• 전기전자학회논문지
• /
• 제24권4호
• /
• pp.998-1003
• /
• 2020

본 논문에서는 기존 감정 인식 시스템의 성능 향상을 위하여 감정 적응을 사용한 감정 학습 방법이 제안되었다. 감정 적응을 위하여 적은 개수의 학습 감정 음성과 감정 적응 방식을 사용하여 감정이 없는 음성 모델로부터 감정 음성 모델이 생성되었다. 이러한 방법은 기존 방법보다 적은 개수의 감정 음성을 사용하여도 우수한 성능을 나타내었다. 학습을 위하여 충분한 감정 음성을 얻는 것은 쉽지 않기 때문에 적은 개수의 감정 음성을 사용하는 것은 실제 상황에서 매우 실용적이다. 4가지 감정이 포함된 한국어 데이터베이스를 사용한 실험 결과에서 감정 적응을 이용한 제안된 방법이 기존 방법보다 우수한 성능을 나타내었다.

• 한국음향학회지
• /
• 제25권1호
• /
• pp.36-42
• /
• 2006

본 논문에서는 화자검증 시스템의 성능향상을 위해서 주성분 분석 (PCA) 기반 Mel-Frequency Discrete Wavelet Coefficients (MFDWC) 추출방법을 제안한다. 제안된 방법에서는 멜척도 (Mel-scale)를 근사화한 각 레벨 (level)의 각 노드 (node) 에너지를 계산하기 위해 기존의 평균치 대신 주성분 분석을 이용한 첫 번째 eigenvector를 이용한다. 이 eigenvecto.의 제곱의 합은 1로서 일반적인 가중 함수 (weighting function)의 조건을 만족하고, 또한 각 화자마다 서로 다른 값을 갖게 되므로, 화자의 특징을 더 잘 나타내는 MFDWC를 추출할 수 있다. 화자검증은 Gaussian Mixture Model (GMM) 기반의 백그라운드 모델과 화자 모델과의 점수를 비교하는 이진 결정 (binary decision) 방법을 이용하여 Universal 백그라운드 모델 (UBM)과 각 화자 모델의 값을 프레임단위로 비교하여 대상 화자의 수락/거부 여부를 결정하는 방법을 채택하였다. 특징 파라미터에 따른 화자 검증 성능변화를 확인하기 위하여 제안된 화자종속 가중함수를 이용한 MFDWC를 특징 파라미터로 이용한 경우와 Mel-Frequency Cepstral Coefficients (MFCC), Linear Predictive Cepstral Coefficients (LPCC), 기존의 MFDWC를 특징 파라미터로 이용한 경우에 대하여 성능비교실험을 수행한 결과 각각 $0.80\%,\;5.14\%,\; 6.69\%$의 향상된 성능을 나타내어 제안한 방법의 유효성을 확인할 수 있었다.

• 대한임베디드공학회논문지
• /
• 제11권5호
• /
• pp.277-285
• /
• 2016

Moving object detection system has been an emerging research field in various advanced driver assistance systems (ADAS) and surveillance system. In this paper, we propose two optical flow based moving object detection methods at dynamic scenes. Both proposed methods consist of three successive steps; pre-processing, foreground segmentation, and post-processing steps. Two proposed methods have the same pre-processing and post-processing steps, but different foreground segmentation step. Pre-processing calculates mainly optical flow map of which each pixel has the amplitude of motion vector. Dense optical flows are estimated by using Farneback technique, and the amplitude of the motion normalized into the range from 0 to 255 is assigned to each pixel of optical flow map. In the foreground segmentation step, moving object and background are classified by using the optical flow map. Here, we proposed two algorithms. One is Gaussian mixture model (GMM) based background subtraction, which is applied on optical map. Another is adaptive thresholding based foreground segmentation, which classifies each pixel into object and background by updating threshold value column by column. Through the simulations, we show that both optical flow based methods can achieve good enough object detection performances in dynamic scenes.

• 한국컴퓨터정보학회논문지
• /
• 제22권12호
• /
• pp.79-85
• /
• 2017

In this paper, we propose a CNN-based swimmer detection algorithm. Every year, water safety accidents have been occurred frequently, and accordingly, intelligent video surveillance systems are being developed to prevent accidents. Intelligent video surveillance system is a real-time system that detects objects which users want to do. It classifies or detects objects in real-time using algorithms such as GMM (Gaussian Mixture Model), HOG (Histogram of Oriented Gradients), and SVM (Support Vector Machine). However, HOG has a problem that it cannot accurately detect the swimmer in a complex and dynamic environment such as a beach. In other words, there are many false positives that detect swimmers as waves and false negatives that detect waves as swimmers. To solve this problem, in this paper, we propose a swimmer detection algorithm using CNN (Convolutional Neural Network), specialized for small object sizes, in order to detect dynamic objects and swimmers more accurately and efficiently in complex environment. The proposed CNN sets the size of the input image and the size of the filter used in the convolution operation according to the size of objects. In addition, the aspect ratio of the input is adjusted according to the ratio of detected objects. As a result, experimental results show that the proposed CNN-based swimmer detection method performs better than conventional techniques.

• The Journal of Asian Finance, Economics and Business
• /
• 제8권11호
• /
• pp.191-201
• /
• 2021

This article examines the role of financial development in economic growth in a number of transitional economies where the financial systems were newly established or reformed only in the early 1990s to facilitate their transition from centrally planned economies to market-based ones. Based on a dataset collected from 29 transitional economies and 5 Asian developing economies covering the period 1990-2020, an empirical endogenous growth model is specified and estimated using the generalized method of moments (GMM). Three measures of financial development are used to investigate the relative role of the banking system and stock exchange market in the process of transition and growth. The results show that the three measures of financial development are crucial determinants of economic growth in transitional economies but the link seems to be in an inverted U-shape. This suggests the existence of thresholds for different channels of the financial sector to expand to positively influence growth. When becoming too large relative to the size of the economy, the financial system would have become a factor not conducive to growth. The growth convergence hypothesis is also confirmed and the impacts of other growth determinants are overall consistent with the extant literature.

• 한국컴퓨터정보학회논문지
• /
• 제23권11호
• /
• pp.51-57
• /
• 2018

In this paper, we studied about the extraction of the parameter and implementation of speechreading system to recognize the Korean 8 vowel. Face features are detected by amplifying, reducing the image value and making a comparison between the image value which is represented for various value in various color space. The eyes position, the nose position, the inner boundary of lip, the outer boundary of upper lip and the outer line of the tooth is found to the feature and using the analysis the area of inner lip, the hight and width of inner lip, the outer line length of the tooth rate about a inner mouth area and the distance between the nose and outer boundary of upper lip are used for the parameter. 2400 data are gathered and analyzed. Based on this analysis, the neural net is constructed and the recognition experiments are performed. In the experiment, 5 normal persons were sampled. The observational error between samples was corrected using normalization method. The experiment show very encouraging result about the usefulness of the parameter.

• 한국해양공학회지
• /
• 제28권6호
• /
• pp.552-559
• /
• 2014

In general, the number of underwater transient signals is very limited for research on automatic recognition. Data-dependent feature extraction is one of the most effective methods in this case. Therefore, we suggest WPCC (Wavelet packet ceptsral coefficient) as a feature extraction method. A wavelet packet best tree for each data set is formed using an entropy-based cost function. Then, every terminal node of the best trees is counted to build a common wavelet best tree. It corresponds to flexible and non-uniform filter bank reflecting characteristics for the data set. A GMM (Gaussian mixture model) is used to classify five classes of underwater transient data sets. The error rate of the WPCC is compared using MFCC (Mel-frequency ceptsral coefficients). The error rates of WPCC-db20, db40, and MFCC are 0.4%, 0%, and 0.4%, respectively, when the training data consist of six out of the nine pieces of data in each class. However, WPCC-db20 and db40 show rates of 2.98% and 1.20%, respectively, while MFCC shows a rate of 7.14% when the training data consists of only three pieces. This shows that WPCC is less sensitive to the number of training data pieces than MFCC. Thus, it could be a more appropriate method for underwater transient recognition. These results may be helpful to develop an automatic recognition system for an underwater transient signal.

• Journal of Korea Trade
• /
• 제25권4호
• /
• pp.146-163
• /
• 2021

Purpose - This paper examines China's impact on Korea's ICT exports considering the direct competition channel, the production shift channel, and the indirect demand channel at once. This paper also takes China's economic rebalancing into account and discusses whether it makes any differences in the effect of the three channels. Design/methodology - To quantify the effect of the three channels, I constructed a linear panel regression model and estimated it with various estimation methods including the system GMM. China's exports toward the same destination as Korea's exports, Korea's exports toward China, and the third countries' exports toward China respectively reflect the three channels. China's GVC indicators are included as well to evaluate the effect of further China's economic rebalancing. Since the present paper has a greater interest in the effect of China rather than the determinant of bilateral trade, a (fixed effect) panel model becomes more appropriate than the gravity model because timeinvariant variables in the gravity model, such as the distance and the language, are eliminated during the estimation process. Findings - The estimation results indicate that Chinese ICT exports are complementary to Korea's ICT exports in general. However, when markets are considered in subgroups, China's ICT exports could have a negative effect in the long run, especially for SITC75 and SITC76 markets, implying a possible competitive threat of China. The production shift effect turns significant during China's economic rebalancing in the markets for the advanced economies and the SITC76 product. China's indirect demand channel is also in effect significantly for the advanced economy and SITC75 commodities during China's economic rebalancing periods. In addition, this paper shows that China's transition toward upstream in the global value chain could have a positive impact on Korea's ICT exports, especially at the Asian market. Originality/value - The contribution of this paper is threefold. First, it focuses on the ICT industry for which Korea increasingly depends on China and China becomes a global hub of the GVC. Second, this paper quantitatively studies three channels in a model in contrast to the literature which mostly examines those channels separately and pays less attention to the GVC aspect. Third, by utilizing relatively recent data from the period of 2001-2017, this paper discusses whether China's economic rebalancing affects the three channels.

• 한국정보통신학회논문지
• /
• 제20권3호
• /
• pp.471-476
• /
• 2016

본 논문에서는 음성 특성 지표를 이용한 음성 인식 성능 예측 실험의 내용을 소개한다. 선행 실험에서 효과적인 음성 인식 성능 예측을 위해 대표적인 음성 인식 성능 지표인 단어 오인식률과 상관도가 높은 여러 가지 특성 지표들을 조합하여 새로운 성능 지표를 제안하였다. 제안한 지표는 각 음성 특성 지표를 단독으로 사용할 때 보다 단어 오인식률과 높은 상관도를 나타내 음성 인식 성능을 예측하는데 효과적임을 보였다. 본 실험에서는 이 결과를 근거하여 조합에 사용된 음성 특성 지표를 채택하여 4차원 특징 벡터를 생성하고 GMM 기반의 음성 인식 성능 예측기를 구축한다. 가우시안 요소를 증가시키며 실험한 결과 제안된 시스템은 babble 잡음, 자동차 잡음에서 모두 SNR이 낮을수록 단어 오인식률을 높은 확률로 예측함을 확인하였다.

• 대한전자공학회:학술대회논문집
• /
• 대한전자공학회 2004년도 ICEIC The International Conference on Electronics Informations and Communications
• /
• pp.237-243
• /
• 2004

A multichannel smart sound sensor capable to detect and identify sound events in noisy conditions is presented in this paper. Sound information extraction is a complex task and the main difficulty consists is the extraction of high­level information from an one-dimensional signal. The input of smart sound sensor is composed of data collected by 5 microphones and its output data is sent through a network. For a real time working purpose, the sound analysis is divided in three steps: sound event detection for each sound channel, fusion between simultaneously events and sound identification. The event detection module find impulsive signals in the noise and extracts them from the signal flow. Our smart sensor must be capable to identify impulsive signals but also speech presence too, in a noisy environment. The classification module is launched in a parallel task on the channel chosen by data fusion process. It looks to identify the event sound between seven predefined sound classes and uses a Gaussian Mixture Model (GMM) method. Mel Frequency Cepstral Coefficients are used in combination with new ones like zero crossing rate, centroid and roll-off point. This smart sound sensor is a part of a medical telemonitoring project with the aim of detecting serious accidents.