• Proceedings of the KSRS Conference
• /
• 2008.03a
• /
• pp.237-240
• /
• 2008

본 연구의 목적은 인공신경망 기법을 이용하여 2006년 전국을 대상으로 구축된 국토환경성평가지도의 환경 생태적항목에 대한 각 항목별 가중치를 결정하는 것이다. 기본 분석 도구로 지리정보시스템(GIS)가 사용되었다. 국토환경성평가지도의 환경 생태적항목은 다양성(생태자연도), 자연성(임상도, 녹지자연도, 생태자연도), 풍부도(생태계변화관찰 지역도), 희귀성(생태자연도), 허약성(수치지형도, 토지피복도), 군집구조의 안정성(임상도)등이 활용되어 구축되었다. 본 연구는 기 구축되어 사용되고 있는 국토환경성평가 지도의 환경 생태적항목을 공간 데이터베이스를 이용하고, 인공신경망 기법을 적용하여 각 평가항목간의 상대적 가중치를 구하였다. 인공신경망의 훈련 지역은 환경 생태적항목중 환경성이 높은 1등급 지역 및 환경성이 낮은 5등급 지역을 추출하였다. 그 결과 50번의 가중치를 산정하였을 경우 허약성이 다른 항목들에 비해 1.58배 정도 높은 상대적 가중치를 나타냈다. 이러한 가중치는 국토환경성평가지도 환경 생태적 항목의 취약성도를 작성하는데 활용될 수 있다.

• Proceedings of the KWS Conference
• /
• 2009.11a
• /
• pp.111-111
• /
• 2009

전 세계적으로 환경 보호의 차원에서 자동차 업체는 자동차의 연비 향상을 위한 차체의 경량화가 큰 이슈로 대두되고 있다. 이를 위해 알루미늄과 같은 경량화 소재를 이용하여 차체 조립에 투입하고자 연구 중에 있다. 이와 같은 레이저 용접 공정이 현장에 적용되기 위해서는 용접부의 품질을 실시간으로 모니터링하고 품질을 판단하여야 생산성을 극대화 할 수 있다. 그러므로 본 연구에서는 알루미늄 AA5182 알루미늄 판재의 용가 와이어를 이용한 레이저 용접에서 용접부를 모니터링 할 수 있는 시스템을 구축하였다. 이를 위하여 레이저는 4kW급 Nd:YAG 레이저를 사용하였고, 차체용 알루미늄 판재 AA5182 1.4t를 AA5356 와이어를 이용하여 용접을 수행하였다. 모니터링 센서로는 반응 범위가 190 mn~680 nm인 센서를 이용하였고, 용접 중 센서로부터 발생된 출력전류를, 신호 증폭기와 DAQ 보드를 통해 초당 10,000 samples/sec로 계측하였다. 다양한 용접조건을 이용하여 실험을 수행하였고 이를 정량적으로 분석하였다. 계측된 신호와 용접 품질은 비선형적 관계를 가지고 있으므로 본 연구에서는 용접 품질을 예측하는 방법으로 퍼지 패턴인식 알고리즘을 이용하는 방법과 계측 신호를 이용한 인장강도 예측모델을 이용하여 병렬로 품질평가를 할 수 있는 알고리즘을 구현하였다. 이를 위하여 계측된 신호와 용접 품질과의 관계를 이용하여 퍼지 규칙 베이스 정의하였고, 신경회로망 모델을 이용하여 인장강도 예측모델을 제시하였다. 또한 품질 평가 알고리즘을 기반으로 레이저 용접부의 품질평가가 가능한 GUI 프로그램을 구현하였다.

• Transactions of the Korean Society of Mechanical Engineers A
• /
• v.32 no.11
• /
• pp.990-996
• /
• 2008

Nowadays, the trend to a lightweight design accelerates the use of advanced high strength steel (AHSS) in automotive industry. Springback phenomena is a hot issue in the sheet metal forming, especially bending process using AHSS. Several analytical methods for that have been proposed in recent years. Each of method has their advantages and disadvantages. There are only a few optimal solutions which can minimize the two objectives simultaneously. In this study, an effective method optimized the multi objective value. The method by the design of experiments(DOE) and artificial neural network(ANN) was presented to compensate springback of bending parts. This method was applied to L and V bending process. The effective method could be optimized to multiple object. It was confirmed that the proposed method was more efficient than traditional manual FEA procedure and the trial and error approach for springback compensation.

• Proceedings of the Korean Society for Agricultural Machinery Conference
• /
• 1996.06c
• /
• pp.773-779
• /
• 1996

Dried oak mushroom have complex and various visual features. Grading and sorting of dried oak mushrooms has been done by the human expert. Though actions involved in human grading looked simple, a decision making underneath the simple action comes from the result of the complex neural processing of the visual image. Through processing details involved in human visual recognition has not been fully investigated yet, it might say human can recognize objects via one of three ways such as extracting specific features or just image itself without extracting those features or in a combined manner. In most cases, extracting some special quantitative features from the camera image requires complex algorithms and processing of the gray level image requires the heavy computing load. This fact can be worse especially in dealing with nonuniform, irregular and fuzzy shaped agricultural products, resulting in poor performance because of the sensitiveness to the crisp criteria or specific ules set up by algorithms. Also restriction of the real time processing often forces to use binary segmentation but in that case some important information of the object can be lost. In this paper, the neuro net based real time recognition algorithm was proposed without extracting any visual feature but using only the directly captured raw gray images. Specially formated adaptable size of grids was proposed for the network input. The compensation of illumination was also done to accomodate the variable lighting environment. The proposed grading scheme showed very successful results.

• Progress in Superconductivity
• /
• v.4 no.1
• /
• pp.53-58
• /
• 2002

Noise from electronic instrumentation is invariably present in biomedical signals, although the art of instrumentation design is such that this noise source may be negligible. And sometimes signals of interest are contaminated or degraded by signals of similar type from another source. Biomedical signals are omni-presently contaminated by these background noises that span nearly all frequency bandwidths. In the magneto-cardiogram (MCG), several digital filters have been designed for the elimination of the power-line interference, broadband white noise, surrounding magnetic noise, and baseline wondering. In addition to the introduced FIR filter, notch, adaptive filter using the least mean square (LMS) algorithm, and recurrent neural network (RNN) filter, a new filtering method for effective noise canceling in MCG signals is proposed in this paper, which is realized by the wavelet packets. The experimental results show that the proposed filter using wavelet packet performs efficiently with respect to noise rejection. To verify this, two characteristics were analyzed and compared with LMS adaptive filter, SNR of filtered signal and attractor pattern using the nonlinear dynamics.

• Journal of Korean Society of Water and Wastewater
• /
• v.27 no.2
• /
• pp.197-206
• /
• 2013

Recently, climate change causes climatic anomaly such as global warming, the typhoon and severe rain storm etc. and it brings damage frequently. Climate change and global warming are prevalent all over the world in this century and many researchers including hydrologists have studied on the climate change. In this study, Seonakdong river watershed in the Nakdong river basin was selected as a study area. Real-time monitoring system was used to draw the rating curves, which has 0.78 to 0.96 of $R^2$. To predict runoff change in Seonakdong river watershed caused by climate change, the change in hydrologic runoff were predicted using the watershed model, SWAT. As a result, the runoff from the Seonakdong river watershed was increased by up to 45 % in summer. Because of the non-point sources from the farmland and the urban area, the water quality will be affected by the climate change. In this study, the operating plan of the water gates in Seonakdong river will be suggested by considering the characteristics of the watershed runoff due to the climate change. The optimal watergate opening plan will solve the water pollution problems in the reservoir-like river.

• Proceedings of the Korean Society for Noise and Vibration Engineering Conference
• /
• 2006.05a
• /
• pp.1437-1442
• /
• 2006

The international competition in car markets has continuously required the research about the sound quality of a car. The domestic carmakers have also invested a lot of money for the research and development of interior sound quality of passenger cars. Therefore, the aim of this research is to predict the customer's evaluation of a new vehicle. There are two major research works to achieve this goal in this research. The first one is to search questionnaires about the sound quality, which customers prefer, to identify the relationship between these questionnaires and sound metrics that is a psychoacoustics parameters, and to development sound indexes for the questionnaires. All tests for this work is proceed on the road test during acceleration. The second one is to balance the sound component (engine noise, booming noise, road noise and wind noise) of a passenger. This wok will be tested on the constant speed. All of research results will be contributed to the development of brand sound quality of a new passenger car.

• Proceedings of the Korea Water Resources Association Conference
• /
• 2004.05b
• /
• pp.1208-1212
• /
• 2004

다목적댐을 효율적이고 체계적으로 운영하기 위해서는 수문순환에 대한 지역별, 기간별 이해와 더불어 댐저수지로의 정확한 유입량 산정이 필요하다. 수문모델링을 비교하기 위해서는 개념적 모형과 추계학적 모형으로 나눌 수 있는데 개념적 모형은 상당히 많은 입력요소로 말미암아 사용자로 하여금 이해를 하는데 있어서 어려움을 겪을 수 밖에 없는 실정이나 추계학적 모형은 확률적 철상 및 기초적 예측이론을 습득하게 되면 쉽고 간단하여 검토를 용이하게 할 수 있는 장점이 있다. 수자원시스템의 설계, 계획, 운영에 있어서 핵심적인 수문변수의 미래거동의 보다 나은 추정치가 필요하다. 예를 들어, 수력발전, 레크리에이션 이용과 하류지역의 오염희석과 같은 다중 목적을 유지하기 위하여 다목적댐을 운영할 때에, 다가오는 미래시간에 대한 계획된 유입량의 예측이 요구된다. 예측의 목적은 미래에 발생한 정확한 예측을 제공하는 것이다. 따라서 월유입량 예측을 위해 추계학적 모형(ARMA(1,1), ARMAX, TFN, SARIMA)과 신경망 모형(BP, CASCADE 등)의 적용을 통해 한강수게 주요 다목적댐에 가장 적합한 방법을 선정하고자 하는데 본 연구의 목적이 있다.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.9 no.8
• /
• pp.2948-2963
• /
• 2015

In the field of computer vision, visual surveillance systems have recently become an important research topic. Growth in this area is being driven by both the increase in the availability of inexpensive computing devices and image sensors as well as the general inefficiency of manual surveillance and monitoring. In particular, the ultimate goal for many visual surveillance systems is to provide automatic activity recognition for events at a given site. A higher level of understanding of these activities requires certain lower-level computer vision tasks to be performed. So in this paper, we propose an intelligent activity recognition model that uses a structure learning method and a classification method. The structure learning method is provided as a K2-learning algorithm that generates Bayesian networks of causal relationships between sensors for a given activity. The statistical characteristics of the sensor values and the topological characteristics of the generated graphs are learned for each activity, and then a neural network is designed to classify the current activity according to the features extracted from the multiple sensor values that have been collected. Finally, the proposed method is implemented and tested by using PETS2013 benchmark data.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.11 no.2
• /
• pp.785-804
• /
• 2017

Due to various factors such as postures, facial expressions and illuminations, face recognition by videos often suffer from poor recognition accuracy and generalization ability, since the within-class scatter might even be higher than the between-class one. Herein we address this problem by proposing a hierarchical cascaded classifier for video face recognition, which is a multi-layer algorithm and accounts for the misclassified samples plus their similar samples. Specifically, it can be decomposed into single classifier construction and multi-layer classifier design stages. In single classifier construction stage, classifier is created by clustering and the number of classes is computed by analyzing distance tree. In multi-layer classifier design stage, the next layer is created for the misclassified samples and similar ones, then cascaded to a hierarchical classifier. The experiments on the database collected by ourselves show that the recognition accuracy of the proposed classifier outperforms the compared recognition algorithms, such as neural network and sparse representation.