检索结果-内蒙古大学图书馆

International IEEE Conference on Intelligent Transportation systems

作者： David Van Hamme Peter Veelaert Wilfried Philips Vision Systems research group University College Ghent Image Processing and Interpretation (IPI) research group department of Telecommunications and Information Processing (TELIN) Ghent University

ISBN: (纸本)9781467330640

GPS navigation is often found undependable in urban situations where tall structures occlude large parts of the sky. To keep accurate position in these situations, we need an alternative method. We propose a novel visual odometry method that is shown to provide reliable relative motion estimation in typical urban road driving using a single camera. While the short-term accuracy is good, relative motion estimation by itself is susceptible to drift and therefore insufficient to provide good long-term absolute position estimates. To overcome this drift, we propose to warp the visual odometry output to a stored map. This warping must be able to cope with temporary discrepancies between visual odometry and map data. The proposed mapping does not make a hard decision about road position, but instead entertains all plausible hypotheses about the followed trajectory and their associated warping costs. Evaluation on real test sequences proves that the method succesfully eliminates drift, and on average stays within 7 metres of simultaneously recorded GPS data. This proves that the combined visual odometry and mapping are sufficient to provide positioning with comparable accuracy to GPS in those situations when GPS is unavailable.

关键词： Cameras Visualization Vehicles Roads Trajectory Global Positioning System Feature extraction

来源：评论

学校读者我要写书评

暂无评论

Classification and numbering of dental radiographs for an automated human identification system

引用

Telkomnika 2012年第1期10卷 137-146页

作者： Yuniarti, Anny Nugroho, Anindhita Sigit Amaliah, Bilqis Arifin, Agus Zainal Laboratory of Vision and Image Processing Department of Informatics Faculty of Information Technology Institut Teknologi Sepuluh Nopember Kampus ITS Sukolilo Surabaya 60111 Indonesia

Dental based human identification is commonly used in forensic. In a case of large scale investigation, manual identification needs a large amount of time. In this paper, we developed an automated human identification system based on dental radiographs. The system developed has two main stages. The first stage is to arrange a database consisting of labeled dental radiographs. The second stage is the searching process in the database in order to retrieve the identification result. Both stages use a number of image processing techniques, classification methods, and a numbering system in order to generate dental radiograph's features and patterns. The first technique is preprocessing which includes image enhancement and binarization, single tooth extraction, and feature extraction. Next, we performed dental classification process which aims to classify the extracted tooth into molar or premolar using the binary support vector machine method. After that, a numbering process is executed in accordance with molar and premolar pattern obtained in the previous process. Our experiments using 16 dental radiographs that consist of 6 bitewing radiographs and 10 panoramic radiographs, 119 teeth objects in total, has shown good performance of classification. The accuracy value of dental pattern classification and dental numbering system are 91.6 % and 81.5% respectively. © 2012 Universitas Ahmad Dahlan.

关键词： Numbering systems

来源：评论

学校读者我要写书评

暂无评论

Realization of 2D Reverse-Lattice discrete filters

Realization of 2D Reverse-Lattice discrete filters

引用

International Symposium on Communications Control and Signal processing (ISCCSP)

作者： G. E. Antoniou Department of Computer Science Image Processing and Systems Laboratory Montclair State University Montclair NJ USA

In this paper state space and circuit realizations are presented for 2D Reverse-Lattice based filters. The proposed state space realizations are based on the corresponding circuit implementation. For the state space realization the 2D Cyclic model, having minimal dimension and delay elements, is used. Two low-order representative examples are presented to illustrate the proposed results using, 2D all-pass and all-pole with a scale factor, realization structures.

关键词： Mathematical model Delay Transfer functions Integrated circuit modeling Aerospace electronics Lattices Equations

来源：评论

学校读者我要写书评

暂无评论

What Can an image of a Smooth Lambertian Surface Tell About Camera Nonlinearity?

What Can an Image of a Smooth Lambertian Surface Tell About ...

引用

Canadian Conference on Computer and Robot vision, CRV

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

This paper investigates the use of a single image of a smooth Lambertian surface to calibrate and remove some image nonlinearities due to the imaging device. To the best of our knowledge, this has not been addressed before in the literature. We show that this is possible, both theoretically and practically, taking advantage of some local shading measures that vary nonlinearly as a function of luminance and geometric nonlinearities (e.g., gamma correction and lens distortion). This can work as a basis for developing a simple method to estimate these nonlinearities from a single image. Several experiments are reported to validate the proposed method.

关键词： Cameras Distortion measurement Lenses Calibration Shape Lighting Surface treatment

来源：评论

学校读者我要写书评

暂无评论

DIRECT METHOD FOR SHAPE RECOVERY FROM POLARIZATION AND SHADING

DIRECT METHOD FOR SHAPE RECOVERY FROM POLARIZATION AND SHADI...

引用

IEEE International Conference on image processing

作者： Ali H. Mahmoud Moumen T. El-Melegy Aly A. Farag Computer Vision and Image Processing Laboratory University of Louisville Louisville KY USA Electrical Engineering Department Assiut University Assiut Egypt

Polarization imaging can give information about surface shape, and roughness. Polarization has been used for shape recovery, but with convex/concave reconstruction ambiguity. In this paper, we present a direct method to shape recovery using both polarization and shading that resolves this ambiguity, without the need for nonlinear optimization routines. Several experiments on synthetic and real datasets are reported to evaluate the proposed method. The method consistently outperforms some well-known methods based on polarization information alone.

关键词： Shape Surface reconstruction Azimuth image reconstruction Surface treatment Cameras

来源：评论

学校读者我要写书评

暂无评论

Heat Kernels for Non-Rigid Shape Retrieval: Sparse Representation and Efficient Classification

Heat Kernels for Non-Rigid Shape Retrieval: Sparse Represent...

引用

Canadian Conference on Computer and Robot vision, CRV

作者： Mostafa Abdelrahman Moumen El-Melegy Aly Farag Computer Vision and Image Processing Laboratory University of Louisville Louisville KY USA Electrical Engineering Department Assiut University Asyut Egypt

One of the major goals of computer vision and machine intelligence is the development of flexible and efficient methods for shape representation. This paper presents an approach for shape retrieval based on sparse representation of scale-invariant heat kernel. We use the Laplace-Beltrami eigen functions to detect a small number of critical points on the shape surface. Then a shape descriptor is formed based on the heat kernels at the detected critical points for different scales, combined with the normalized eigen values of the Lap lace-Beltrami operator. Sparse representation is used to reduce the dimensionality of the calculated descriptor. The proposed descriptor is used for classification via the collaborative representation-based classification with regularized least square algorithm. We compare our approach to two well-known approaches on two different data sets: the nonrigid world data set and the SHREC 2011. The results have indeed confirmed the improved performance of the proposed approach, yet reducing the time and space complicity of the shape retrieval problem.

关键词： Shape Heating Eigenvalues and eigenfunctions Kernel Vectors Manifolds Heat transfer

来源：评论

学校读者我要写书评

暂无评论

Large Pose Invariant Face Recognition Using Feature-based Recurrent Neural Network

Large Pose Invariant Face Recognition Using Feature-based Re...

引用

International Joint Conference on Neural Networks

作者： Teddy Salan Khan M. Iftekharuddin Intelligent Systems and Image Processing Lab Department of Electrical and Computer Engineering University of Memphis Vision Lab Department of Electrical and Computer Engineering Old Dominion University

ISBN: (纸本)9781467314886

Cellular Simultaneous Recurrent Network (CSRN) is a novel bio-inspired recurrent neural network that mimics reinforcement learning in the brain. CSRN has been proven to be a powerful tool for learning and predicting temporal information in face image sequences. In this work, we propose a novel implementation of feature-based CSRN for large-scale pose invariant face recognition. We also report systematic evaluation and performance comparison of our feature-based CSRN method with other well-known standard algorithms (PCA, LDA, Bayesian Classifier and EBGM) using face recognition technology standards for large-scale pose invariant face recognition.

关键词： Large pose invariant face recognition Cellular Simultaneous Neural Network Feature extraction Motion unit Face Recognition Vendor Test Face Recognition Technology (FERET) program

来源：评论

学校读者我要写书评

暂无评论

Coral reef mosaicking using Teardrop and Fast image Labeling

Coral reef mosaicking using Teardrop and Fast Image Labeling

引用

OCEANS 2012 MTS/IEEE Yeosu Conference: The Living Ocean and Coast - Diversity of Resources and Sustainable Activities

作者： Corpuz, Francis James Naval, Prospero Capili, Eusebio Jauod, Jaylord Judilla, Roel John Soriano, Maricor Vision and Image Processing Group National Institute of Physics University of the Philippines Diliman Quezon City Philippines Department of Computer Science College of Engineering University of the Philippines Diliman Quezon City Philippines Mapua Robotics Center Special Projects Laboratory Mapua Institute of Technology Intramuros Manila Philippines

ISBN: (纸本)9781457720895

In this paper, we present an inexpensive system for diverless video capture and fast image stitching of image frames for rapid reef assessment of shallow coral reefs. Our system has two main components: 1) Teardrop, a boat-towable, winged hull apparatus designed to house a commercial digital camera, and 2) a mosaicking algorithm to stitch the coral reef video into mosaics for further appreciation and analysis. The captured reef video is then separated into image frames which are to be stitched sequentially using Fast image Labeling. The overlap between image frames is estimated using Single-Step DFT, an efficient sub-pixel estimation algorithm. The estimated overlap is used to compute for the area to be added to the mosaic space. The overlapping section between succeeding image-pairs are stitched along a seam determined by a minimal-cost path using dynamic programming. The visibility of the seam boundaries is further minimized by utilizing blending on multi-resolution splines. Experimental results on automated reef mosaics creation from actual coral reef video taken using Teardrop shows the performance of the system described. The main contribution of this work is the demonstration of a rapid reef visualization system using a diverless system and commercially available, non-research-grade imaging equipment. © 2012 IEEE.

关键词： Reefs

来源：评论

学校读者我要写书评

暂无评论

Human action concentric video retrieval system using features weight updating method as relevance feedback

Human action concentric video retrieval system using feature...

引用

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

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

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：