检索结果-内蒙古大学图书馆

2004 7th International Conference on Solid-State and Integrated Circuits Technology Proceedings, ICSICT 2004

作者： Xiong, Chengyi Tian, Jinwen Liu, Jian Institute of Pattern Recognition and Artificial Intelligence Key Laboratory of Eduaction Ministry for Image Processing and Intelligent Control Huazhong University of Science and Technology Wuhan 430074 China College of Electronic Information Engineering South-Center University for Nationalities Wuhan 430074 China

This paper proposes a novel fast architecture for two-dimensional discrete wavelet transform by using lifting scheme. The parallel and embedded decimation techniques are employed to optimize the architecture, which is mainly composed of two horizontal filter modules and one vertical filter module, working in parallel and pipeline fashion with 100% hardware utilization. The architecture is designed to generate two outputs at one working clock cycle, with every two subbands coefficients alternately. The total time of computing J levels of decomposition for N×N image is approximately 2N2(I-4 J)/3 clock cycles. In comparison with the other devices reported in previous literature, the design has many advantages including lower hardware complexity and area and power efficient. The design is also fast, regular and simple, as well as well suited for VLSI implementation. © 2004 IEEE.

关键词： VLSI circuits

来源：评论

学校读者我要写书评

暂无评论

Edge preserving post-processing of wavelet compressed image

Edge preserving post-processing of wavelet compressed image

引用

International Conference on Signal processing Proceedings (ICSP)

作者： Yihua Tan Jinwen Tian Jian Liu State Key Laboratory of Education Ministry for Image Processing and Intelligent Control Huazhong University of Science and Technology Wuhan China

This paper proposes a new post-processing algorithm with edge preserving. To control the granular and edge preserving in the process of post-processing, we introduce a new potential function. To avoid the non-linear obstacle, the object energy is converted with half-quadratic regularization. At last, the implementation of the algorithm is improved to speed up the computation. Experiments show our method can achieve the design object that effectively remove artifacts at the same time maintain the edges.

关键词： image coding Degradation image restoration Discrete cosine transforms Decoding Bit rate Quantization image reconstruction Laboratories image processing

来源：评论

学校读者我要写书评

暂无评论

The cook projection index estimation using the wavelet kernel function

引用

International Symposium on Neural Networks, ISNN 2004

作者： Lin, Wei Zheng, Tian He, Fan Wen, Xian-Bin Department of Applied Mathematics Northwestern Polytechnical University Xi’an710072 China Department of Computer Science and Technology Northwestern Polytechnical University Xi’an710072 China Key Laboratory of Education Ministry for Image Processing and Intelligent Control Huazhong University of Science and Technology Wuhan430074 China

ISBN: (纸本)3540228411

The key procedure of exploratory projection pursuit is to optimize a criterion function, which is called the projection pursuit index. The cook family index estimated by the wavelet kernel function is given in this paper. And the asymptotic unbiasedness and the convergence property of the projection index are proved. Also, as the fast computing of this kind projection index, it is suited for the processing of a large data. Some results of projection index based on the wavelet kernel estimation are compared with that of the Gauss kernel estimation. © Springer-Verlag Berlin Heidelberg 2004.

关键词： Artificial intelligence

来源：评论

学校读者我要写书评

暂无评论

One efficient facet-based small target detection technique

One efficient facet-based small target detection technique

引用

International Conference on Signal processing Proceedings (ICSP)

作者： Sheng Zheng ChengYi Xiong JinWen Tian Jian Liu State Education Commission Key Laboratory of Image Processing and Intelligent Control Institute of Pattern Recognition and Artificial Intelligence Huazhong University of Science and Technology Wuhan China

This paper presents a method for the detection of small objects from the infrared images. The detection is performed on the intensity surface well fitted by the cubic facet model. The small target energy distribution presents as a convex surface on the image intensity surface and the target center is the maximal extremum points of the convex surface. According to the extremum theory, the possible small target position is analytically determined by directly convolving the original image with the derivative operators deduced from the bivariate cubic function. With the available coarse target locations, the potential target is separated from the background by examining the intensity features of the target cluster. Experimental results on the sample infrared images demonstrate the proposed algorithm provides a robust and efficient performance.

关键词： Face detection Object detection Target tracking Infrared detectors Surface topography Infrared imaging Maintenance intelligent control Pattern recognition Artificial intelligence

来源：评论

学校读者我要写书评

暂无评论

An SVM-based small target segmentation and clustering approach

An SVM-based small target segmentation and clustering approa...

引用

International Conference on Machine Learning and Cybernetics (ICMLC)

作者： Sheng Zheng Jian Liu Jin-Wen Tian State Education Commission Key Laboratory for Image Processing and Intelligent Control Institute for Pattern Recognition and Artificial Intelligence Huazhong University of Science and Technology Wuhan China

ISBN: (纸本)0780384032

Segmentation and clustering of infrared small target images in a sky or sea-sky background is considered in this paper, which is the preprocessing part of the detection and recognition of the moving small targets in an infrared image sequence. The infrared image intensity surface is well fitted by the least squares support vector machines (LS-SVM), and then the maximum extremum points are detected on the well fitted intensity surface by convolving the image with the second order directional derivative operators deduced from the mapped LS-SVM with mixtures of kernels. With the coarse locations, the possible targets are extracted by the clustering analysis. The computer experiments are carried out for the real and simulated sky and sea-sky infrared images. The experimental results demonstrate the proposed approach is effective.

关键词： Infrared detectors Infrared imaging image segmentation Sea surface image recognition Target recognition image sequences Least squares methods Support vector machines Kernel

来源：评论

学校读者我要写书评

暂无评论

controlling the false positive detection rate in fuzzy clustering of fMRI data

Controlling the false positive detection rate in fuzzy clust...

引用

IEEE International Symposium on Biomedical Imaging

作者： H. Jahanian H. Soltanian-Zadeh G.A. Hossein-Zadeh Signal and Image Processing Group School of Intelligent Systems Institute of Studies in Theoretical Physics and Mathematics Tehran Iran Elec. Eng. Department Control and Intelligent Processing Center of Excellence University of Tehran Tehran Iran Image Analysis Laboratory Radiology Department Henry Ford Health System Detroit MI USA

Despite its potential advantages for fMRI analysis, fuzzy C-means (FCM) clustering suffers from limitations such as the need for a priori knowledge of the number of clusters, and unknown statistical significance and instability of the results. We propose a randomization-based method to control the false positive rate and estimate statistical significance of the FCM results. Using this novel approach, we develop an fMRI activation detection method. The ability of the method in controlling the false positive rate is shown by analysis of false positives in activation maps of resting-state fMRI data. controlling the false positive rate in FCM allows comparison of different fuzzy clustering methods, using different feature spaces, to other fMRI detection methods. In this paper, using simulation and real fMRI data, we compare a novel feature space that takes the variability of the hemodynamic response function into account (HRF-based feature space) to the conventional cross-correlation analysis and FCM using the cross-correlation feature space.

关键词： Fuzzy control Clustering methods Hemodynamics Clustering algorithms Brain Signal processing image processing intelligent systems Process control control systems

来源：评论

学校读者我要写书评

暂无评论

Multiscale centerline extraction of angiogram vessels using gabor filters

Lecture Notes in Computer Science (including subseries Lectu...

引用

Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) 2004年 3314卷 570-575页

作者： Sang, Nong Tang, Qiling Liu, Xiaoxiao Weng, Wenjie Institute for Pattern Recognition and Artificial Intelligence Huazhong University of Science and Technology Wuhan 430074 China Key Laboratory of Ministry of Education for Image Processing and Intelligent Control Huazhong University of Science and Technology Wuhan 430074 China

In this paper, we propose a new automated approach to extract the centerlines from 2-D angiography. The centerline extraction is the basis of 3-D reconstruction of the blood vessels, so the accurate localization of centerlines counts for much. The characteristics of multiscale Gabor even filter, flexible frequency bands and enhancement effects, are fully utilized to detect centerlines of the blood vessels in various size. © Springer-Verlag 2004.

关键词： Blood vessels

来源：评论

学校读者我要写书评

暂无评论

An efficient algorithm and implementation for residue to binary number conversion

An efficient algorithm and implementation for residue to bin...

引用

International Conference on Communications, Circuits and Systems (ICCCAS)

作者： Chengyi Xiong Zhirong Gao Jinwen Tian Key Laboratory of Education Commission for Image Processing and Intelligent Control Institute of Pattern Recognition & Artificial Intelligence Huazhong University of Science and Technology Wuhan China Department of Computer Science Wuhan University of Science & Engineering Wuhan China Key Laboratory of Education Commission for Image Processing Huazhong University of Science & Technology China

The residue number system (RNS) has computational advantages in addition and multiplication compared with weighted number systems, such as the binary number system (BNS), since operations on residue digits are performed independently and these processes can be performed in parallel. Thus they are widely used in digital signal processing etc. Since residue to binary conversion is critical and difficult for the practicality of RNS, in this paper, a novel residue to binary (R/B) conversion algorithm for the restricted moduli set (2/sup n/ -1, 2/sup n/, 2n+1), based on exploring the periodicity of modulo (2/sup n/ /spl plusmn/ 1) operations is presented. A new 2n-bit adder based R/B converter is also proposed. The performance comparison results demonstrate that the new converter is faster and requires less area compared with the others reported in the previous literature.

关键词： Signal processing algorithms image converters Educational institutions Computer science Concurrent computing Digital signal processing Arithmetic Laboratories image processing intelligent control

来源：评论

学校读者我要写书评

暂无评论

Blurred image recognition based on complex moment invariants

Blurred image recognition based on complex moment invariants

引用

IEEE International Conference on image processing

作者： Zhang Tianxu Liu Jin Institute for Pattern Recognition and Artificial Intelligence Laboratory of Image Processing and Intelligent Control Huazhong University of Science and Technology China Institute for Pattern Recognition and Artificial Intelligence Huazhong University of Science and Technology China

ISBN: (纸本)0780385543

An important class of radiometric degradations we are faced with often in practice is image blurring. Special attention is paid to the recognition of the blurred image by moment invariant approach. Some important rules of complex moments for the blurred image are presented. Based on these rules, a useful subset of moment invariants is introduced, that are not affected by the blur, rotation, scale, and translation of the images. The experiments have shown that these invariants can be successfully used in recognition of the blurred image.

关键词： image recognition Layout Convolution Radiometry Character recognition Target recognition Computer vision Biomedical imaging Degradation Cameras

来源：评论

学校读者我要写书评

暂无评论

Clustering-based framework for comparing fMRI analysis methods

Clustering-based framework for comparing fMRI analysis metho...

引用

IEEE International Symposium on Biomedical Imaging

作者： G.-A. Hossein-Zadeh A.-M. Golestani H. Soltanian-Zadeh Control and Intelligent Processing Center of Excellence Electrical and Computer Engineering Department Faculty of Engineering University of Tehran Tehran Iran Image Analysis Laboratory Radiology Department Henry Ford Health System Detroit MI USA

In this paper, a cluster-based framework is introduced for comparing analysis methods of functional magnetic resonance images (fMRI). In the proposed framework, fMRI data is replaced with a feature space and each method considered as a clustering method in the new space. As a result, different methods can be compared by means of a cluster validity measure. The feature space is computed using a non-parametric method (principal component analysis-PCA). Four subjects have been analyzed with three methods and the proposed cluster-based framework has evaluated performance of the methods. The results are identical to those of the modified receiver operating characteristics (ROC). This validates the proposed approach.

关键词： Clustering algorithms Algorithm design and analysis image analysis Clustering methods Magnetic analysis Testing Accuracy Reproducibility of results Radiology Magnetic resonance

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：