By Topic

Neural Network Applications in Electrical Engineering, 2006. NEUREL 2006. 8th Seminar on

Date 25-27 Sept. 2006

Filter Results

Displaying Results 1 - 25 of 68
  • [Front cover]

    Page(s): C1
    Save to Project icon | Request Permissions | PDF file iconPDF (818 KB)  
    Freely Available from IEEE
  • Organizing Committee

    Page(s): nil1
    Save to Project icon | Request Permissions | PDF file iconPDF (758 KB)  
    Freely Available from IEEE
  • Foreword

    Page(s): nil2
    Save to Project icon | Request Permissions | PDF file iconPDF (1201 KB) |  | HTML iconHTML  
    Freely Available from IEEE
  • [Copyright notice]

    Page(s): nil3
    Save to Project icon | Request Permissions | PDF file iconPDF (490 KB)  
    Freely Available from IEEE
  • Table of contents

    Page(s): nil4 - nil6
    Save to Project icon | Request Permissions | PDF file iconPDF (2261 KB)  
    Freely Available from IEEE
  • [Breaker page]

    Page(s): nil7
    Save to Project icon | Request Permissions | PDF file iconPDF (171 KB)  
    Freely Available from IEEE
  • Full text access may be available. Click article title to sign in or learn about subscription options.
  • [Breaker page]

    Page(s): nil8
    Save to Project icon | Request Permissions | PDF file iconPDF (163 KB)  
    Freely Available from IEEE
  • Full text access may be available. Click article title to sign in or learn about subscription options.
  • [Breaker page]

    Page(s): nil9
    Save to Project icon | Request Permissions | PDF file iconPDF (175 KB)  
    Freely Available from IEEE
  • Asymmetric and Normalized Cuts for Image Clustering and Segmentation

    Page(s): 5 - 9
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (6141 KB) |  | HTML iconHTML  

    Over the last few years spectral clustering has emerged as a powerful model for data partitioning and segmentation. Spectral clustering techniques use eigenvalues and eigenvectors of the matrix representation of a suitable graph representing the original data. In this paper a new spectral clustering method is proposed: the asymmetric cut. It allows extraction of relevant information from a dataset by making just one cut over the database. The approach is tailored to the image classification task where a given image class is to be extracted from an image database containing an unknown number of classes. The main goal of this paper is to show that the proposed technique outperforms standard spectral methods under given circumstances. The technique is compared against the conventional and well-known normalized cut algorithm View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Minor Component Analysis (MCA) Applied to Image Classification in CBIR Systems

    Page(s): 11 - 16
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (7442 KB) |  | HTML iconHTML  

    A content-based image retrieval system with query image classification prior to retrieving procedure is proposed. Query image is compared to representative patterns of image classes, not to all images from database, accelerating thus initial retrieving step. Such procedure is possible when images from database are grouped into classes with similar content. Classification is performed using minor component (MC) analysis. Since it is expectable that MCs mainly depend on image details, not on an image background, this approach seems to be more efficient than classic CBIR. Minor components may be calculated by using single-layer neural network. The efficiency of proposed system is tested over images from Corel dataset View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Evaluating the Influence of Image Modifications upon Content-Based Multimedia Retrieval

    Page(s): 17 - 20
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (4595 KB) |  | HTML iconHTML  

    In this paper, evaluation of properties of MUVIS software package is presented. Images are modified in several ways (e.g. blurred, noise added, etc.) and added into the existing MUVIS databases. Simulations show that a satisfactory retrieval performance can be obtained from small set of extracted features (in comparison with large default set), even when significantly different number of images in databases are observed. YUV feature (color), GLCM (texture) and CANN (shape) feature are extracted from images. Applied image modifications have greatest impact on extracted YUV feature (color), which differs most in comparison with original image, while the lowest impact is observed on GLCM feature (gray level co-occurrence matrix), which shows that texture was not significantly changed during image modifications. CANN feature (shape and edges extraction feature) is only slightly different from original image. These results are similar for databases with various numbers of elements View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • GA-Based Feature Extraction for Clapping Sound Detection

    Page(s): 21 - 25
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (4653 KB) |  | HTML iconHTML  

    Automatically extracting semantic content from audio streams can be helpful in many multimedia applications. In this paper, we introduce a framework for automatic feature subspace selection from a common feature vector. The selected features build a new representation which is better suitable for a given learning task and recognition. In order to solve this problem, we propose the GA-based (genetic algorithm) method to improve the representativeness and robustness of the features generic audio recognition task View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Real-time Face Detection and Tracking of Animals

    Page(s): 27 - 32
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (6090 KB) |  | HTML iconHTML  

    This paper presents a real-time method for extracting information about the locomotive activity of animals in wildlife videos by detecting and tracking the animals' faces. As an example application, the system is trained on lions. The underlying detection strategy is based on the concepts used in the Viola-Jones detector, an algorithm that was originally used for human face detection utilising Haar-like features and AdaBoost classifiers. Smooth and accurate tracking is achieved by integrating the detection algorithm with a low-level feature tracker. A specific coherence model that dynamically estimates the likelihood of the actual presence of an animal based on temporal confidence accumulation is employed to ensure a reliable and temporally continuous detection/tracking capability. The information generated by the tracker can be used to automatically classify and annotate basic locomotive behaviours in wildlife video repositories View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • [Breaker page]

    Page(s): nil10
    Save to Project icon | Request Permissions | PDF file iconPDF (159 KB)  
    Freely Available from IEEE
  • On Rotation Invariant Texture Classification Using Two-Grid Coupled CNNs

    Page(s): 33 - 36
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (4014 KB) |  | HTML iconHTML  

    This paper presents several results of rotation-invariant texture classification using a bank of 2D band-pass CNN filters with approximately circular frequency response. The filters are autonomous two grid coupled CNNs, capable of producing Turing patterns used in the central linear part of their characteristic. The classification performances of the CNN filters are compared with the performances of the ideal circular filters View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Improvements in Image Segmentation by Applying Hopfield Neural Networks

    Page(s): 37 - 40
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (3745 KB) |  | HTML iconHTML  

    Hundreds of algorithms for different aspects of image segmentation have been developed. In this paper goals of using segmentation in video and multimedia systems are considered. Thus direction of improvements is towards forming masks and matte signals. The idea is to spread locally dominant segment in already segmented image. This problem is set as optimization problem and solution can be achieved by Hopfield neural network. Presented results are obtained applied to color detected segmentation, performed by mean shift algorithm View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Synchronization of Chaotic Cellular Neural Networks based on Rössler Cells

    Page(s): 41 - 43
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (3089 KB) |  | HTML iconHTML  

    Using and extending the approach in previous studies we demonstrate synchronization of two hyper chaotic cellular neural networks consisting of 25 cells governed by chaotic Rossler dynamics. We guarantee global asymptotic stability of the synchronization manifold by designing a nonlinear observer in such a way that the resulting error system is linear and time invariant. This linear error system is evaluated and a state feedback is designed to accomplish full state synchronization. Analytical as well as numerical simulation results are presented View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Face Detection Approach in Neural Network Based Method for Video Surveillance

    Page(s): 44 - 47
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (4872 KB) |  | HTML iconHTML  

    Neural networks are adaptive information processing systems that offer attractive solutions for video surveillance. This application aims at identifying particular patterns. Also, MPEG-4 standard profiling strategy in facial animation guarantees that the standard can provide adequate solutions for video surveillance. The main goal of this presentation is to provide face detection for video surveillance using neural network based method. After providing the corresponding architecture for face detection, the emphasize is on the detector which is trained with multilayer back propagation neural networks. Three different face representations are taken into account, i.e. pixel representation, partial profile representation and eigenface representation. Based on this, three independent sub-detectors are generated. The detection rates are measured. The circle at about 94% indicates the position where the neural network achieves the optimal performance View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Face Recognition by Using Unitary Vector Spaces

    Page(s): 49 - 51
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (2291 KB) |  | HTML iconHTML  

    Dynamic developments of science and technology have demanded necessity of interdisciplinary approach and appearance of novel scientific disciplines. In this respect, face recognition using quantum mechanical methods of unitary vector spaces, represents very interesting field due to possible applications in the field of quantum informatics. Thus traditional quantum mechanical methods widely applied to microsystems during the past century are now successfully extrapolated in macroscopic information framework as well View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • [Breaker page]

    Page(s): nil11
    Save to Project icon | Request Permissions | PDF file iconPDF (160 KB)  
    Freely Available from IEEE
  • Gaussian Sum Filters for Recurrent Neural Networks training

    Page(s): 53 - 57
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (7173 KB) |  | HTML iconHTML  

    We consider the problem of recurrent neural network training as a Bayesian state estimation. The proposed algorithm uses Gaussian sum filter for nonlinear, non-Gaussian estimation of network outputs and synaptic weights. The performances of the proposed algorithm and other Bayesian filters are compared in noisy chaotic time series long-term prediction View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Applications of Neural Networks in Network Intrusion Detection

    Page(s): 59 - 64
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (7359 KB) |  | HTML iconHTML  

    In this paper, we discuss the applications of multilayer perceptrons for classification of network intrusion detection data characterized by skewed class distributions. We compare several methods for learning from such skewed distributions by manipulating data records. The investigated methods include oversampling, undersampling and generating artificial data records using SMOTE technique. The presented methods are tested on KDDCup99 network intrusion dataset and compared using various classification performance metrics. In addition, the influence of decision margin on recall and misclassification rates is also examined View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Routing in Optical Networks by Using Neural Network

    Page(s): 65 - 68
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (4353 KB) |  | HTML iconHTML  

    Routing in optical, especially wavelength division multiplexing networks, is very hard task. This paper defines a new routing algorithm, based on Hopfield neural network. It is improvement of previous research, now applied to optical communications View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.