KMITL

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

Browse

Search Results

Now showing 1 - 10 of 10

Some of the metrics are blocked by your
consent settings
The preliminary investigation of ear recognition using hybrid technique
(2017-02-21)
Taertulakarn, S.
;
Pintavirooj, C.
;
Tosranon, P.
;
Hamamoto, K.
Ear recognition is one of the new patterns of biometrics. Ear structure is believed to contain specific and unique anatomical markers, which can be used both to distinguish it from others. In this paper, we derive preliminary of Ear identification based on the geometric features on 3D ear surface for 2D ear image. Principal components analysis (PCA) is used in this study for feature extraction to identify a person. The result is shown 92% recognition rate of 50 volunteers.
Some of the metrics are blocked by your
consent settings
3D ear alignment based on geometry invariant
(2016-02-04)
Taertulakarn, S.
;
Tosranon, P.
;
Pintavirooj, C.
Nowadays, ear shape study is one of the most significant trends in the biometric research. In this study, we present a novel surface fiducial point's detection that is computed from the differential surface geometry. The fiducial points are intrinsic, local, and relative invariants, i.e., they are preserved under similarity, affine, and nonlinear transformations that are piecewise affine. In our experiment, the fiducial points are used in a non-iterative geometric-based method for 3D ear alignment. The matching is achieved by establishing correspondences between fiducial points after a sorting based on a set of absolute local affine invariants derived from them. Experimental results showed that our purposed surface feature is suitable for further application to 3D ear identification.
Some of the metrics are blocked by your
consent settings
Using intrinsic surface geometry invariant for 3D ear alignment
(2015-01-01)
Taertulakarn, S.
;
Tosranon, P.
;
Pintavirooj, C.
In this study we derive novel surface fiducial point's detection that is computed from the differential surface geometry. The fiducial points are intrinsic, local, and relative invariants, i.e., they are preserved under similarity, affine, and nonlinear transformations that are piecewise affine. In our experiment, the fiducial points, computed from high order surface shape derivatives, are used in a non-iterative geometric-based method for 3D ear registration and alignment. The matching is achieved by establishing correspondences between fiducial points after a sorting based on a set of absolute local affine invariants derived from them. Experimental results showed that our purposed surface feature is suitable for further application to 3D ear identification because its robustness to geometric transformation.
Some of the metrics are blocked by your
consent settings
Gaussian curvature-based geometric invariance for ear recognition
(2014-01-20)
Taertulakarn, S.
;
Tosranon, P.
;
Pintavirooj, C.
Ear recognition is one of the new patterns of biometrics. In this paper we derive a novel geometric invariance on ear surfaces that it is preserved under affine and weak perspective transformations. Our 3D shape features are based on the Gaussian curvature and Mean curvature. When a surface undergoes an affine transformation, the shape features are the affine transformed shape features of the original surface; they are preserved and hence can be for shape matching. We have tested robustness of the shape feature on the 3D ear data for various linear geometric transformations. The experiment results show that our purposed shape feature is suitable for further application to 3D ear identification because its robustness to geometric transformation.
Some of the metrics are blocked by your
consent settings
Robustness of geodesics to affine transformation
(2011-12-01)
Panyindee, C.
;
Tosranon, P.
;
Pintavirooj, C.
This paper will determine the geometric invariance of geodesic on the surface that can handle the affine transform. The unique characteristics of the calculated from the differential geometry contained curvature surface that lies above the standard of Gaussian curve. When applied the affine transformed to the surface according to the condition, the surface characteristics will also be transformed. Therefore, the characteristics of the surface will still contain original form, in which we can use as the comparable properties for matching. In this work, the method is performed with 3 dimensional data and provided a satisfactory result. © 2011 IEEE.
Some of the metrics are blocked by your
consent settings
Robustness of novel surface invariance to geometric transformation
(2008-10-06)
Tosranon, P.
;
Sanpanish, A.
;
Bunluechokchai, S.
;
C.Pintavirooj
In this paper we explore the novel geometric invariance on surfaces based on the set of invariant normal vectors that are relatively preserved under geometric transformations, are local, intrinsic and computed from the differential geometry of the surface. To reduce the sensitivity of the computation of the geometric invariance to noise, we use a B-Spline surface representation that smoothes out the surface prior to the computation of these invariant points. The robustness of the geometric invariance is shown for a variety of geometric transformation. The result is very promising. ©2008 IEEE.
Some of the metrics are blocked by your
consent settings
Miniatured computed tomography system and calibration
(2006-12-01)
Pititheerapab, Y.
;
Chanwimalueang, T.
;
Pintavirooj, C.
;
Tosranon, P.
In this paper, we focus on a miniature computed tomographic system with application in 3D modelling of bony structure of a small animal. Our system consists of an x-ray source, a rotating platform and an x-ray array detector unit. The rotating platform is controlled by a personal computer which can rotate the sitting object to arbitrary angle. The x-ray array detector is used to capture the 2D x-ray signal that traversing the object placed on the platform. The x-ray detector is an image intensifier tube of which the 2D image is coupled to the computer via a CCD camera. Feldkamp Conebeam technique is engaged for reconstructing tomograms due to its simplicity. Volume rendering technique together with the shading effects is performed on a stack of cross-sectional image to realize the data into 3D visualization. The algorithms are applied to the practical situation where a series of x-ray radiographs of an animal's bone are collected from the system. The 3D modelling of such bone is performed on the cross-sectional images reconstructed with Feldkamp Conebeam. The results are very satisfactory. © 2006 IEEE.
Some of the metrics are blocked by your
consent settings
Ultrasonic reflection mode tomography using frequency-shift method
(2006-12-01)
Tosranon, P.
;
Onemanisone, T.
;
Greesuradej, P.
;
Pintavirooj, C.
;
Jaruwongrungsee, K.
Tomogram of the soft tissue can be reconstructed by using its projections resolved from the ultrasonic broadband pulsed waves which are passed through the tissue. While a data from ultrasonic transmission mode tomography cannot be used to image the character of soft tissue where shadowed by a bone, an ultrasonic reflection mode tomography is realized to be a solution for this. There are many ways to extract the projections, in most case the integrated attenuation coefficients of the tissue from the pulsed wave. Almost all of these methods were proved to implement with the transmission mode signal, but however, some adaptation may takes benefit from those method to implement with the reflection mode. We choose the frequency-shift method to be analyzed because its advantage in a computational viewpoint. The simple filtered backprojection algorithm was implemented to reconstruct the tomographic images of the soft tissue suffered by bone. The results are shown the successful of this method over the transmission mode tomography. Also the simulations for noisy data were analyzed. © 2006 IEEE.
Some of the metrics are blocked by your
consent settings
Acceleration of genetic algorithm with parallel processing with application in medical image registration
(2005-12-01)
Laksanapanai, B.
;
Withayachumnankul, W.
;
Pintavirooj, C.
;
Tosranon, P.
Generally, image registration using genetic algorithm is a time-consuming process since the algorithm needs to evaluate the objective function several hundred times depending on the vastness of search space. The situation appears worse if the registration is intensity-based due to the interpolation loops prior to each objective function. However, with the availability of parallel processing method, one can accelerate the application of genetic algorithm for iterative-based image registration process of up 80 % for multi-modality alignment. Copyright UNION Agency - Science Press.
Some of the metrics are blocked by your
consent settings
3D modeling from radiograph with Conebeam geometry
(2004-12-01)
Kawikitwitcha, S.
;
Pintavirooj, C.
;
Tosranon, P.
;
Kiriratnikbm, T.
;
Anuntaseree, S.
In order to render 3D model of the bone, the stack of cross-sectional images must be reconstructed from a series of X-ray radiographs, served as the projections. In the case where the distance between x-ray source and detector is not infinite, image reconstruction from projection based on parallel-beam geometry provides an error in the cross-sectional image. In such case, image reconstruction from projection based on conebeam geometry must be exercised instead. In this paper, the Simultaneous Algebraic Reconstruction Technique (SART Conebeam) is engaged for reconstructing tomograms in case of limited views of radiographs. Compared with Feldkamp Conebeam technique, the SART Conebeam theoretically gives the better quality of image for the same limited set of projections Volume rendering technique together with the shading effects is performed on a stack of cross-sectional image to realize the data into 3D visualization. The algorithms are applied to the practical situation where a series of x-ray radiographs of human femur bone are collected from a C-Arm x-ray apparatus. The 3D modeling of such bone is performed on the cross-sectional images reconstructed with SART conebeam. The results are very satisfactory. © 2004IEEE.

KMITL

Browse

Filters

Settings

Sort By

Results per page

Search Results