检索结果-内蒙古大学图书馆

8th Asian Conference on computer vision

作者： Huang, Yonggang Wang, Yunhong Tan, Tieniu Chinese Acad Sci Inst Automat Natl Lab Pattern Recognit Beijing Peoples R China Beihang Univ Sch Comp Sci & Engn Beijing Peoples R China

ISBN: (纸本)9783540763895

In this paper, we propose an efficient 3D face recognition method based on statistics of range image differences. Each pixel value of range image represents normalized depth value of corresponding point on facial surface, and so depth differences between two range images' pixels of the same position on face can straightforwardly describe the differences between two faces' structures. Here, we propose to use histogram proportion of depth differences to discriminate intra and inter personal differences for 3D face recognition. Depth differences are computed from a neighbor district instead of direct subtraction to avoid the impact of non-precise registration. Furthermore, three schemes are proposed to combine the local rigid region(nose) and holistic face to overcome expression variation for robust recognition. Promising experimental results are achieved on the 3D dataset of FRGC2.0, which is the most challenging 3D database so far.

关键词： Face recognition

来源：评论

学校读者我要写书评

暂无评论

computer vision and digital inclusion of persons with special needs: Overview and state of art

Computer vision and digital inclusion of persons with specia...

引用

International Symposium on Computational Modelling of Objects Represented in Images (CompIMAGE 2006)

作者： Pistori, Hemerson Univ Catolica Dom Bosco Campo Grande Brazil

ISBN: (纸本)9780415433495

This survey paper addresses some issues related to the application of computer vision techniques to improve the welfare of people with special needs. The main problems and current work on topics like sign language processing and wheelchair control will be presented. The paper also introduces an ongoing project that aims at creating a free software environment that will include implementations of a large amount of computer vision, pattern recognition and machine learning techniques, tuned to the problems related to the digital inclusion of people with special needs. The software will also serve as an experimental environment, where new techniques will be implemented, tested and compared.

关键词： Learning systems

来源：评论

学校读者我要写书评

暂无评论

Segmentation of printed Farsi/Arabic words

Segmentation of printed Farsi/Arabic words

引用

5th IEEE/ACS International Conference on computer Systems and Applications (AICCSA-07)

作者： Broumandnia, A. Shanbehzadeh, J. Nourani, M. Islamic Azad Univ Tehran S Branch Tehran Iran Tarbiat Moallem Univ Tehran Iran Univ Tehran Tehran 14174 Iran

ISBN: (纸本)9781424410309

Characters connectivity is a problem in automated printed Farsi/Arabic script recognition. This paper introduces a novel scheme based on wavelet transform to solve segmentation of printed Farsi/Arabic words into characters. Our novel algorithm employs a new wavelet transform by which the extracted wavelet coefficients are exploited, in detecting, underlying horizontal edges and base line. Projection of horizontal edges and their location on base line provide the segmentation points. A classification method distinguishes true segmenting points. New algorithm is robust against noise, gray level, font and size of characters. Simulation results provide a comparison between new algorithm and three schemes, closed contour, structural and holistic, in terms Of precision, speed and robustness against Gaussian noise. Experimental Results indicate superiority of our scheme in terms of precision and show that new algorithm improves recognition speed by a factor of at least 2.5 times.

关键词： pattern recognition OCR image processing machine vision wavelet transform

来源：评论

学校读者我要写书评

暂无评论

Energy-based models in document recognition and computer vision

Energy-based models in document recognition and computer vis...

引用

9th International Conference on Document Analysis and recognition

作者： LeCun, Yann Chopra, Sumit Ranzato, Marc'Aurelio Huang, Fu-Jie NYU Courant Inst Math Sci New York NY 10011 USA

ISBN: (纸本)9780769528229

The Machine Learning and pattern recognition communities are facing two challenges: solving the normalization problem, and solving the deep learning problem. The normalization problem is related to the difficulty of training probabilistic models over large spaces while keeping them properly normalized. In recent years, the ML and Natural Language communities have devoted considerable efforts to circumventing this problem by developing "unnormalized" learning models for tasks in which the output is highly structured (e.g. English sentences). This class of models was in fact originally developed during the 90's in the handwriting recognition community, and includes Graph Tran former Networks, Conditional Random Fields, Hidden Markov SVMs, and Maximum Margin Markov Networks. We describe these models within the unifying framework, of "Energy-Based Models" (EBM). The Deep Learning Problem is related to the issue of training all the levels of a recognition system (e.g. segmentation, feature extraction, recognition, etc) in an integrated fashion. We first consider "traditional" methods for deep learning, such as convolutional networks and back-propagation, and show that, although they produce very low error rates for handwriting and object recognition, they require many training samples. We show that using unsupervised learning to initialize the layers of a deep network dramatically reduces the required number of training samples, particularly for such tasks as the recognition of everyday objects at the category level.

关键词： Object recognition

来源：评论

学校读者我要写书评

暂无评论

Stochastic local search for omnidirectional catadioptric stereovision design

Stochastic local search for omnidirectional catadioptric ste...

引用

3rd Iberian Conference on pattern recognition and Image Analysis

作者： Dequen, G. Devendeville, L. Mouaddib, E. Univ Picardie CNRS CREA LaRIA 33 Rue St Leu F-80039 Amiens 1 France

ISBN: (纸本)9783540728481

This paper deals with a compact catadioptric omnidirectional stereovision system based on a single camera and multi-mirrors (at least two mirrors). Many configurations were empirically designed in previous works with the aim to obtain a good 3D reconstruction accuracy. In this paper, we propose to use optimization techniques for omnidirectional catadioptric stereovision design, by using a stochastic local search method in order to find a good sensor (number, relative positions and sizes of mirrors). We explain principles of our approach and provide automatically designed sensors with a number of mirrors from two to nine. We finally simulate the 3D-reconstruction of a real environment modeled under a ray-tracing software with some of these sensors.

关键词： Stereo vision

来源：评论

学校读者我要写书评

暂无评论

Identification of drawing tools by classification of textural and boundary features of strokes

引用

pattern recognition LETTERS 2007年第6期28卷 710-718页

作者： Kammerer, Paul Lettner, Martin Zolda, Ernestine Sablatnig, Robert Vienna Univ Technol Pattern Recognit & Image Proc Grp Inst Comp Aided Automat A-1040 Vienna Austria

Recent developments in computer vision provide powerful tools for the examination and classification of data of our cultural heritage. it is generally recognized that the cultural heritage we are preserving for future generations will profit considerably from passing over to state of the art technologies. New camera hardware allows new insights into cultural heritage, especially if infrared cameras are concerned, since they allow the study of structures that are visually hidden. In this paper a strategy for the analysis of underdrawing strokes in infrared reflectograms is presented. Underdrawings are the basic concept or "primal sketch" of the artist before the complete painting is created. We focus on infrared reflectograms of medieval panel paintings, since their underdrawings are common and help art historians to study the school of the old masters. The purpose of the stroke analysis is the determination of the drawing tool used to draft the painting. This information allows significant support for a systematic stylistic approach in the analysis of paintings. Stroke segmentation in paintings is related to the extraction and recognition of handwriting, therefore similar techniques to segment the strokes from the background incorporating boundary information are used. Following the segmentation of single strokes, a classification of strokes with respect to the drawing tool used to create the strokes is performed. Two different classification methods, one texture-based and one based on active contour models are combined in order to improve the classification results, which are presented and discussed for strokes on selected test panels. (c) 2006 Elsevier B.V. All rights reserved.

关键词： panel paintings underdrawings stroke analysis textural features geometric features classification

来源：评论

学校读者我要写书评

暂无评论

Markov random field modeled level sets method for object tracking with moving cameras

引用

8th Asian Conference on computer vision

作者： Zhou, Xue Hu, Weiming Chen, Ying Hu, Wei Inst Automat Natl Lab Pattern Recognit Beijing Peoples R China

ISBN: (纸本)9783540763857

Object tracking using active contours has attracted increasing interest in recent years due to acquisition of effective shape descriptions. In this paper, an object tracking method based on level sets using moving cameras is proposed. We develop an automatic contour initialization method based on optical flow detection. A Markov Random Field (MRF)-like model measuring the correlations between neighboring pixels is added to improve the general region-based level sets speed model. The experimental results on several real video sequences show that our method successfully tracks objects despite object scale changes, motion blur, background disturbance, and gets smoother and more accurate results than the current region-based method.

关键词： Object recognition

来源：评论

学校读者我要写书评

暂无评论

Fast multi-scale template matching using binary features

Fast multi-scale template matching using binary features

引用

7th IEEE Workshop on Applications of computer vision, WACV 2007

作者： Tang, Feng Tao, Hai Department of Computer Engineering University of California Santa Cruz United States

ISBN: (纸本)0769527949

Template matching is one of the key problems in computer vision and has been widely used in tracking, recognition and many other applications. Traditional methods are usually slow because the template needs to be matched to every location in the image and the matching involves element-by-element floating point multiplications. The process is even slower when multi-scale matching is needed. This makes it not suitable for time-critical applications. In this paper, we present a novel approach to accelerate multi-scale template matching. The main computation saving is achieved by representing the template as a linear combination of a small number of Haar-like binary features. This representation replaces the element-by-element floating point multiplications with several additions thus significantly improves the speed. In addition, such simple features can easily adapt to template scale changes with negligible extra computation cost. Experiments show that the proposed method can achieve speed improvement up to two orders of magnitude. © 2007 IEEE.

关键词： pattern matching

来源：评论

学校读者我要写书评

暂无评论

Continuously tracking objects across multiple widely separated cameras

引用

8th Asian Conference on computer vision

作者： Cai, Yinghao Chen, Wei Huang, Kaiqi Tan, Tieniu Chinese Acad Sci Inst Automat Natl Lab Pattern Recognit Beijing 100080 Peoples R China

ISBN: (纸本)9783540763857

In this paper, we present a new solution to the problem of multi-camera tracking with non-overlapping fields of view. The identities of moving objects are maintained when they are traveling from one camera to another. Appearance information and spatio-temporal information are explored and combined in a maximum a posteriori (MAP) framework. In computing appearance probability, a two-layered histogram representation is proposed to incorporate spatial information of objects. Diffusion distance is employed to histogram matching to compensate for illumination changes and camera distortions. In deriving spatio-temporal probability, transition time distribution between each pair of entry zone and exit zone is modeled as a mixture of Gaussian distributions. Experimental results demonstrate the effectiveness of the proposed method.

关键词： Object recognition

来源：评论

学校读者我要写书评

暂无评论

New coarse region segmentation used in computer-aided diagnosis of liver cancer from ultrasound images

New coarse region segmentation used in computer-aided diagno...

引用

Conference on Medical Imaging, Parallel Processing of Images, and Optimization Techniques

作者： Huang, Xiaoyue Ding, Mingyue Xu, Tiantian Zhang, Songgeng Huazhong Univ Sci & Technol Inst Pattern Recognit & Artificial Intelligence State Key Lab Image Proc & Intelligent Control Wuhan 430074 Hubei Peoples R China HUST Dept Biomed Engn State Key Lab Image Proc & Intelligent Control Wuhan Peoples R China Beijing Tsinghua R & D Ind Inst Beijing 100085 Peoples R China

ISBN: (纸本)9780819469533

In this paper a coarse region segmentation of liver cancer in ultrasound Images is introduced. The reason employing coarse region segmentation is to reflect the inhomogeneous distribution of the image gray levels and provide the features such as the distribution, shape and size of the suspect region of liver cancer. Then combine with the prior knowledge we can divide the image into three different classes, which the results of the analysis of the region's location can be used by a classifier in a multilayer classifier. Furthermore, the result of the coarse region segmentation will support the texture analysis for further classification. The segmentation is based on watershed algorithm in order to receive an integrated region and two processing techniques are adopted to avoid the over segmentation of watershed algorithm.

关键词： segmentation liver cancer computer-aided diagnosis image processing watershed

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：