检索结果-内蒙古大学图书馆

IEEE International Conference on Control System, Computing and Engineering (ICCSCE)

作者： Munaf Rashid S. A. R. Abu-Bakar Musa Mokji Aliyu Abdu Micro Electronics and Computer Engineering Department Computer Vision Videos and image processing Laboratory (CVVIP) Universiti Teknologi Malaysia Johor Bahru Malaysia

Retrieving videos based on its contents is becoming an increasingly popular area of research, because of enormous growth in the availability of multimedia information on public databases like Google and YouTube. Usually videos contain large variety of data but majority of online videos contain human as a subject of interest. In this paper, a human action based video retrieval system is presented which can be used to retrieve videos based on the contents of the query. The proposed system can search videos containing particular action on large databases efficiently. Furthermore, it is also shown that by using features weight updating approach as a Relevance feedback (RF), it is possible to involve user concepts interactively so that complex human action queries can be searched quickly to achieve useful results. Three popular Human action datasets namely Weizmann, KTH and UCF (sports) have been utilized in order to validate the performance of the proposed system. Experimental results and simulations show the efficacy of the proposed system. Even with number of visual challenges proposed approach will manage to get better accuracy as compare to other existing methods.

关键词：

来源：评论

学校读者我要写书评

暂无评论

From early biological models to CNNs: do they look where humans look?

From early biological models to CNNs: do they look where hum...

引用

International Conference on Pattern Recognition

作者： Marinella Iole Cadoni Andrea Lagorio Enrico Grosso Tan Jia Huei Chan Chee Seng Computer Vision Laboratory University of Sassari Sassari Italy Faculty of Computer Science and Information Technology Center of Image and Signal Processing University of Malaya

Early hierarchical computational visual models as well as recent deep neural networks have been inspired by the functioning of the primate visual cortex system. Although much effort has been made to dissect neural networks to visualize the features they learn at the individual units, the scope of the visualizations has been limited to a categorization of the features in terms of their semantic level. Considering the ability humans have to select high semantic level regions of a scene, the question whether neural networks can match this ability, and if similarity with humans attention is correlated with neural networks performance naturally arise. To address this question we propose a pipeline to select and compare sets of feature points that maximally activate individual networks units to human fixations. We extract features from a variety of neural networks, from early hierarchical models such as HMAX up to recent deep convolutional neural netwoks such as Densnet, to compare them to human fixations. Experiments over the ETD database show that human fixations correlate with CNNs features from deep layers significantly better than with random sets of points, while they do not with features extracted from the first layers of CNNs, nor with the HMAX features, which seem to have low semantic level compared with the features that respond to the automatically learned filters of CNNs. It also turns out that there is a correlation between CNN's human similarity and classification performance.

关键词： Visualization Correlation Databases Neural networks Semantics Pipelines Visual systems

来源：评论

学校读者我要写书评

暂无评论

2D PCA-based shape prior for level sets segmentation framework of the vertebral body

2D PCA-based shape prior for level sets segmentation framewo...

引用

Cairo International Biomedical Engineering Conference, CIBEC

作者： Ahmed Shalaby Melih Aslan Hossam Abdelmunim Aly A. Farag Computer Vision and Image Processing Laboratory University of Louisville KY USA Computer & Systems Engineering Department Ain Shams University Cairo Egypt

In this paper, a novel statistical shape modeling method is developed for the vertebral body (VB) segmentation framework. Two-dimensional principle component analysis (2D-PCA) technique is exploited to extract the shape prior. The obtained shape prior is then embedded into the image domain to develop a new shape-based segmentation approach. Our framework consists of four main steps: i) shape model construction using 2D-PCA, ii) Detection of the VB region using the Matched filter, iii) Initial segmentation using the graph cuts which integrates the intensity and spatial interaction models, and iv) Registration of the shape prior and initially segmented region to obtain the final segmentation. The proposed method is validated on a Phantom as well as clinical CT images with various Gaussian noise levels. The experimental results show that the noise immunity and the segmentation accuracy of 2D-PCA based approach are much higher than conventional PCA approach.

关键词： Shape image segmentation Training Principal component analysis Accuracy Computed tomography Vectors

来源：评论

学校读者我要写书评

暂无评论

Gender estimation based on physiological features of the face

Gender estimation based on physiological features of the fac...

引用

International Symposium on Signal processing and Its Applications (ISSPA)

作者： C.H. Ting U.U. Sheikh S.A.R Abu-Bakar Computer Vision Video and Image Processing Research Laboratory Department of Microelectronics and Computer Engineering Faculty of Electrical Engineering Universiti Teknologi Malaysia Malaysia

In this paper, we present a gender estimation technique that will determine whether a person looking at the camera is a male or female. Several facial features extracted from a face are passed to post-image-processing operations before they are sent to a rule-base classifier. We applied our technique to both static passport-like photos as well as several video sequences consisting of a single face each. Our database also contains images of females wearing headscarf as well as bald men. Our initial results showed a very promising outcome. For static face images, we obtained an accuracy of 100% correct gender estimation for females and 98% for males while for video sequences, the accuracy obtained was 87.5% for females and 70% for males.

关键词： Face Hair image resolution Estimation computers Accuracy

来源：评论

学校读者我要写书评

暂无评论

Produce recognition system using data mining algorithm

Produce recognition system using data mining algorithm

引用

IEEE International Conference on Control System, Computing and Engineering (ICCSCE)

作者： J. K. Chaw M. M. Mokji Computer Vision Video and Image Processing (CvviP) Laboratory Department of Microelectronic and Computer Engineering Faculty of Electrical Engineering Universiti Teknologi Malaysia Malaysia

Produce recognition system is a system that can categorize types of vegetables and fruits based on features extracted from the images. However, there are numerous features that can be extracted from fruits and vegetables such as colour, texture and shape. As a result, it is effort consuming to identify suitable features ad hoc. Thus, data mining is required to discover the most discriminative features for recognition. This paper aims to extend the usage of data mining algorithm to image domain. Data mining algorithm is preferred to other feature selection algorithms because it discovers nuggets of knowledge that can be understood by human whereas classic feature selection techniques provide outputs that can only be managed by learning algorithms.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Maximum likelihood thresholding algorithm based on four-parameter gamma distributions

Maximum likelihood thresholding algorithm based on four-para...

引用

International Conference on Electrical Engineering, Computing Science and Automatic Control (CCE)

作者： Peter De-Ford Geovanni Martinez Image Processing and Computer Vision Research Laboratory (IPCV-LAB) Universidad de Costa Rica San José Costa Rica

In this contribution, we present a segmentation algorithm based on thresholding to subdivide an intensity image in the regions of object and background. The optimal threshold is found by maximizing a likelihood function derived from a novel intensity probability density function model, which consists of the sum of two weighted four-parameter gamma distributions, as a more flexible alternative to currently used models consisting of the sum of two weighted two-parameter Gaussian distributions. According to our experiments with 132 images, the proposed algorithm is in average slightly better than the best found in the scientific literature, performing particularly good in low contrast images. The additional parameters and complexity of its likelihood function resulted in an increase of the processing time by a factor of 3, from 0.003 sec/image to 0.009 sec/image.

关键词： image segmentation Probability density function Approximation algorithms Gaussian distribution Histograms Shape Pattern recognition

来源：评论

学校读者我要写书评

暂无评论

Model-based multiview stereo via level sets with statistical shape prior

Model-based multiview stereo via level sets with statistical...

引用

IEEE International Conference on image processing

作者： Moumen El-Melegy Nagi Al-Ashwal Aly A. Farag Electrical Engineering Department Assiut University Asyut Egypt Computer Vision and Image Processing Laboratory University of Louisville KY USA

In this paper, we propose a methodology to incorporate 3D shape prior information in multi-view stereo. This is important for applications that deal with specific category of objects. The methodology is based on a new formulation of a level-set based energy functional. Shape prior model is then embedded in the energy functional to allow the reconstruction of an object with shape variations consistent with the training model examples. Several experiments to evaluate the proposed methodology are presented.

关键词： Shape image reconstruction Level set Stereo vision Solid modeling Training Ellipsoids

来源：评论

学校读者我要写书评

暂无评论

Variational-based method to extract parametric shapes from images

Variational-based method to extract parametric shapes from i...

引用

International Conference on computer vision (ICCV)

作者： M.T. El-Melegy N.H. Al-Ashwal A.A. Farag Electrical Engineering Department Assiut University Asyut Egypt Computer Vision and Image Processing Laboratory University of Louisville KY USA

In this paper, we propose a variational method to segment image objects, which have a given parametric shape based on a level-set formulation of the Mumford-Shah functional, and the shape parameters. We define an energy functional composed by two complementary terms. The first one detects object boundaries using a Chan-Vese-like method. The second term constrains the contour to find a shape compatible with the parametric shape. The segmentation of the object of interest is given by the minimum of our energy functional. This minimum is computed with the calculus of variation and the gradient descent method that provide a system of evolution equations solved with the well-known level set method. We focus in this paper on the parametric category of image linear objects. Applications of the proposed model are presented on synthetic and real images

关键词： Shape image segmentation Level set computer vision Focusing image processing Object detection Calculus Application software Equations

来源：评论

学校读者我要写书评

暂无评论

Real-time vision-based approach for probe tracking in operating room

引用

International Congress Series 2004年 1268卷 467-472页

作者： Aly A. Farag M.Sabry Hassouna Ayman El-Baz S. Hushek Computer Vision and Image Processing Laboratory Electrical and Computer Engineering University of Louisville 403 Lutz Hall Bldg. Louisville KY 40292 USA Department of Neurological Surgery University of Louisville USA

A hand-held surgical probe is an essential component of any image-guided neurosurgery (IGNS). During the surgical procedure, the IGNS tracks the probe position and displays the anatomy beneath it as three orthogonal image slices on a workstation-based 3D imaging system. Existing IGNS systems use different tracking techniques including mechanical, optical, ultrasonic, and electromagnetic systems. In this paper, we present a new computational vision-based probe tracking technique, which provides both position and orientation. We also propose another new computational vision-based technique to track the patient head and thus compensate for its movement during probing procedure. The proposed system is completely passive, its stereo cameras detect the presence of the probe using available visible light, and reports on their exact location and orientation in patient physical space in real time. We have validated the system using a skull phantom and a hand made probe.

关键词： image-guided neurosurgery Head motion tracking Probe tracking vision tracking

来源：评论

学校读者我要写书评

暂无评论

Automatic detection and recognition of lung abnormalities in helical CT images using deformable templates

Automatic detection and recognition of lung abnormalities in...

引用

Medical image Computing and computer-Assisted Intervention, MICCAI 2004 - 7th International Conference, Proceedings

作者： Farag, Aly El-Baz, Ayman Gimel'Farb, Georgy G. Falk, Robert Hushek, Stephen G. Computer Vision and Image Processing Laboratory University of Louisville Louisville KY 40292 United States Department of Computer Science Tamaki Campus University of Auckland Auckland New Zealand Medical Imaging Division Jewish Hospital Louisville KY United States iMRI Department Norton Hospital Louisville KY United States

ISBN: (纸本)3540229779

Automatic detection of lung nodules is an important problem in computer analysis of chest radiographs. In this paper we propose a novel algorithm for isolating lung nodules from spiral CT scans. The proposed algorithm is based on using four different types of deformable templates describing typical geometry and gray level distribution of lung nodules. These four types are (z) solid spherical model of large-size calcified and non-calcified nodules appearing in several successive slices;(ii) hollow spherical model of large lung cavity nodules;(Hi) circular model of small nodules appearing in only a single slice;and (iv) semicircular model of lung wall nodules. Each template has a specific gray level pattern which is analytically estimated in order to fit the available empirical data. The detection combines the normalized cross-correlation template matching by genetic optimization and Bayesian post-classification. This approach allows for isolating abnormalities which spread over several adjacent CT slices. Experiments with 200 patients' CT scans show that the developed techniques detect lung nodules more accurately than other known algorithms. © Springer-Verlag Berlin Heidelberg 2004.

关键词： computer vision

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：