• 한국소음진동공학회:학술대회논문집
• /
• 한국소음진동공학회 2013년도 추계학술대회 논문집
• /
• pp.316-317
• /
• 2013

• 한국소음진동공학회:학술대회논문집
• /
• 한국소음진동공학회 2014년도 춘계학술대회 논문집
• /
• pp.254-254
• /
• 2014

에어컨 실내기는 소비자와 근접해 있는 실내에 설치되며, 필요성에 의해 가동되므로 소비자들의 소음에 대한 반응이 민감하다. 실내기 소음에는 Fan Noise, Cricking Noise, Refringerant Noise 등이 존재한다. 이중 Refringerant Noise 의 경우 Overall Level 은 낮지만 거슬리는 소음이 발행하여 소비자의 가장 큰 불만을 초래한다. Refringerant Noise 에 대해 발생원인 및 해결 방안 측면으로의 연구는 활발히 진행 중이다. 그러나, 소음에 대한 평가 방법에 대한 연구가 이루어지지 않고 있다. 이에 본 연구에서는 에어컨 실내기 Refringerant Noise 을 대상으로 각 소음 별 특성 분석을 수행하였다. 이를 바탕으로 각 소음 별 영향인자를 도출하였다.

• 전자공학회논문지B
• /
• 제29B권7호
• /
• pp.70-78
• /
• 1992

In this paper, an approach for 3-D object segmentation and classification, which is based on eigen-values of polynomial function as their surface features, using neural network is proposed. The range images of 3-D objects are classified into surface primitives which are homogeneous in their intrinsic eigenvalue properties. The misclassified regions due to noise effect are merged into correct regions satisfying homogeneous constraints of Hopfield neural network. The proposed method has advantage of processing both segmentation and classification simultaneously.

• 제어로봇시스템학회:학술대회논문집
• /
• 제어로봇시스템학회 1992년도 한국자동제어학술회의논문집(국내학술편); KOEX, Seoul; 19-21 Oct. 1992
• /
• pp.100-105
• /
• 1992

Texture is one of the important characteristics that has been used to identify objects or regions of interest in an image. Tire tread patterns can be considered as a kind of texture, and these are classified with a texture analysis method. In this sense, this paper proposes a new algorithm for the classification of tire tread pattern. For the classification, cooccurrence matrix for the binary image is used. The performances are tested by experimentally 8 different tire tread pattern and the robustness is examined by including some kinds on noise.

• 한국소음진동공학회:학술대회논문집
• /
• 한국소음진동공학회 2007년도 추계학술대회논문집
• /
• pp.273-277
• /
• 2007

In case of the Hydraulic Turbine Dynamo operating for Waterpower Generation, it makes very huge and loud noises, and it influences bad effect physically as same as mentally to those people who are working inside of power plant, and brings the decline of an effective working efficiency. However, its evaluation method or measure about such noise reflects merely its physical attribute which is sensuous Loudness of the Noise itself, since the accumulation effect of Noise or the meaning connected with psychological response did not reflect, it is the actual state that a rational evaluation is unable to expect. Consequently, this Study has attempted to evaluate the Noise of Hydraulic Turbine Dynamo by analyzing the sound quality using Zwicker‘s Psychological Acoustic Parameter, after classification by its positions of the Noise occurring at Hydraulic Turbine Dynamo.

• 대한원격탐사학회지
• /
• 제37권4호
• /
• pp.719-731
• /
• 2021

This study proposes a two-stage hybrid classification model for crop classification using multi-temporal remote sensing images; the model combines feature embedding by using an autoencoder (AE) with a convolutional neural network (CNN) classifier to fully utilize features including informative temporal and spatial signatures. Long short-term memory (LSTM)-based AE (LAE) is fine-tuned using class label information to extract latent features that contain less noise and useful temporal signatures. The CNN classifier is then applied to effectively account for the spatial characteristics of the extracted latent features. A crop classification experiment with multi-temporal unmanned aerial vehicle images is conducted to illustrate the potential application of the proposed hybrid model. The classification performance of the proposed model is compared with various combinations of conventional deep learning models (CNN, LSTM, and convolutional LSTM) and different inputs (original multi-temporal images and features from stacked AE). From the crop classification experiment, the best classification accuracy was achieved by the proposed model that utilized the latent features by fine-tuned LAE as input for the CNN classifier. The latent features that contain useful temporal signatures and are less noisy could increase the class separability between crops with similar spectral signatures, thereby leading to superior classification accuracy. The experimental results demonstrate the importance of effective feature extraction and the potential of the proposed classification model for crop classification using multi-temporal remote sensing images.

• 환경영향평가
• /
• 제20권2호
• /
• pp.123-131
• /
• 2011

The objective of this study is to prepare the effective management plan of environmental noise with a noise map, and the guideline on the management plan of environmental noise was suggested through the review of existing application examples. The management plan of environmental noise with a noise map includes the practical contents in the stages of subdivision of management areas, establishment of reduction measures, opinion collection, post investigation, and reformulation of management plan. First, the classification of management regions is performed considering the excess degree of noise standard and the facility type in the phase of subdivision of management areas. Second, the optimal management plan is established through the investigation of regional characteristics and various noise reduction measures in the phase of establishment of reduction measures, which includes the examination of noise reduction effects with a noise map and the budget planning with the costing of noise reduction measures. Third, the opinion survey with a local resident and a expert is carried out in order to prove the validity of the management plan in the phase of opinion collection, and the management plan is modified with gathered opinions. Fourth, the post examination plan with noise measurement is performed in order to verify the real effect of noise reduction measures according to the management plan in the phase of post investigation. Finally, the amendment of the management plan as well as the improvement of a noise map is carried out at a regular cycle in the phase of reformulation of management plan.

• 한국음향학회지
• /
• 제38권1호
• /
• pp.39-46
• /
• 2019

본 논문에서는 합성곱 신경망(Convolutional Neural Network, CNN) 구조를 이용하여 잡음 환경에서 음향신호를 분류할 때, 인식률을 높이는 결합 특징을 제안한다. 반면, Wiener filter를 이용한 강인한 log Mel-filter bank와 PNCCs(Power Normalized Cepstral Coefficients)는 CNN 구조의 입력으로 사용되는 2차원 특징을 형성하기 위해 추출됐다. 자연환경에서 43종의 조류 울음소리를 포함한 ebird 데이터베이스는 분류 실험을 위해 사용됐다. 잡음 환경에서 결합 특징의 성능을 평가하기 위해 ebird 데이터베이스를 3종류의 잡음을 이용하여 4개의 다른 SNR (Signal to Noise Ratio)(20 dB, 10 dB, 5 dB, 0 dB)로 합성했다. 결합 특징은 Wiener filter를 적용한 log-Mel filter bank, 적용하지 않은 log-Mel filter bank, 그리고 PNCC와 성능을 비교했다. 결합 특징은 잡음이 없는 환경에서 1.34 % 인식률 향상으로 다른 특징에 비해 높은 성능을 보였다. 추가적으로, 4단계 SNR의 잡음 환경에서 인식률은 shop 잡음 환경과 schoolyard 잡음 환경에서 각각 1.06 %, 0.65 % 향상했다.

• 한국음향학회:학술대회논문집
• /
• 한국음향학회 1997년도 영남지회 학술발표회 논문집 Acoustic Society of Korean Youngnam Chapter Symposium Proceedings
• /
• pp.68-74
• /
• 1997

Ship radiated noise is an infortant parameter which dtermines Anti Submarine Warfare(ASW) countermeansure or passive Sonar detection and classification performance. Its measurement should be performed under controlled ocean acoustic environment. In data reduction of the measured data from hydrophone array, theeffect fo ambient noise, surface reflection and bottom reflection etc. should be compensated to obtain the source level of the ship radiated noise. This study describes the measurement hydrophone array design criteria based on the analysis of transimission anomaly due to the surface reflection.

• 융합보안논문지
• /
• 제17권3호
• /
• pp.21-29
• /
• 2017

지역 패턴을 정확하게 부호화 하는 방법은 텍스처 분류 연구에 매우 중요한 요소다. 하지만 기존 널리 연구된 LBP기반 방법들은 잡음에 취약한 근본적인 문제점이 있다. 최근 표정인식 분야에서 에지반응 값과 방향 정보를 활용한 LDP방법이 제안되었다. LDP방법은 LBP보다 잡음에 강하고 더 많은 정보를 코드에 수용할 수 있는 장점이 있지만 텍스처 분류에 적용하기에는 치명적인 회전 변화에 민감한 단점이 있다. 본 논문에서는 LDP 방법에 회전 불변 특성을 결합하고 기존 LDP가 가지고 있던 부호 정보를 수용하지 않은 단점과 밝기 값 차이가 적은 영역에서 의미 없는 코드가 생성되는 단점을 극복한 새로운 지역 패턴 부호화 방법인 Rotation Invariant Local Directional Pattern 방법을 제안한다. 본 논문에서 제안된 방법의 텍스처 분류 성능을 입증하기 위해 널리 사용되는 UIUC, CUReT 데이터 셋에서 텍스처 분류를 수행했다. 그 결과 제안된 RILDP방법이 기존 방법보다 우수한 성능을 보여주었다.