• Speech Sciences
• /
• v.11 no.2
• /
• pp.211-216
• /
• 2004

In this paper, we present a fast encoding algorithm for vector quantization with an approximate Euclidean distance calculation. An approximation is performed by converting floating point to the near integer. An inequality between the approximate Euclidean distance and the nearest distance is developed to avoid unnecessary distance calculations. Since the proposed algorithm rejects those codewords that are impossible to be the nearest codeword, it produces the same output as conventional full search algorithm.

• Proceedings of the Korean Information Science Society Conference
• /
• 2005.07b
• /
• pp.748-750
• /
• 2005

Isomap [1] is a manifold learning algorithm, which extends classical multidimensional scaling (MDS) by considering approximate geodesic distance instead of Euclidean distance. The approximate geodesic distance matrix can be interpreted as a kernel matrix, which implies that Isomap can be solved by a kernel eigenvalue problem. However, the geodesic distance kernel matrix is not guaranteed to be positive semidefinite. In this paper we employ a constant-adding method, which leads to the Mercer kernel-based Isomap algorithm. Numerical experimental results with noisy 'Swiss roll' data, confirm the validity and high performance of our kernel Isomap algorithm.

• The Transactions of the Korean Institute of Electrical Engineers D
• /
• v.52 no.5
• /
• pp.305-310
• /
• 2003

In this paper, we proposed an IJADE(Improved joint approximate diagonalisation of eigenmatrices) which use high order statistics instead of second order statistics for data whitening. For simulation, we artificially construct signals mixed with two ECG signals, 60Hz power line interference and 16Hz sine signal and then put them into a JADE and an IJADE. To evaluate the performance of separated ECG signal in each algorithm, we have adopted indices such as kurtosis, standard deviation ratio, correlation coefficient and euclidean distance. As a results, IJ ADE showed theimproved performances as kurtosis of $2\%,$ standard deviation ratio of 0.2194, and Euclidean distance of 0.07 except correlation coefficient showing similar value. In conclusion, the proposed IJADE showed a good performance in separating ECG and a possibilities in applying to the various biological signal.

• Journal of the Korean Institute of Telematics and Electronics B
• /
• v.33B no.7
• /
• pp.138-146
• /
• 1996

In this paper, w epropose a new mehtod to represent the shape and to recognize the object. The shape description and the matching is implemented by using the distance transform on the morphological skeleton. The employed distance transform is the chamfer (3,4) distance transform, because the chamfer distance transform (CDT) has an approximate value to the euclidean distance. The 2-D object can be represented by means of the distribution of the distance transform on the morphological skeleton, the number of skeletons, the sum of the CDT, and the other features are employed as the mtching parameters. The matching method has the invariant features (rotation, translation, and scaling), and then the method is used effectively for recognizing the differently-posed objects and/or marks of the different shape and size.

• Korean Journal of Computational Design and Engineering
• /
• v.17 no.4
• /
• pp.282-293
• /
• 2012

This paper presents an error-bounded B-spline fitting technique to approximate unorganized data within a prescribed error tolerance. The proposed approach includes two main steps: leastsquares minimization and error-bounded approximation. A B-spline hypervolume is first described as a data representation model, which includes its mathematical definition and the data structure for implementation. Then we present the least-squares minimization technique for the generation of an approximate B-spline model from the given data set, which provides a unique solution to the problem: overdetermined, underdetermined, or ill-conditioned problem. We also explain an algorithm for the error-bounded approximation which recursively refines the initial base model obtained from the least-squares minimization until the Euclidean distance between the model and the given data is within the given error tolerance. The proposed approach is demonstrated with some examples to show its usefulness and a good possibility for various applications.