By Topic

Computer Vision, IET

Issue 3 • Date September 2009

Filter Results

Displaying Results 1 - 6 of 6
  • Vision-based monitoring method using gray relational analysis

    Page(s): 103 - 111
    Save to Project icon | Click to expandQuick Abstract | PDF file iconPDF (630 KB)  

    Intelligent surveillance system has become an important research topic in the field of computer vision. The authors propose a monitoring method based on the cellular model to monitor human activities in the indoor environment. The measured area of an indoor room is divided into several unit areas in which each unit area is considered as a simple cell in the cellular model. A rectangular box is then used to group those neighbouring active cells into a unit to represent a moving object. Since people usually walk without a fixed style and the colour of objects may be similar to that of the background, the distribution of active cells is often uncertain and incomplete. The authors therefore apply the gray relational analysis to detect and track multiple moving objects. Several experiments have been conducted to evaluate the performance of the proposed system. The experimental results show that the proposed system is highly effective in verifying and tracking multiple objects in real time. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Combining classifiers through fuzzy cognitive maps in natural images

    Page(s): 112 - 123
    Save to Project icon | Click to expandQuick Abstract | PDF file iconPDF (512 KB)  

    A new automatic hybrid classifier for natural images by combining two base classifiers through the fuzzy cognitive maps (FCMs) approach is presented in this study. The base classifiers used are fuzzy clustering (FC) and the parametric Bayesian (BP) method. During the training phase, different partitions are established until a valid partition is found. Partitioning and validation are two automatic processes based on validation measurements. From a valid partition, the parameters of both classifiers are estimated. During the classification phase, FC provides for each pixel the supports (membership degrees) that determine which cluster the pixel belongs to. These supports are punished or rewarded based on the supports (probabilities) provided by BP. This is achieved through the FCM approach, which combines the different supports. The automatic strategy and the combined strategy under the FCM framework make up the main findings of this study. The analysis of the results shows that the performance of the proposed method is superior to other hybrid methods and more accurate than the single usage of existing base classifiers. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Wand-based calibration of 3D kinematic system

    Page(s): 124 - 129
    Save to Project icon | Click to expandQuick Abstract | PDF file iconPDF (200 KB)  

    This study proposes a method to calibrate 3D kinematic systems. The authors briefly address typical wand-based calibration, that is, calibration using a rigid bar, both from the computational point of view and the procedural perspective, and then they define their alternative way of calibration. The proposed method takes advantage of a feature of the presently used calibration tools, that is, the orthogonality of calibration wands. The usual two steps in wand calibration are reduced to a single one. In addition, the authors propose an alternative for how to enforce a typically available geometric constraint, that is, the known wand length, during the parameter optimisation procedure (the well-known bundle adjustment). The obtained 3D reconstruction accuracy has proven to be comparable with results of commercial 3D kinematic systems. The relative simplicity of the proposed method offers potential for an implementation by a large number of researchers. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Robust face recognition using posterior union model based neural networks

    Page(s): 130 - 142
    Save to Project icon | Click to expandQuick Abstract | PDF file iconPDF (670 KB)  

    Face recognition with unknown, partial distortion and occlusion is a practical problem, and has a wide range of applications, including security and multimedia information retrieval. The authors present a new approach to face recognition subject to unknown, partial distortion and occlusion. The new approach is based on a probabilistic decision-based neural network, enhanced by a statistical method called the posterior union model (PUM). PUM is an approach for ignoring severely mismatched local features and focusing the recognition mainly on the reliable local features. It thereby improves the robustness while assuming no prior information about the corruption. We call the new approach the posterior union decision-based neural network (PUDBNN). The new PUDBNN model has been evaluated on three face image databases (XM2VTS, AT&T and AR) using testing images subjected to various types of simulated and realistic partial distortion and occlusion. The new system has been compared to other approaches and has demonstrated improved performance. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Estimating fibre orientation in spruce using lighting direction

    Page(s): 143 - 158
    Save to Project icon | Click to expandQuick Abstract | PDF file iconPDF (1603 KB)  

    Softwood is almost entirely composed of fibres and its physical properties depend on their orientation. A method is proposed to estimate the average fibre orientation at each pixel in the inspection image. In this study it is shown that finished wood has distinctive reflectance properties that are a consequence of the microstructure of the wood surface. If the fibres lie parallel to the image plane these properties can be used to estimate fibre orientation. It is argued that the reflectance behaviour reported in this study generalises to a wide range of materials with directional microstructures. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Discriminant analysis based on modified generalised singular value decomposition and its numerical error analysis

    Page(s): 159 - 173
    Save to Project icon | Click to expandQuick Abstract | PDF file iconPDF (256 KB)  

    Generalised singular value decomposition (GSVD) has been used in the literature for linear discriminant analysis (LDA) to solve the small sample size problem in pattern recognition. However, this method, commonly known as LDA/GSVD algorithm, suffers from excessive computational load when the sample dimension is high. Here the GSVD framework used in the LDA/GSVD algorithm is modified by replacing the SVD of a high-dimension matrix with the eigen-decomposition of a small size inner product matrix, thus circumventing the direct calculation of a high-dimension singular vector matrix. It is established by a theorem that if the samples are linearly independent in the feature space, the samples in each class are degenerated into a distinct single point of a discriminative space derived from the GSVD-based algorithms, and the distances between the points depend only on the respective numbers of the samples in the corresponding classes. In order to overcome the over-fitting problem, a method to orthogonalise the basis of the discriminative subspace is proposed. The proposed linear algorithm is kernelised for the discriminant analysis of samples that are not linearly independent as the non-linear kernel mapping can establish linear independence. The results of the above theorem are used to develop a method to measure the numerical error. This measure can also be used to decide the kernel parameters to minimise the numerical error in the non-linear algorithm. View full abstract»

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

Aims & Scope

IET Computer Vision seeks original research papers in a wide range of areas of computer vision. The vision of the journal is to publish the highest quality research work that is relevant and topical to the field, but not forgetting those works that aim to introduce new horizons and set the agenda for future avenues of research in Computer Vision.

Full Aims & Scope