检索结果-内蒙古大学图书馆

International Conference on computer Networks and Information Technology (ICCNIT)

作者： Tasweer Ahmad Ahlam Jameel Balal Ahmad Government College University Lahore Pakistan

pattern recognition has become an attractive research oriented field of the computer vision and machine learning for the last few decades. Neural pattern recognition techniques are also being exercised for pattern recognition, showing promising results. In this paper, a comparison is made between statistical and neural pattern recognition techniques and tried to realize how neural techniques reveal far better results than statistical techniques. In this comparison, Discriminant Analysis (DA) and Principal Component Analysis (PCA) are used for pattern recognition, which are a statistical technique. Discriminant Analysis engrosses the problem of huge data dimensions and small sample size. To evade these problems, pattern recognition task is also implemented using Generalized Regression Neural Network (GRNN) and Back-propagation Neural Network (BPNN) techniques. The task of pattern recognition is conceded on a data base of face images of 400 people. Neural networks proved results for better than statistical methods.

关键词： pattern recognition Support vector machine classification Databases Labeling Accuracy

来源：评论

学校读者我要写书评

暂无评论

Biologically inspired template matching using scene context

Biologically inspired template matching using scene context

引用

IEEE computer Society Conference on computer vision and pattern recognition Workshops (CVPRW)

作者： Changxin Gao Nong Sang Institute of Pattern Recognition and Artificial Intelligence Huazhong University of Science and Technology Wuhan China

Template matching is widely used in pattern recognition and computer vision. However, the performance of traditional template matching approaches is often sensitive to large intraclass variance, occlusion, minor variety of poses, low-resolution conditions, background clutter etc. To resolve the problem, this paper presents a scene context based template matching method, which is inspired from two biological plausible ways, scene-centered viewpoint and change blindness phenomenon. The proposed method is quite different form conventional template matching methods, since this technique measures the similarity of a template and a sub-image by putting both into scene context, in a semantic representation. This work focuses on object detection task using only one object instance without training. Extensive experiments on three datasets and a video clearly show the superior robustness of the proposed method.

关键词： Context Layout Humans Robustness Clutter Blindness Databases

来源：评论

学校读者我要写书评

暂无评论

pattern recognition by affine Legendre moment invariants

Pattern recognition by affine Legendre moment invariants

引用

IEEE International Conference on Image Processing

作者： Hui Zhang Q. M. Jonathan Wu Department of Electrical and Computer Engineering University of Windsor Canada

Affine moment invariants are important shape descriptors in pattern recognition and computer vision. Existing affine invariants methods are based on geometric and complex moments. In this paper, we propose a set of affine invariants extracted from Legendre moments. These invariants are derived by the relationship between the Legendre moment of the affine transformed image and that of the original image. The performance of the proposed descriptor is evaluated with a set of binary and gray images. Experimental results show that the proposed method behaves better than existing methods in terms of pattern recognition accuracy.

关键词： pattern recognition Noise Conferences Testing Image processing Accuracy Databases

来源：评论

学校读者我要写书评

暂无评论

Parsimonious classification of binary lacunarity data computed from food surface images using kernel principal component analysis and artificial neural networks

引用

MEAT SCIENCE 2011年第2期87卷 107-114页

作者： Iqbal, Abdullah Valous, Nektarios A. Sun, Da-Wen Allen, Paul Univ Coll Dublin Sch Agr Food Sci & Vet Med Agr & Food Sci Ctr FRCFT Grp Dublin 4 Ireland TEAGASC Ashtown Food Res Ctr Dublin 15 Ireland

Lacunarity is about quantifying the degree of spatial heterogeneity in the visual texture of imagery through the identification of the relationships between patterns and their spatial configurations in a two-dimensional setting The computed lacunarity data can designate a mathematical index of spatial heterogeneity therefore the corresponding feature vectors should possess the necessary Inter-class statistical properties that would enable them to be used for pattern recognition purposes The objectives of this study is to construct a supervised parsimonious classification model of binary lacunarity data-computed by Valous et al (2009)-from pork ham slice surface images with the aid of kernel principal component analysis (KPCA) and artificial neural networks (ANNs) using a portion of informative salient features At first the dimension of the initial space (510 features) was reduced by 90% in order to avoid any noise effects in the subsequent classification Then using KPCA the first nineteen kernel principal components (99 04% of total variance) were extracted from the reduced feature space and were used as input in the ANN An adaptive feedforward multilayer perceptron (MLP) classifier was employed to obtain a suitable mapping from the input dataset The correct classification percentages for the training test and validation sets were 86 7% 86 7% and 85 0% respectively The results confirm that the classification performance was satisfactory The binary lacunarity spatial metric captured relevant information that provided a good level of differentiation among pork ham slice images (C) 2010 The American Meat Science Association Published by Elsevier Ltd All rights reserved

关键词： computer vision Pork ham Gliding box binary lacunarity Salient features Supervised classification Chi square Mutual information

来源：评论

学校读者我要写书评

暂无评论

Real-time Ellipse fitting, 3D spherical object localization, and tracking for the ICUB simulator

Real-time Ellipse fitting, 3D spherical object localization,...

引用

作者： Greggio, Nicola Bernardino, Alexandre Laschi, Cecilia Dario, Paolo Santos-Victor, José ARTS Lab - Scuola Superiore S. Anna Polo S. Anna Valdera Viale R. Piaggio 34 56025 Pontedera Italy Instituto de Sistemas e Robótica Instituto Superior Técnico 1049-001 Lisboa Portugal

ISBN: (纸本)9789898425751

This paper presents the implementation of real-time tracking algorithm for following and evaluating the 3D position of a generic spatial object. The key issue of our approach is the development of a new algorithm for pattern recognition in machine vision, the Least Constrained Square-Fitting of Ellipses (LCSE), which improves the state of the art ellipse fitting procedures. It is a robust and direct method for the least-square fitting of ellipses to scattered data. Although it has been ellipse-specifically developed, our algorithm demonstrates to be well suitable for the real-time tracking any spherical object, and it presents also robustness against noise. In this work we applied it to the RobotCub humanoid robotics platform simulator. We compared its performance with the Hough Transform and with its original formulation, made by Fitzgibbon et Al. in 1999, in terms of robustness (success/failure in the object detection) and fitting precision. We performed several tests to prove the robustness of the algorithm within the overall system. Finally we present our results.

关键词： computer vision

来源：评论

学校读者我要写书评

暂无评论

Dust storm detection using a neural network with uncertainty and ambiguity output analysis

Dust storm detection using a neural network with uncertainty...

引用

Lecture Notes in computer Science

作者： Chacon-Murguía, Mario I. Quezada-Holguín, Yearim Rivas-Perea, Pablo Cabrera, Sergio DSP and Vision Laboratory Chihuahua Institute of Technology Chihuahua Mexico ECE University of Texas El Paso United States

ISBN: (纸本)9783642215865

Dust storms are meteorological phenomena that may affect human life. Therefore, it is of great interest to work towards the development of a stand-alone dust storm detection system that may help to prevent and/or counteract its negative effects. This work proposes a dust storm detection system based on an Artificial Neural Network, ANN. The ANN is designed to identify not just dust storm areas but also vegetation and soil. The proposed ANN works on information obtained from multispectral images acquired with the Moderate Resolution Imaging Spectroradiometer (MODIS) instrument. Before the multispectral information is fed to the ANN a process to remove cloud regions from images is performed in order to reduce the computational burden. A method to manage undefined and ambiguous ANN outputs is also proposed in the paper which significantly reduces the false positives rate. Results of this research present a suitable performance at detecting the dust storm events. © 2011 Springer-Verlag Berlin Heidelberg.

关键词： Neural networks

来源：评论

学校读者我要写书评

暂无评论

Registered Localization Method-based on OpenCV for Augmented Reality

Registered Localization Method-based on OpenCV for Augmented...

引用

International Conference of Soft Computing and pattern recognition (SOCPAR)

作者： Jun Dai Li-fen Zhang College of Information Science and Technology Jiujiang University Jiujiang China

Augmented Reality is a kind of technique combining a virtual world with a real environment. This paper tells that OpenCV is applied to the realization of Registration Technique which is key to Augmented Reality. It studies the implementation of the camera calibration, the extraction of characteristic target and its recognition. This paper improved the efficiency of the exploitation and laid a foundation for the exploitation and application of Augmented Reality.

关键词： Feature extraction Cameras Augmented reality Calibration Target recognition Lenses computer vision

来源：评论

学校读者我要写书评

暂无评论

A system for change detection and human recognition in voxel space using the Microsoft Kinect sensor

A system for change detection and human recognition in voxel...

引用

Applied Imagery pattern recognition Workshop (AIPR)

作者： T. Gill J. M. Keller D. T. Anderson R. H. Luke Electrical and Computer Engineering University of Missouri Columbia MO USA Electrical and Computer Engineering Mississippi State University MS USA U.S. Army Night Vision and Electronic Sensors Directorate-Countermine Div. Fort Belvoir VA USA

Previously, we put forth a new computer vision system for indoor well-being monitoring of elderly populations based on the use of multiple stereo camera pairs. That approach involves combining the strengths of image space with three dimensional volume element (voxel) space techniques. However, that system is fundamentally limited because it is based on color imagery from visible light cameras. In this article, we extend our prior research and consider a new, inexpensive infrared depth camera device, the Microsoft Kinect. Advantages, such as the ability to operate 24-7 in low-to-no light conditions, and shortcomings are detailed. In addition, we discuss necessary algorithmic extensions to our mixed image and voxel space framework for the Kinect sensor. Experiments are performed in a laboratory designed to resemble an elders living quarter. vision findings are evaluated using our prior high-level linguistic summarization of human activity work. Preliminary results indicate that the Kinect sensor does indeed work in a wider range of operating conditions and it can produce activity descriptions that match that of a human.

关键词： Cameras Humans Stereo vision Image color analysis Lighting Monitoring Skin

来源：评论

学校读者我要写书评

暂无评论

The Effect of Block Size, Training Set and K-Value in the Classification of Food Grains Using HSI Color Model

The Effect of Block Size, Training Set and K-Value in the Cl...

引用

National Conference on computer vision, pattern recognition, Image Processing and Graphics (NCVPRIPG)

作者： Neelamma K. Patil Ravi M. Yadahalli Department of Telecommunication Engineering KLE Society's College of Engineering and Technology Belgaum Karnataka India Department of Electronics & Communication Engineering PES Institute of Technology & Management Shivamogga Karnataka India

The method proposed for the classification of food grains can be divided into two phases: 1) color, global and local feature extraction 2) classification using extracted features. This paper presents the effect of block size, training set and K-value in the classification of food grains using HSI color model by combining color and texture information without pre-processing. The first phase in the proposed system is feature extraction. The features are computed locally and globally. The co-occurrence matrix helps to extract features locally. The non-uniformity of RGB color space is eliminated by Hue, Saturation and Intensity (HSI) color space. Further, minimum distance and K nearest neighbour algorithms are used for classification. Percentage of correct classification and error analysis are carried out by confusion matrix. The proposed work attains maximum average accuracy of 95.83% and 70.37% for block size 512×512 and 256×256 with K=5 and K=1 respectively.

关键词： Feature extraction Image color analysis Training Testing Databases Machine vision Histograms

来源：评论

学校读者我要写书评

暂无评论

A pattern framework driven by the Hamming distance for structured light-based reconstruction with a single image

A pattern framework driven by the Hamming distance for struc...

引用

IEEE Conference on computer vision and pattern recognition

作者： Xavier MAURICE Pierre GRAEBLING Christophe DOIGNON LSIIT (UMR CNRS 7005) Strasbourg University

ISBN: (纸本)9781457703942

Structured light based patterns provide a means to capture the state of an object shape. However it may be inefficient when the object is freely moving, when its surface contains high curvature parts or in out of depth of field situations. For image-based robotic guidance in unstructured and dynamic environment, only one shot is required for capturing the shape of a moving region-of-interest. Then robust patterns and real-time capabilities must be targeted. To this end, we have developed a novel technique for the generation of coded patterns directly driven by the Hamming distance. The counterpart is the big amount of codes the coding/decoding algorithms have to face with a high desired Hamming distance. We show that the mean Hamming distance is a useful criterion for driving the patterns generation process and we give a way to predict its value. Furthermore, to ensure local uniqueness of codewords with consideration of many incomplete ones, the Perfect Map theory is involved. Then, we describe a pseudorandom/exhaustive algorithm to build patterns with more than 200×200 features, in a very short time, thanks to a splitting strategy which performs the Hamming tests in the codeword space instead of the pattern array. This leads to a significant reduction of the computational complexity and it may be applied to other purposes. Finally, real-time reconstructions from single images are reported and results are compared to the best known which are outperformed in many cases.

关键词： Hamming distance Automobile driving codeword Reconstructive Surgical Procedures

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：