• 대한전기학회:학술대회논문집
• /
• 대한전기학회 2001년도 하계학술대회 논문집 D
• /
• pp.2622-2624
• /
• 2001

In this thesis, speaker separation is employed when speech from two talkers has been summed into one signal and it is desirable to recover one or both of the speech signals from the composite signal. This paper proposed the method that separated the summed speeches and proved the similarity between the signals by the cross correlation between the signals for exact between original signal and separated signal. This paper uses frequency sampling method based on sinusoidal model to separate the composite signal with vocalic speech and vocalic speech and noise masking method based on psycoacoustics model to separate the composite signal with vocalic speech and nonvocalic speech.

• 한국수산과학회지
• /
• 제55권2호
• /
• pp.195-206
• /
• 2022

The separation of spectrally overlapped broadband echo signals from free-swimming Japanese needlefish Hypohamphus sajori using the fractional Fourier transform (FrFT) was investigated. The broadband echo signals were measured over frequency ranges of 40-80 and 110-220 kHz. The overlapped echo signals were separated after eliminating noise signals in the smoothed pseudo-Wigner-Ville distribution domain. The echo signal from a 40 mm WC sphere suspended just below a chirp transducer was used to calibrate the broadband of the chirp echo sounder and estimate the frequency dependence of target strength for the separated echo signals. The experimental results show that the proposed FrFT method can analyze the time-frequency image of broadband echo signals from free-swimming individual fish effectively and can be used as a quantitative tool for extracting the acoustic features used for fish species identification.

• Journal of Information Processing Systems
• /
• 제19권6호
• /
• pp.858-869
• /
• 2023

As one of the pivotal techniques of image restoration, single-channel blind source separation (SCBSS) is capable of converting a visual-only image into multi-source images. However, image degradation often results from multiple mixing methods. Therefore, this paper introduces an innovative SCBSS algorithm to effectively separate source images from a composite image in various mixed modes. The cornerstone of this approach is a novel triple generative adversarial network (TriGAN), designed based on dual learning principles. The TriGAN redefines the discriminator's function to optimize the separation process. Extensive experiments have demonstrated the algorithm's capability to distinctly separate source images from a composite image in diverse mixed modes and to facilitate effective image restoration. The effectiveness of the proposed method is quantitatively supported by achieving an average peak signal-to-noise ratio exceeding 30 dB, and the average structural similarity index surpassing 0.95 across multiple datasets.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제9권10호
• /
• pp.4072-4091
• /
• 2015

A separation of text and non-text elements plays an important role in document layout analysis. A number of approaches have been proposed but the quality of separation result is still limited due to the complex of the document layout. In this paper, we present an efficient method for the classification of text and non-text components in document image. It is the combination of whitespace analysis with multi-layer homogeneous regions which called recursive filter. Firstly, the input binary document is analyzed by connected components analysis and whitespace extraction. Secondly, a heuristic filter is applied to identify non-text components. After that, using statistical method, we implement the recursive filter on multi-layer homogeneous regions to identify all text and non-text elements of the binary image. Finally, all regions will be reshaped and remove noise to get the text document and non-text document. Experimental results on the ICDAR2009 page segmentation competition dataset and other datasets prove the effectiveness and superiority of proposed method.

• 한국음향학회지
• /
• 제26권7호
• /
• pp.353-360
• /
• 2007

급속한 도시화와 인구밀도의 증가로 도시의 주거지역은 점점 고층화, 고밀화 되어가는 추세이다. 그로 인하여 인구의 공간적 집중으로 발생하는 교통수요를 감당하기 위하여 도시내부 도로는 점차 증가하였으며, 주거지역은 도로에 바로 인접하여 개발되고 있다. 이에 따라 도로에 연접한 주거지역의 도로교통소음 문제를 해결하기 위하여 방음벽에 의존하는 실정이었으나, 경관저해, 소음차단성능저해 등 다양한 역기능이 발생하고 있다. 본 연구에서는 방음벽의 역기능을 해소하고 바람직한 개선방안을 제시하기 위하여 도로변에 위치한 주거지역의 현장측정을 통하여 시뮬레이션의 정확도를 검증한 후 방음시설별 성능평가 및 전체 지역을 분석하였다. 그리고 개선방안으로 건축물의 배치형태 변화와 도로단으로부터의 이격거리를 20m 증가시켰다. 분석 결과 도로변과 평행한 공동주택의 피해가 가장 심각하였고, 시설별 분석에서는 방음벽의 성능이 우수하였으며, 건물 배치형태는 $30^{\circ}$ 배치가 우수하며, 이격거리가 증가함에 따라 교통소음이 감소하는 것으로 나타났다.

• 한국소음진동공학회:학술대회논문집
• /
• 한국소음진동공학회 2000년도 춘계학술대회논문집
• /
• pp.1513-1520
• /
• 2000

Complex modal testing method for rotating disks with support motion is introduced which handles the pairs of two point excitation and responses of the disk as complex input and output, respectively. This method utilizes the directivity information and the separation over the rotational speed of forward and backward traveling wave modes or bending coupled modes in the directional frequency response functions(dFRFs). This method synthesizes the normal/reverse dFRFs and complex wave dFRF, which were originally applied to rotating shaft and rotating disk, respectively, and is applied to complex system with dynamically coupled rotating disks and shaft. Experiments with a commercial hard disk drive spindle system demonstrate the validity of this method.

• 한국소음진동공학회:학술대회논문집
• /
• 한국소음진동공학회 2004년도 춘계학술대회논문집
• /
• pp.535-541
• /
• 2004

For housing estate of a new administrative capital city, Noise reduction method is a important design factor. As measuring a noise level of traffic noise according to separation from road, it can be created a quite housing estate. Analyzing of merits and demerits in sound barrier walls and tunnels can be proposed environment friendly soundproofing facilities. Number of measurement was performed to know what kind of layout of housing estate is good for noise reduction. Through this measurement, ㄷ shaped layout or parallel layout has the advantage of sound insulation rather than right angled layout. In this case (ㄷ shaped layout or parallel layout) buildings neighboring to the road should be designed to insulate sound. Evergreen trees should be planted between housing estate and road more than 30m (at least 7～8m) in order to reduce noise and have masking effects. If broad-leaved trees are planted more than 30m, approximately 10dB noise is reduced and 2～4dB if 7～8m. Roads in the estate should be designed considering pedestrians first, and special roads for moving and ambulance should be designed as skew road, if possible. The result shows that 15$^{\circ}$-sloped‘S’road reduces 1～2dB noise and 30$^{\circ}$-sloped road reduces 4～7dB. If noise barrier is inevitably installed, it should be designed to go well wit neighboring environment so as to install Environment Friendly Noise Barrier using materials and trees including wood and soil. Through this study the results are used to guideline for construction of environment friendly housing estate

• 한국소음진동공학회:학술대회논문집
• /
• 한국소음진동공학회 2010년도 추계학술대회 논문집
• /
• pp.201-207
• /
• 2010

In this study, the SIP (Soil-cement Injected precast Pile) method among the Low-vibration & Low-noise pile driving methods was decided into study compensation. Ground vibrations by the SIP methods step by step divide and were analyzed. Quantitative response values and ground vibration equations with reliability were presented from findings of this study. Also, vibration responses that are occurred by the SIP method of construction were compared as quantitative with vibration responses by general method of construction that are presented in existent study. Ground vibration values by the SIP method correspond to level of 17 ~ 57% of values that are assumed by the Attewell & Famer's equation, respectively, and these result compares in reliability 50% and separated distance 10 ~ 50 m. Also, those values were analyzed that correspond to level of 14 ~ 96% of ground vibration values by the Prof. Park's equation, respectively. Construction limit extents, separation distances from vibration occurs position, were presented that can satisfy domestic criteria for vibration control for the SIP methods. Those presented in this paper were divided newly according to reliability.

• 한국정보과학회논문지:소프트웨어및응용
• /
• 제26권1호
• /
• pp.130-130
• /
• 1999

In this paper, we propose a novel design method for improving performance of existing FCM-type clustering algorithms. First, we define the performance measure which focuses on bothcompactness and separation of clusters. Next, we optimize this measure using evolution program.Especially the proposed method has following merits: ① using evolution program, it solves suchproblems as initialization, number of clusters, and convergence to local optimum ② it reduces searchspace and improves convergence speed of algorithm since it represents chromosome with possiblepotential centers which are selected possible candidates of centers by density measure ③ it improvesperformance of clustering algorithm with the performance index which embedded both compactnessand separation Properties ④ it is robust to noise data since it minimizes its effect on center search.

• Structural Engineering and Mechanics
• /
• 제78권3호
• /
• pp.369-378
• /
• 2021

This paper reviewed a few output-only system identification algorithms and identified the shortcomings of those popular blind source separation methods. To address the issues such as less sensors than the targeted modal modes (under-determinate problem), repeated natural frequencies as well as systems with complex mode shapes, this paper proposed a complex wavelet modified second order blind identification method (CWMSOBI) by transforming the time domain problem into time-frequency domain. The wavelet coefficients with different dominant frequencies can be used to address the under-determinate problem, while complex mode shapes are addressed by introducing the complex wavelet transformation. Numerical simulations with both high and low signal-to-noise ratios validate that CWMSOBI can overcome the above-mentioned issues while obtaining more accurate identified results than other blind identification methods.