• Journal of KIISE:Software and Applications
• /
• v.26 no.1
• /
• pp.157-157
• /
• 1999

Researches on the gesture recognition have become a very interesting topic in the computer vision area, Gesture recognition from visual images has a number of potential applicationssuch as HCI (Human Computer Interaction), VR(Virtual Reality), machine vision. To overcome thetechnical barriers in visual processing, conventional approaches have employed cumbersome devicessuch as datagloves or color marked gloves. In this research, we capture gesture images without usingexternal devices and generate a gesture trajectery composed of point-tokens. The trajectory Is spottedusing phase-based velocity constraints and recognized using the discrete left-right HMM. Inputvectors to the HMM are obtained by using the LBG clustering algorithm on a polar-coordinate spacewhere point-tokens on the Cartesian space .are converted. A gesture vocabulary is composed oftwenty-two dynamic hand gestures for editing drawing elements. In our experiment, one hundred dataper gesture are collected from twenty persons, Fifty data are used for training and another fifty datafor recognition experiment. The recognition result shows about 95% recognition rate and also thepossibility that these results can be applied to several potential systems operated by gestures. Thedeveloped system is running in real time for editing basic graphic primitives in the hardwareenvironments of a Pentium-pro (200 MHz), a Matrox Meteor graphic board and a CCD camera, anda Window95 and Visual C++ software environment.

• The Journal of the Acoustical Society of Korea
• /
• v.13 no.3
• /
• pp.25-32
• /
• 1994

This paper presents the Korean digit recognition method under noise environment using the spectral mapping training based on static supervised adaptation algorithm. In the presented recognition method, as a result of spectral mapping from one space of noisy speech spectrum to another space of speech spectrum without noise, spectral distortion of noisy speech is improved, and the recognition rate is higher than that of the conventional method using VQ (vector quatization) and DTW(dynamic time warping) without noise processing, and even when SNR level is 0dB, the recognition rate is 10 times of that using the conventional method. It has been confirmed that the spectral mapping training has an ability to improve the recognition performance for speech in noise environment.

• The KIPS Transactions:PartB
• /
• v.10B no.6
• /
• pp.673-680
• /
• 2003

In this paper, we study the enhancement of VQ (Vector Quantization) design for text independent speaker recognition. In a concrete way, we present a non-iterative method which makes a vector quantization codebook and this method performs non-iterative learning so that the computational complexity is epochally reduced The proposed Classified Space VQ (CSVQ) design method for text Independent speaker recognition is generalized from Semi-noniterative VQ design method for text dependent speaker recognition. CSVQ contrasts with the existing desiEn method which uses the iterative learninE algorithm for every traininE speaker. The characteristics of a CSVQ design is as follows. First, the proposed method performs the non-iterative learning by using a Classified Space Codebook. Second, a quantization region of each speaker is equivalent for the quantization region of a Classified Space Codebook. And the quantization point of each speaker is the optimal point for the statistical distribution of each speaker in a quantization region of a Classified Space Codebook. Third, Classified Space Codebook (CSC) is constructed through Sample Vector Formation Method (CSVQ1, 2) and Hyper-Lattice Formation Method (CSVQ 3). In the numerical experiment, we use the 12th met-cepstrum feature vectors of 10 speakers and compare it with the existing method, changing the codebook size from 16 to 128 for each Classified Space Codebook. The recognition rate of the proposed method is 100% for CSVQ1, 2. It is equal to the recognition rate of the existing method. Therefore the proposed CSVQ design method is, reducing computational complexity and maintaining the recognition rate, new alternative proposal and CSVQ with CSC can be applied to a general purpose recognition.

• Proceedings of the IEEK Conference
• /
• 2006.06a
• /
• pp.387-388
• /
• 2006

In this paper, we propose an efficient algorithm based on PCA and LDA combining K-means clustering method, which has better accuracy of face recognition than Eigenface and Fisherface. In this algorithm, PCA is firstly used to reduce the dimensionality of original face image. Secondly, a truncated face image data are sub-clustered by K-means clustering method based on Euclidean distances, and all small subclusters are labeled in sequence. Then LDA method project data into low dimension feature space and group data easier to classify. Finally we use nearest neighborhood method to determine the label of test data. To show the recognition accuracy of the proposed algorithm, we performed several simulations using the Yale and ORL (Olivetti Research Laboratory) database. Simulation results show that proposed method achieves better performance in recognition accuracy.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.19 no.6
• /
• pp.821-827
• /
• 2009

Human emotion can be reflected by their facial expressions. So, it is one of good ways to understand people's emotions by recognizing their facial expressions. General recognition system of facial expressions had selected interesting points, and then only extracted features without analyzing physical meanings. They takes a long time to find interesting points, and it is hard to estimate accurate positions of these feature points. And in order to implement a recognition system of facial expressions on real-time embedded system, it is needed to simplify the algorithm and reduce the using resources. In this paper, we propose a real-time recognition algorithm of facial expressions that project the grid points on an expression space based on Gabor wavelet feature. Facial expression is simply described by feature vectors on the expression space, and is classified by an neural network with its resources dramatically reduced. The proposed system deals 5 expressions: anger, happiness, neutral, sadness, and surprise. In experiment, average execution time is 10.251 ms and recognition rate is measured as 87~93%.

• The Transactions of the Korea Information Processing Society
• /
• v.6 no.4
• /
• pp.1129-1135
• /
• 1999

isolated word recognition using the Dynamic Time warping algorithm has shown good recognition rate on speaker dependent environment. But, practically, since the searching time of the dynamic Time Warping algorithm is rapidly increased as searching data is increased. it is hard to implement. In the context-dependent-short-query system such as educational children's workbook on the Web, the number of responses to the specific questions is limited. Therefore, the searching space for the answers can be reduced depending on the questions. In this paper, low cost implementation method using DTW for the Web has been proposed. To cover the weakness of DTW, the searching space is reduced by the context. the searching space, depends on the specific questions, is chosen from interest searchable candidates. In the real implementation, the proposed method show better performance of both time and recognition rate.

• Journal of Korea Multimedia Society
• /
• v.23 no.8
• /
• pp.1019-1029
• /
• 2020

In this paper, we propose a novel knowledge distillation algorithm to create an compressed deep ensemble network coupled with the combined use of local and global features of face images. In order to transfer the capability of high-level recognition performances of the ensemble deep networks to a single deep network, the probability for class prediction, which is the softmax output of the ensemble network, is used as soft target for training a single deep network. By applying the knowledge distillation algorithm, the local feature informations obtained by training the deep ensemble network using facial subregions of the face image as input are transmitted to a single deep network to create a so-called compressed ensemble DCNN. The experimental results demonstrate that our proposed compressed ensemble deep network can maintain the recognition performance of the complex ensemble deep networks and is superior to the recognition performance of a single deep network. In addition, our proposed method can significantly reduce the storage(memory) space and execution time, compared to the conventional ensemble deep networks developed for face recognition.

• Journal of the Institute of Electronics Engineers of Korea CI
• /
• v.47 no.1
• /
• pp.92-103
• /
• 2010

In this paper, a method for the recognition of dynamic hand gestures in various backgrounds using Dynamic Space Time Warping(DSTW) algorithm is proposed. The existing method using DSTW algorithm compares multiple candidate hand regions detected from every frame of the query sequence with the model sequences in terms of the time. However the existing method can not exactly recognize the models because a false path can be generated from the candidates including not-hand regions such as background, elbow, and so on. In order to solve this problem, in this paper, we use the invariant moments extracted from the candidate regions of hand and compare the similarity of invariant moments among candidate regions. The similarity is utilized as a weight and the corresponding value is applied to the matching cost between the model sequence and the query sequence. Experimental results have shown that the proposed method can recognize the dynamic hand gestures in the various backgrounds. Moreover, the recognition rate has been improved by 13%, compared with the existing method.

• KSCE Journal of Civil and Environmental Engineering Research
• /
• v.28 no.2D
• /
• pp.147-154
• /
• 2008

A variety of activity happens around the sidewalk in the city. Particularly, a large variety of activity happens in shopping area, but it causes an obstruction of economical revitalization since the pedestrians require time and cost to find what they want. So, this study will develop the path searching method to minimize the economical loss of shoppers by providing the significant path and supporting the walking movement. Firstly, consider existing network expression techniques and approach three points which are physical and environmental factor, the recognition of the pedestrians' space when changing the direction, and the recognition of restriction of vision and accessibility. Try to design the network DB and simulate the algorithm. As a result, it is now possible to do the path searching that considers variety of recognition factors and show the method how to make the path-searching algorithm for pedestrian.

• Journal of Internet Computing and Services
• /
• v.19 no.3
• /
• pp.43-48
• /
• 2018

In this paper, we propose an indoor-space recognition using DNN and super-pixel. In order to recognize the indoor space from the image, segmentation process is required for dividing an image Super-pixel is performed algorithm which can be divided into appropriate sizes. In order to recognize each segment, features are extracted using a proposed method. Extracted features are learned using DNN, and each segment is recognized using the DNN model. Experimental results show the performance comparison between the proposed method and existing algorithms.