• 대한전자공학회논문지
• /
• 제25권9호
• /
• pp.1019-1026
• /
• 1988

A parametric detection scheme for determenistic signals is obtained in a generalized observation model which contains non-additive noise. The model employed in this paper includes several special cases such as those describing purely-additive noise, multiplicative noise, and signal dependent noise and allows the consideration of deterministic and random signals. Locally optimum detectors for known deterministic signals in the model are derived and analyzed for performance. It is shown that the locally optimum detectors are interesting generalizations of those for the purely-additive noise model. Performance of the locally optimum detectors designed for the generalized observation model is compared to that of other common detectors.

• 한국컴퓨터정보학회논문지
• /
• 제29권5호
• /
• pp.21-29
• /
• 2024

CT 촬영 시 방사선량을 줄이면 피폭 위험성을 낮출 수 있으나, 영상 해상도가 크게 저하 될 뿐아니라 잡음(noise) 발생으로 인해 진단의 효용성이 떨어진다. 따라서, CT 영상에서의 잡음제거는 영상복원 분야에 있어 매우 중요하고 필수적인 처리 과정이다. 영상 영역에서 잡음과 원래 신호를 분리하여 잡음만을 제거하는 것은 한계가 있다. 본 논문에서는 웨이블릿 변환 기반 GAN 모델 즉, WT-GAN(wavelet transform-based GAN) 모델을 이용하여 CT 영상에서 효과적으로 잡음 제거하고자 한다. 여기서 사용된 GAN 모델은 U-Net 구조의 생성자와 PatchGAN 구조의 판별자를 통해 잡음제거 영상을 생성한다. 본 논문에서 제안된 WT-GAN 모델의 성능 평가를 위해 다양한 잡음, 즉, 가우시안 잡음(Gaussian noise), 포아송 잡음 (Poisson noise) 그리고 스펙클 잡음 (speckle noise)에 의해 훼손된 CT 영상을 대상으로 실험하였다. 성능 실험 결과, WT-GAN 모델은 전통적인 필터 즉, BM3D 필터뿐만 아니라 기존의 딥러닝 모델인 DnCNN, CDAE 모형 그리고 U-Net GAN 모형보다 정성적이고, 정량적인 척도 즉, PSNR (Peak Signal-to-Noise Ratio) 그리고 SSIM (Structural Similarity Index Measure) 면에서 우수한 결과를 보였다.

• The Journal of the Acoustical Society of Korea
• /
• 제18권3E호
• /
• pp.3-8
• /
• 1999

In speech recognition for real-world applications, the performance degradation due to the mismatch introduced between training and testing environments should be overcome. In this paper, to reduce this mismatch, we provide a hybrid method of spectral subtraction and residual noise masking. We also employ multiple model approach to obtain improved robustness over various noise environments. In this approach, multiple model sets are made according to several noise masking levels and then a model set appropriate for the estimated noise level is selected automatically in recognition phase. According to speaker independent isolated word recognition experiments in car noise environments, the proposed method using model sets with only two masking levels reduced average word error rate by 60% in comparison with spectral subtraction method.

• 한국소음진동공학회:학술대회논문집
• /
• 한국소음진동공학회 2007년도 추계학술대회논문집
• /
• pp.12-22
• /
• 2007

This paper presents to predict the powertrain structure-borne noise which is primary resource of interior noise. As the first step, it is built up a hybrid powertrain model which is based on the real powertrain which is verified with static and dynamic properties. The methods for verifying are modal analysis and running vibration testing which are experimentally implemented. Based on the Hybrid powertrain component model, an initial predictive assembly model is simulated. As the second step, the characteristic transfer functions are measured that are dynamic stiffness of rubber mounts and vibro-acoustic transfer function based on the acoustic reciprocity. Several techniques utilizing special experimental devices have been proposed for this research. Finally, the structure-borne noise by powertrain will be predict and verify with dynamic simulation and experiment.

• 제어로봇시스템학회:학술대회논문집
• /
• 제어로봇시스템학회 1995년도 Proceedings of the Korea Automation Control Conference, 10th (KACC); Seoul, Korea; 23-25 Oct. 1995
• /
• pp.43-46
• /
• 1995

Adaptive algorithms based on gradient adaptation have been extensively investigated and successfully jointed with active noise/vibration control applications. The Filtered-X LMS algorithm became one of the basic feedforward algorithms in such applications, but still is not fully understood. The error path model effect on the Filtered-X LMS algorithm has been under the investigation and some useful properties related stability has been discovered. We are interested in utilizing the fact that the model error caused by the way optimizing the error path model in a view point of convergence speed of Filtered-X LMS algorithm for pure tone noise suppression application without any performance loss at steady state.

• 한국자동차공학회논문집
• /
• 제1권3호
• /
• pp.55-62
• /
• 1993

Autoregressive moving average(ARMA) model which is a time domain parametric modeling method is implemented for modeling and reproducing characteristics of exhaust noise of an automobile in various RPM range. Experiments have been carried out using 9 set of exhaust noise signals measured at 1,000-3,000 RPM range. Characteristics of sampled signals were estimated using ARMA modeling and Akaike's FPE(final prediction error) criterion to define exact model structure and for model validation. The digital filter consisted of the esitmated ARMA(70,1) model parameters was programed to reproduce exhaust noise. The spectral analysis of reproduced noise is very close to original. The results show that our approaching technique for reproducing acoustic characteristics is valid and feasible to apply in the field of noise quality control.

• 소음진동
• /
• 제9권6호
• /
• pp.1131-1136
• /
• 1999

NIC@E is the software developed by authors. The program provides the noise level in outdoors due to various noise source types : construction machines including blast sources, railroad vehicles and automobiles. It operates in the Windows system. In this paper, a highway traffic noise has been evaluated using various types of approach : Ray-tracing method, NIRI method, JAS method. In order to compare the noise estimation performance for various models, a measurement is conducted on a 8 lane express highway at the distance of 25 m and 50 m from the lane. The result shows that the ray tracing and JAS model predict the measured value well within 2dB deviaton. The NIRI model, however, underestimates the highway noise level, as the distance between the source and receiver increases.

• The Journal of the Acoustical Society of Korea
• /
• 제20권2E호
• /
• pp.9-16
• /
• 2001

A psychoacoustic model based noise shaping method which shapes the noise in the frequency domain is proposed, where its presence with a host signal will not be perceptually noticeable. The derivation of imperceptible noise levels from the masking thresholds of the signal involves a deconvolution associated with the spreading function in the psychoacoustic model, which results in an ill-conditioned problem. In this paper, the problem is formulated as a constrained optimization, and it is demonstrated that the solution provides noise shaping where the noise excitation level conforms to the masking thresholds of the signal, and thus the noises embedded in the signal will not be perceived by human ear.

• 한국소음진동공학회:학술대회논문집
• /
• 한국소음진동공학회 2012년도 춘계학술대회 논문집
• /
• pp.763-768
• /
• 2012

Active window system which can reduce the environmental noises, such as traffic noise and construction noise, from an open window into a room was proposed in the previous works. The key idea of the proposed active window system was that the control sources are approximately collocated with the primary noise source in terms of the acoustic power for global noise reduction throughout the interior room. Moreover, because it is important not to intrude into the living space in the building environment, no error sensors were used and an open-loop control method using control sources at the window frame and the reference sensors outside the room was used for the proposed system. The open-loop control gain was calculated by the interior room model assumed as the semi-infinite space, and the interior sound field was estimated by Rayleigh integral equation under the baffled window model assumption. However, windows with a finite thickness should were considered for the calculation of the open-loop control gain of the active window system since these are representative of most window cases. Therefore, the finite thickness window model based on the Sgard's model was derived and the open-loop control gain using the interior sound field estimated by that model was calculated for active window system. To compare the performance of these two models, a scale-model experiment was performed in an anechoic chamber according to noise source directions. Experimental results showed that the performance for the thickness window model is better than the baffled window model as the angle with respect to the perpendicular direction is larger.

• 대한교통학회지
• /
• 제5권2호
• /
• pp.73-80
• /
• 1987

Various procedures for evaluation of traffic noise annoyance have been proposed. However, most of the studies of this type are restricted for improving traffic flow. In this paper, a method to predict the road traffic noise is proposed in terms of equivalent continuous A-Weighted sound pressure level (Leq), based on a probability model. First, distribution of the road traffic noise level are investigated. second, the weibull distribution parameters are estimated by using the quantification theory. Finally, a prediction model of the road traffic noise is proposed based on the weibull distribution model The predicted values of the Leq are closely matched the measured data.