• Journal of Internet Computing and Services
• /
• v.5 no.2
• /
• pp.69-73
• /
• 2004

A speaker model In speaker recognition system is to be trained from a large data set gathered in multiple sessions. Large data set requires large amount of memory and computation, and moreover it's practically hard to make users utter the data inseveral sessions. Recently the incremental adaptation methods are proposed to cover the problems, However, the data set gathered from multiple sessions is vulnerable to the outliers from the irregular utterance variations and the presence of noise, which result in inaccurate speaker model. In this paper, we propose an incremental robust adaptation method to minimize the influence of outliers on Gaussian Mixture Madel based speaker model. The robust adaptation is obtained from an incremental version of M-estimation. Speaker model is initially trained from small amount of data and it is adapted recursively with the data available in each session, Experimental results from the data set gathered over seven months show that the proposed method is robust against outliers.

• Journal of Navigation and Port Research
• /
• v.32 no.10
• /
• pp.851-856
• /
• 2008

Construction of large-scale container ports with the productivity improvements in container cranes shortened time of staying port to increase the level of service it harbors efforts accelerated. About container crane system exerted on the input, which is designed to look good performance considering the states feedback control system. The states observer designed of container cranes state variables that are expected to measurement noise or particular measurement signal. In the status of existing research, the feedback gain matrix and the state observer gain matrix are searched by being separated solving. But the feedback gain matrix and the state observer gain matrix are searched by RCGAs at once that be used robust search method in this paper.

• Proceedings of the Korea Water Resources Association Conference
• /
• 2015.05a
• /
• pp.377-377
• /
• 2015

부유사 자료는 유사이송해석에 필수적인 요소로 하천의 흐름 변화 및 하상 변동을 발생시키고, 하천 구조물의 설계, 수자원 개발 및 관리를 위한 하천계획의 전반에 있어 매우 중요한 자료이다. 부유사 농도는 수자원의 이용뿐만 아니라 하천 생태계에까지 피해를 미친다는 점에서 하천의 유지 관리 및 보수와도 밀접한 관련이 있다. 부유사량을 산정하는 방법에는 수리량 및 하상토 특성 자료를 유사량 공식에 대입하여 계산하는 간접적인 방법과 유사량을 직접 측정하는 방법으로 나뉜다. 현재 국내에서는 유사량 채집기를 사용하여 실제 하천의 유사량을 채집하는 방식으로 많이 사용되고 있으나, 많은 인력과 시간이 소모되기 때문에 다지점 계측과 지속적인 계측이 힘들다는 한계점를 보이고 있다. 또한 국내 하천에서는 홍수기를 거치면서 하천의 수리학적 특성이 변화하는 경향을 보여주고 있어 유량-부유사 관계식을 자주 갱신해야 한다는 어려움이 있다. 이에 본 연구에서는 현재 국내에서 사용하고 있는 직접적인 측정 방법의 한계점을 보완하고자 직접적인 측정 방법 중 초음파를 이용하여 횡단면 전체의 유사량 측정을 연속적으로 할 수 있는 수평초음파도플러유속계(H-ADCP)를 활용하여 유사량을 추정할 수 있는 기법을 개발하고자 한다. 본 연구의 연구는 건설기술연구원 하천실험센터의 직선수로에서 수행되었다. H-ADCP (SonTek SL3000, 셀 크기 4 cm)를 사용하여 자연상태 흐름조건 (유속 0.7 m/s)에서 초음파산란도(Backscatter, 혹은 신호대잡음비 SNR) 및 유속자료를 2분 간격으로 확보하였다. 그리고 부유사 농도(SSC)의 측정 정확도가 높다고 평가되고 있는 레이저부유사측정기(LISST-100)를 활용하여 부유사 농도를 실측하여 초음파산란도와 실측 SSC의 관계를 도출하고 그 경향을 분석하였다. 또한 초음파산란도의 흡수 등을 보정하고 실측 부유사자료와의 관계식을 기반으로 H-ADCP를 활용하여 실시간으로 부유사 농도를 산정할 수 있는 소프트웨어를 개발하였다.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.19 no.11
• /
• pp.31-37
• /
• 2018

This paper proposes an effective single-image haze-removal algorithm with low complexity by using a dual dark channel prior (DCP) and an adaptive brightness correction technique. The dark channel of a small patch preserves the edge information of the image, but is sensitive to noise and local brightness variations. On the other hand, the dark channel of a large patch is advantageous in estimation of the exact haze value, but halo effects from block effects deteriorate haze-removal performance. In order to solve this problem, the proposed algorithm builds a dual DCP as a combination of dark channels from patches with different sizes, and this meets low-memory and low-complexity requirements, while the conventional method uses a matting technique, which requires a large amount of memory and heavy computations. Moreover, an adaptive brightness correction technique that is applied to the recovered image preserves the objects in the image more clearly. Experimental results for various hazy images demonstrate that the proposed algorithm removes haze effectively, while requiring much fewer computations and less memory than conventional methods.

• The Journal of the Acoustical Society of Korea
• /
• v.40 no.3
• /
• pp.213-218
• /
• 2021

It is known that total least-squares method shows better estimation performance than least-squares method when noise is present at the input and output at the same time. When total least squares method is applied to data with time series characteristics, Recursive Total Least Squares (RTS) algorithm has been proposed to improve the real-time performance. However, RTLS has numerical instability in calculating the inverse matrix. In this paper, we propose an algorithm for reducing numerical instability as well as having similar convergence to RTLS. For this algorithm, we propose a new RTLS using Iterative Wiener Filter (IWF). Through the simulation, it is shown that the convergence of the proposed algorithm is similar to that of the RTLS, and the numerical robustness is superior to the RTLS.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.25 no.7
• /
• pp.910-916
• /
• 2021

In modern society, various digital communication equipment is being used due to the influence of the 4th industrial revolution. Accordingly, interest in removing noise generated in a data transmission process is increasing, and research is being conducted to efficiently reconstruct an image. In this paper, we propose a filtering algorithm to remove the AWGN generated in the digital image transmission process. The proposed algorithm classifies pixels with high similarity by selecting regions with similar patterns around the input pixels according to block matching to remove the AWGN that appears strongly in the image. The selected pixel determines the estimated value by applying the weight obtained by the local steering kernel, and obtains the final output by adding or subtracting the input pixel value according to the standard deviation of the center mask. In order to evaluate the proposed algorithm, it was simulated with existing AWGN removal algorithms, and comparative analysis was performed using enlarged images and PSNR.

• The Journal of the Institute of Internet, Broadcasting and Communication
• /
• v.22 no.3
• /
• pp.1-8
• /
• 2022

Recently, Google MedeiaPipe presents a novel approach for neural network-based hair segmentation from a single camera input specifically designed for real-time, mobile application. Though neural network related to hair segmentation is relatively small size, it produces a high-quality hair segmentation mask that is well suited for AR effects such as a realistic hair recoloring. However, it has undesirable segmentation effects according to hair styles or in case of containing noises and holes. In this study, the energy function of the test image is constructed according to the estimated prior distributions of hair location and hair color likelihood function. It is further optimized according to graph cuts algorithm and initial hair region is obtained. Finally, clustering algorithm and image post-processing techniques are applied to the initial hair region so that the final hair region can be segmented precisely. The proposed method is applied to MediaPipe hair segmentation pipeline.

• Journal of Digital Policy
• /
• v.2 no.3
• /
• pp.1-7
• /
• 2023

In this paper, I proposes a method for performing single image super resolution by separating texture-spatial domains and then classifying features based on detailed information. In CNN (Convolutional Neural Network) based super resolution, the complex procedures and generation of redundant feature information in feature estimation process for enhancing details can lead to quality degradation in super resolution. The proposed method reduced procedural complexity and minimizes generation of redundant feature information by splitting input image into two channels: texture and spatial. In texture channel, a feature refinement process with step-wise skip connections is applied for detail restoration, while in spatial channel, a method is introduced to preserve the structural features of the image. Experimental results using proposed method demonstrate improved performance in terms of PSNR and SSIM evaluations compared to existing super resolution methods, confirmed the enhancement in quality.

• Journal of Advanced Navigation Technology
• /
• v.26 no.6
• /
• pp.522-527
• /
• 2022

Object tracking is a field of signal processing that sequentially tracks the location of an object based on the previous-time location estimations and the present-time observation data. In this paper, we propose an adaptive scaling neural network that can track and adjust the scale of the input data with three recursive neural network (RNN) submodules. To evaluate object tracking performance, we compare the proposed system with the Kalman filter and the maximum likelihood object tracking scheme under an one-dimensional object movement model in which the object moves with piecewise constant acceleration. We show that the proposed scheme is generally better, in terms of root mean square error (RMSE) performance, than maximum likelihood scheme and Kalman filter and that the performance gaps grow with increased observation noise.

• The Journal of the Acoustical Society of Korea
• /
• v.43 no.2
• /
• pp.253-259
• /
• 2024

This paper proposes a new speech enhancement model based on a complex nested U-Net with a dual-branch decoder. The proposed model consists of a complex nested U-Net to simultaneously estimate the magnitude and phase components of the speech signal, and the decoder has a dual-branch decoder structure that performs spectral mapping and time-frequency masking in each branch. At this time, compared to the single-branch decoder structure, the dual-branch decoder structure allows noise to be effectively removed while minimizing the loss of speech information. The experiment was conducted on the VoiceBank + DEMAND database, commonly used for speech enhancement model training, and was evaluated through various objective evaluation metrics. As a result of the experiment, the complex nested U-Net-based speech enhancement model using a dual-branch decoder increased the Perceptual Evaluation of Speech Quality (PESQ) score by about 0.13 compared to the baseline, and showed a higher objective evaluation score than recently proposed speech enhancement models.