperation of some other security components,the mechanism guarantees the reso urce availab.lity and reliability of the end system against some attacks to the specific ports or hosts.

关键词： multi-homed SOCKSv5 security connection context virtual private network （VPN）

来源：评论

学校读者我要写书评

暂无评论

Fast algorithm of discrete Walsta-Haar transformation

Fast algorithm of discrete Walsta-Haar transformation

引用

International Conference on Space Information Technology

作者： Lu, Li Shi, Baochang Tian, Jinwen Liu, Jian State Education Commission Key Lab for Image Processing and Intelligent Control Huazhong University of Science and Technology Wuhan Hubei 430074 China Inst. for Pattern Recognition and Artificial Intelligence Huazhong University of Science and Technology Wuhan Hubei 430074 China Department of Mathematics Huazhong University of Science and Technology Wuhan Hubei 430074 China

Walsh-Haar function system that was first intruoduced by us is a new kind of function systems, and has a good global/local property. This function system is called Walsh ordering function system since its generation kernel functions belong to Walsh ordering Walsh function system. We worked out a recursive property of the matrix WHKRm+1 corresponding to the first KR m+1 Walsh-Haar functions in Walsh-Haar function system, and we also proved that Walsh-Haar function system is perfect and orthogonal similar to Walsh function system and Haar function system. Thus, discrete Walsh-Haar transformation (DW-HT) is an orthogonal transformation that can be widely used in signal processing. In this paper, using the recursive property of the matrix WHKRm+1 and the fast algorithm of discrete Walsh transformation (DWT) in Walsh ordering, we have designed a fast algorithm of Walsh ordering DW-HT based on the bisection technique. The idea and method used in this paper can be used for designing fast algorithms of other ordering DW-HTs and other discrete orthogonal transformations.

关键词： Mathematical transformations

来源：评论

学校读者我要写书评

暂无评论

Multi-homed VPN architecture based on extended SOCKS+TLS protocols

引用

Journal of Shanghai Jiaotong University (Science) 2005年第3期10 E卷 250-255页

作者： Zhang, Bao-Liang Hu, Han-Ping Chen, Xiang State Education Ministry Key Lab. of Image Information and Intelligent Control Inst. of Pattern Recognition and Artificial Intelligence Huazhong Univ. of Science and Technology Wuhan 430074 China

A multi-homed VPN architecture based on extended SOCKSv5 and TLS was proposed. The architecture employs a dynamic connection mechanism for multiple proxies in the end system, in which the security-demanded transmission connections can switch smoothly among the multiple proxies by maintaining a coherent connection context. The mechanism is transparent to application programs and can support the building of VPN. With the cooperation of some other security components, the mechanism guarantees the resource availab.lity and reliability of the end system against some attacks to the specific ports or hosts.

关键词： Virtual reality

来源：评论

学校读者我要写书评

暂无评论

Multifeature distance map based detection of small infrared targets with low contrast in image sequences

Multifeature distance map based detection of small infrared ...

引用

International Conference on Space Information Technology

作者： Wang, Zhicheng Gao, Chenqiang Tian, Jinwen Liu, Jian Chen, Xiangping State Education Commission Key Lab for Image Processing and Intelligent Control Huazhong University of Science and Technology Wuhan Hubei 430074 China Inst. for Pattern Recognition and Artificial Intelligence Huazhong University of Science and Technology Wuhan Hubei 430074 China Electronic and Engineering Department Huazhong University of Science and Technology Wuhan Hubei 430074 China

Target detection techniques play an important role in automatic target recognition (ATR) systems because overall ATR performance depends closely on detection results. In this paper, a novel method for fusion detection of infrared weak targets based on multifeature distance map (MFDM) in image sequences is proposed. As for small weak targets, there are many features, such as local entropy, average gradient strength. These features depict the characteristics of small infrared targets and can be extracted. Multifeature-based fusion techniques are applied to detect such weak targets. The problem of detecting small targets is converted to search peak values in specified feature space where multifeature vectors space (MFVS) is considered. Distance map (DM) can be derived according to feature vectors and target detection is performed in DM. In order to accumulate energy of targets deeply and suppress background and clutters to a great extent, five distance maps obtained by corresponding five consecutive frames are utilized to fuse with average weight, which results in the fact that the contrast between targets and background including clutters are enlarged and that the feature peaks of targets are obvious different from background and clutters. After these steps, a contrast segmentation method is used to extract targets from complicated background on the fused DM. Actual infrared image sequences in background of sea and sky are applied to validate the proposed approach. Experimental results demonstrate the robustness of the proposed method with high performance.

关键词： Automatic target recognition

来源：评论

学校读者我要写书评

暂无评论

Knowledge-based segmentation: Using simultaneous shape priori and histogram information to segment brain structures

Knowledge-based segmentation: Using simultaneous shape prior...

引用

Seventh IASTED International Conference on Signal and image processing, SIP 2005

作者： Batmanghelich, Nematollah Soltanian-Zadeh, Hamid Araabi, Babak Najaar Elec. and Comp. Eng. Dept. Control and Intelligent Processing Center of Excellence University of Tehran Tehran Iran School of Cognitive Sciences Institute for Studies in Theoretical Physics and Mathematics Tehran Iran Image Analysis Lab. Radiology Dept. Henry Ford Health System Detroit MI United States

ISBN: (纸本)0889865183

In this paper, we propose a new method integrating both a priori shape information and our knowledge about gray levels of the desired structure. We describe an approach inspired from tracking to deal with non-uniform gray levels. We define focus region to consider both interior and exterior of the desired object. We utilize signed distance function to consider shape information. Embedding a priori shape and gray level knowledge in a statistical platform, we use correlation between changes in shape and histogram to improve the results. Our method successfully segments Thalamus and other brain structures.

关键词： image segmentation

来源：评论

学校读者我要写书评

暂无评论

Noise suppression of fMRI time-series in wavelet domain

Noise suppression of fMRI time-series in wavelet domain

引用

Seventh IASTED International Conference on Signal and image processing, SIP 2005

作者： Jahanian, Hesamoddin Soltanian-Zadeh, Hamid Hossein-Zadeh, Gholam Ali Elec. and Comp. Eng. Dept. Control and Intelligent Processing Center of Excellence University of Tehran Tehran 14399 Iran School of Cognitive Sciences Institute for Studies in Theoretical Physics and Mathematics Tehran Iran Image Analysis Lab. Radiology Dept. Henry Ford Health System Detroit MI United States

ISBN: (纸本)0889865183

Because of poor signal-to-noise ratio (SNR) of the fMRI time series and confounding effects, the results of fMRI analysis are often unsatisfactory. Existence of significant noise and artifacts in fMRI time-series as well as their unknown structure, complicates the problem of activation detection in the time domain. This makes the fMRI noise suppression a challenging problem. Based on some assumptions, different parametric denoising methods such as wavelet based denoising methods have been introduced in the literature. But these assumptions may not necessarily hold for the fMRI data. To remedy this problem, using randomization analysis, we propose a novel wavelet-based denoising method for fMRI analysis. The proposed denoising method is employed to build a feature space for fMRI cluster analysis and its efficiency is shown using simulated and experimental datasets.

关键词： Noise abatement

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：