KMITL

Permanent URI for this communityhttps://dspace.kmitl.ac.th/handle/123456789/1

Browse

Settings

Author: search.filters.author.Thourn, Kosorl

Search Results

Now showing 1 - 5 of 5
  • Loading...
    Thumbnail Image
    Some of the metrics are blocked by your 
    consent settings
    Item type:Item,
    Shape retrieval using Eigen and Fisher Barycenter contour
    (2011-01-01)
    Thourn, Kosorl
    ;
    Kitjaidure, Yuttana
    ;
    Kondo, Shozo
    To achieve a good performance for shape retrieval, it requires both shape representation and classifier. In this paper, the algorithm for shape matching and retrieval is developed by using Eigen Barycenter Contour (EBcC) and Fisher Barycenter Contour (FBcC). In our algorithm, the Signed Enclosed Area (SEA) signature (formed by two adjacent points of contour and its center point), computed at each scale level of Barycenter contour (BcC), is utilized as the shape representation. The BcC technique is robust to moderate amount of noise and occlusion. Furthermore, the SEA signature is invariant to general affine transformation including translation, rotation, scale and shear. Because of high dimension of the shape representation, thus, in the matching step, two classifiers have been studied. The first classifier, Eigen face technique, is employed for dimensionality reduction while the second classifier, Fisher face technique, is used for reducing dimension as well and making discrimination. Then, the similarity among shapes is measured by the normalized crosscorrelation (NCC). The performance of our technique is evaluated onto the affine shape database and two well-known databases, the MPEG-7 shape database part B and the Kimia's database. The experimental results illustrate that our approach gives very high retrieval efficiency over all published methods.
  • Loading...
    Thumbnail Image
    Some of the metrics are blocked by your 
    consent settings
    Item type:Item,
    Eigen and fisher barycenter contour for 2D shape classification
    (2009-11-16)
    Thourn, Kosorl
    ;
    Kitjaidure, Yuttana
    ;
    Kondo, Shozo
    To achieve a good performance for shape classification, it requires both shape representation and classifier. In this paper, the so-called Eigen Barycenter Contour (EBcC) and Fisher Barycenter Contour (FBcC) techniques are presented for 2D shape classification. The representation utilizes the area of triangles at different scale level of Barycenter Contour (BcC). However, it is not invariant to starting point selection, so the phase normalization is applied. After that, we linearly project the shape feature in 3D format onto a subspace based on EBcC technique into low dimensional subspace. The FBcC, another similar method, also produces well separated classes in low dimensional subspace. Finally, the normalized cross correlation is used to measure the similarity among shapes. The experimental results demonstrate that the FBcC method outperforms the EBcC method and achieves high retrieval efficiency over other recent methods in the literature for tests on three different databases, the affine shape database, the MPEG-7 database CE-1 part B and the Kimia's database. © 2009 IEEE.
  • Loading...
    Thumbnail Image
    Some of the metrics are blocked by your 
    consent settings
    Item type:Item,
    Multi-view shape recognition based on principal component analysis
    (2009-04-24)
    Thourn, Kosorl
    ;
    Kitjaidure, Yuttana
    In this paper, the principal component analysis (PCA) for multi-view shape recognition is proposed. Our algorithm presents the signed enclosed area signature as the shape representation. In our method, the barycenter contour is used for decomposing the shape boundary into multiscale level. At each scale level, the signed enclosed area signatures are obtained. After that, the principal component analysis (PCA) is used as the recognition strategy. This method is independent to starting point of the contour by exploiting the property of the discrete Fourier transform (DFT). In the experimentation, the various number of the sample shapes are used as the training set and the rest are used as the testing set. The experimental results indicate that the recognition accuracy are high enough even one sample shape per class is used as the training set. As the more training sample shapes per class are used, the higher recognition will be. © 2008 IEEE.
  • Loading...
    Thumbnail Image
    Some of the metrics are blocked by your 
    consent settings
    Item type:Item,
    Affine invariant shape recognition based on multi-level of barycenter contour
    (2008-12-01)
    Thourn, Kosorl
    ;
    Kitjaidure, Yuttana
    ;
    Kondo, Shozo
    In this paper, a new multiresolution created from multi-level of barycenter contour is proposed in order to reduce the moderate amount of noise and to improve the retrieval efficiency of the recognition task in computer vision. Then, the Triangle Area Representation with two points (TAR-2p) signature at each level of barycenter contour is introduced as the shape representation. Finally, the normalized cross-correlation function at each level is used for measuring the similarity among the shapes. Our experiment has been performed on database consisting of 560 affine distorted shapes, chosen from MPEG-7 contour shape database CE-1. The results illustrate that our algorithm is invariant to affine transformation, robustness to the noise. Moreover, it achieves high retrieval efficiencies when compares to those of the Triangle Area Representation with three points (TAR-3p) signature and the centroid distance signature. © 2008 IEEE.
  • Loading...
    Thumbnail Image
    Some of the metrics are blocked by your 
    consent settings
    Item type:Item,
    Robust multi-TAR of barycenter contour for multiple views shape matching and retrieval
    (2008-01-01)
    Thourn, Kosorl
    ;
    Kitjaidure, Yuttana
    In this paper, we present a one dimensional descriptor for the two dimensional object silhouettes associated with each level of barycenter contour for multiple views shape matching and retrieval. Firstly, the barycenter contour is applied onto the shape contour. Then the averaging multitriangle area representation (AMTAR) at each level of barycenter contour is computed as the shape descriptor. Finally, to classify among the shapes, the normalized crosscorrelation function is used. Our algorithm is implemented on database consisting of 560 affine distorted shapes, chosen from MPEG-7 contour shape database CE-1. The experimental results indicate that the proposed method is invariant to affine transformation, robustness to the noise, and achieves high retrieval performance. © 2008 IEEE.