• Journal of the Korean Institute of Intelligent Systems
• /
• v.13 no.4
• /
• pp.403-408
• /
• 2003

A general meaning of caricaturing is that a representation, especially pictorial or literary, in which the subject's distinctive features or peculiarities are deliberately exaggerated to produce a comic or grotesque effect. In other words, a caricature is defined as a rough sketch(dessin) which is made by detecting features from human face and exaggerating or warping those. There have been developed many methods which can make a caricature image from human face using computer. In this paper, we propose a new caricaturing system. The system uses a real-time image or supplied image as an input image and deals with it on four processing steps and then creates a caricatured image finally. The four Processing steps are like that. The first step is detecting a face from input image. The second step is extracting special coordinate values as facial geometric information. The third step is deforming the face image using local warping method and the coordinate values acquired in the second step. In fourth step, the system transforms the deformed image into the better improved edge image using a fuzzy Sobel method and then creates a caricatured image finally. In this paper , we can realize a caricaturing system which is simpler than any other exiting systems in ways that create a caricatured image and does not need complex algorithms using many image processing methods like image recognition, transformation and edge detection.

• Journal of Broadcast Engineering
• /
• v.14 no.3
• /
• pp.322-331
• /
• 2009

This paper proposes an efficient data representation of stereo images using edge-based mesh optimization. Mash-based two dimensional warping for stereo images mainly depends on the performance of a node selection and a disparity estimation of selected nodes. Therefore, the proposed method first of all constructs the feature map which consists of both strong edges and boundary lines of objects for node selection and then generates a grid-based mesh structure using initial nodes. The displacement of each nodal position is iteratively estimated by minimizing the predicted errors between target image and predicted image after two dimensional warping for local area. Generally, iterative two dimensional warping for optimized nodal position required a high time complexity. To overcome this problem, we assume that input stereo images are only horizontal disparity and that optimal nodal position is located on the edge include object boundary lines. Therefore, proposed iterative warping method performs searching process to find optimal nodal position only on edge lines along the horizontal lines. In the experiments, we compare our proposed method with the other mesh-based methods with respect to the quality by using Peak Signal to Noise Ratio (PSNR) according to the number of nodes. Furthermore, computational complexity for an optimal mesh generation is also estimated. Therefore, we have the results that our proposed method provides an efficient stereo image representation not only fast optimal mesh generation but also decreasing of quality deterioration in spite of a small number of nodes through our experiments.

• Proceedings of the IEEK Conference
• /
• 2000.09a
• /
• pp.157-160
• /
• 2000

In the generally a voice conversion has used VQ(Vector Quantization) for partitioning the spectral feature and has performed by adding an appropriate offset vector to the source speaker's spectral vector. But there is not represented the target speaker's various characteristics because of discrete characteristics of transformed parameter. In this paper, these problems are solved by using the LMR(Linear Multivariate Regression) instead of the mapping codebook which is determined to the relationship of source and target speaker vocal tract characteristics. Also we propose the method for solved the discontinuity which is caused by applying to time aligned parameters using Dynamic Time Warping the time or pitch-scale modified speech. In our proposed algorithm for overcoming the transitional discontinuities, first of all, we don't change time or pitch scale and by using the LMR change a speaker's vocal tract characteristics in speech with non-modified time or pitch. Compared to existed methods based on VQ and LMR, we have much better voice quality in the result of the proposed algorithm.

• Proceedings of the Acoustical Society of Korea Conference
• /
• 1994.06c
• /
• pp.133-137
• /
• 1994

In this paper, a method for designing an optimal weight function for the weighted cepstral distance measure is proposed. A conventional weight function or cepstral lifter is obtained eperimentally depending on the spectral components to be emphasized. The proposed method minimizes the error between word reference patterns and the traning data. To compare the proposed optimal weight function with conventional function, speech recognition systems based on Dpynamic Time Warping and Hidden Markov Models were constructed to conduct speaker independent isolated word necogination eperiment. Results show that the proposed method gives better performance than conventional weight functions.

• Proceedings of the Korean Information Science Society Conference
• /
• 2001.04b
• /
• pp.202-204
• /
• 2001

본 논문에서는 대형 시퀀스 데이터베이스에서 타임 워핑을 지원하는 유사 검색을 효과적으로 처리하는 방안에 관하여 논의한다. 타임 워핑은 시퀀스의 길이가 서로 다른 경우에도 유사한 패턴을 갖는 시퀀스들을 찾을 수 있도록 해 주는 변환이다. 타임 워핑 거리는 삼각형 부등식 성질을 만족하지 못하므로 기존의 기법들은 착오 기각 없이 다차원 인덱스를 사용할 수 없었다. 본 논문에서는 타임 워핑을 지원하는 새로운 인덱스 기반 유사 검색 기법을 제안한다. 제안된 주요 목표는 착오 기각 없이 대형 데이터베이스에서도 좋은 검색 성능을 보장하는 것이다. 다양한 실험을 통하여 제안된 기법의 우수성을 규명한다. 실험 결과에 의하면, 제안된 기법은 기존의 기법과 비교하여 약 4배에서 43배까지의 성능 개선 효과를 가지는 것을 나타났다.

• Proceedings of the Korean Institute of Intelligent Systems Conference
• /
• 2005.04a
• /
• pp.419-423
• /
• 2005

This paper is a research on feature extraction and comparison method of dynamic (on-line) signature verification. We suggest desirable feature information and modified DTW(Dynamic Time Warping) and describe the performance results of our enhanced dynamic signature verification system.

• The Transactions of the Korean Institute of Electrical Engineers
• /
• v.40 no.4
• /
• pp.351-359
• /
• 1991

This paper describes performances of speaker adaptation using a probabilistic spectral mapping matrix in hidden-Markov model(HMM) -based Korean isolated word recognition. Speaker adaptation based on probabilistic spectral mapping uses a well-trained prototype HMM's and is carried out by Viterbi, dynamic time warping, and forward-backward algorithms. Among these algorithms, the best performance is obtained by using the Viterbi approach together with codebook adaptation whose improvement for isolated word recognition accuracy is 42.6-68.8 %. Also, the selection of the initial values of the matrix and the normalization in computing the matrix affects the recognition accuracy.

• Transactions of Materials Processing
• /
• v.20 no.3
• /
• pp.250-256
• /
• 2011

Vacuum-assisted thermoforming is one of the critical steps for the successful application of film insert molding(FIM) to parts of complex shapes. If the thickness distribution of the formed film is non-uniform, cracking, deformation, warping, and wrinkling can easily occur at the injection molding stage. In this study, the effects of processing parameters, which include the film heating time, plug depth, plug speed and vacuum delay time, on film thickness distribution were investigated. It was found that the film thickness at the part sidewall decreases with increasing the film heating time and plug depth, but the thickness at the bottom was found to exhibit the opposite behavior. The film thickness of the sidewall was observed to increase at higher plug speed and vacuum delay time of 0 ~ 0.3sec.

• 제어로봇시스템학회:학술대회논문집
• /
• 1993.10a
• /
• pp.870-876
• /
• 1993

This paper presents a binary clustering network (BCN) and a heuristic algorithm to detect pitch for recognition of keywords in continuous speech. In order to classify nonlinear patterns, BCN separates patterns into binary clusters hierarchically and links same patterns at root level by using the supervised learning and the unsupervised learning. BCN has many desirable properties such as flexibility of dynamic structure, high classification accuracy, short learning time, and short recall time. Pitch Detection algorithm is a heuristic model that can solve the difficulties such as scaling invariance, time warping, time-shift invariance, and redundance. This recognition algorithm has shown recognition rates as high as 95% for speaker-dependent as well as multispeaker-dependent tests.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2022.10a
• /
• pp.93-96
• /
• 2022

DTW can eliminate gaps between sequences of different lengths and find out the similarity of patterns, but due to the time and space complexity, it requires a high computational cost on large datasets. In this paper, we propose a DDTW algorithm that not only reduces computational costs but also has no error in the results. In addition, the algorithm complexity of DTW and DDTW is compared by measuring the computational time according to the length of the sequence. Simulation results show a noticeable reduction in computational time in DDTW compared to DTW.