제주대학교와 창원대학교의 지체장애인 편의시설 비교분석

고상선,박철민,김태일 제주대학교 공과대학 첨단기술연구소 2002 尖端技術硏究所論文集 Vol.13 No.2

Colleges and Universities are one of our whole life educational facilities for the disabled regarding job training opportunities are academic goal achievement. However it is usual for the disabled students to be deprived of their opportunities to study and to prepare their future life on campus because of physical obstacles of campus environment. Therefore this study provides what are the physical obstacles on campus, how to use the college's facilities for the disabled to achieve their academic goals and to have job training opportunities just like ordinary students, and how to improve the campus environment which is based on the concept of Universal Design.

다채널 오디오 특징값 및 게이트형 순환 신경망을 사용한 다성 사운드 이벤트 검출

고상선,조혜승,김형국,Ko, Sang-Sun,Cho, Hye-Seung,Kim, Hyoung-Gook 한국음향학회 2017 韓國音響學會誌 Vol.36 No.4

본 논문에서는 다채널 오디오 특징값을 게이트형 순환 신경망(Gated Recurrent Neural Networks, GRNN)에 적용한 효과적인 다성 사운드 이벤트 검출 방식을 제안한다. 실생활의 사운드는 여러 사운드 이벤트가 겹쳐있는 다성사운드로, 기존의 단일 채널 오디오 특징값으로는 다성 사운드에서 개별적인 이벤트의 검출이 어렵다는 한계가 있다. 이에 본 논문에서는 다채널 오디오 신호를 기반으로 추출된 특징값을 사용하여 다성 사운드 이벤트 검출에 적용하였다. 또한 본 논문에서는 현재 순환 신경망에서 가장 높은 성능을 보이는 장단기 기억 신경망(Long Short Term Memory, LSTM) 보다 간단한 GRNN을 분류에 적용하여 다성 사운드 이벤트 검출의 성능을 더욱 향상시키고자 하였다. 실험결과는 본 논문에서 제안한 방식이 기존의 방식보다 성능이 더 뛰어나다는 것을 보인다. In this paper, we propose an effective method of applying multichannel-audio feature values to GRNNs (Gated Recurrent Neural Networks) in polyphonic sound event detection. Real life sounds are often overlapped with each other, so that it is difficult to distinguish them by using a mono-channel audio features. In the proposed method, we tried to improve the performance of polyphonic sound event detection by using multi-channel audio features. In addition, we also tried to improve the performance of polyphonic sound event detection by applying a gated recurrent neural network which is simpler than LSTM (Long Short Term Memory), which shows the highest performance among the current recurrent neural networks. The experimental results show that the proposed method achieves better sound event detection performance than other existing methods.

교통사고 발생원인 인식과 감소대책 인지 영향요인 판별?분류에 관한 연구

고상선,배기목,이원규,정헌영 대한교통학회 2002 대한교통학회지 Vol.20 No.7

교통사고 발생지점의 유형화와 원인인지.감소대책 선호모델 구축에 관한 연구

고상선,오석기 대한교통학회 1995 大韓交通學會誌 Vol.13 No.1

Traffic has a very important function but has caused such social problems as traffic congestion parking and traffic accidents in metropolitan areas. It is difficult to examine the causes of traffic accidents related to human life, which occur by human, vehicle and environmental factors. But human factor is the only measure requlating these factors together an analyzing factors influencing establishment of counterplan of traffic accidents. Consequently , this study employs the principal component analysis and stepwise multiple regression analysis to estimate the characteristics and influential factors of traffic accidents and defines the typical patterns of happening lots of traffic accidents. Accordingly, this study establishes an acknowledgement model of the causes and preference model of the counterplan of traffic accidents using Multi-Dimension Preference(MDPREF) method.

대형교통사고 판별모델 구축에 관한 연구

고상선,이원규,배기목,노유진 한국항해항만학회 1999 韓國港灣學會誌 Vol.13 No.1

Traffic accidents increase with the increase of the vehicles in operation on the street. Especially big traffic accidents composed of over 3 killed or 20 injured accidents with the property damage become one of the serious problems to be solved in most of the cities. The purpose of this study is to build the discrimination model on big traffic accidents using the Quantification II theory for establishing the countermeasures to reduce the big traffic accidents. The results are summarized as follows. 1)The existing traffic accident related model could not explain the phenomena of the current traffic accident appropriately. 2) Based on the big traffic accident types vehicle-vehicle, vehicle-alone, vehicle-pedestrian and vehicle-train accident rates 73%, 20.5% 5.6% and two cases respectively. Based on the law violation types safety driving non-fulfillment center line invasion excess speed and signal disobedience were 48.8%, 38.1% 2.8% and 2.8% respectively. 3) Based on the law violation types major factors in big traffic accidents were road and environment, human, and vehicle in order. Those factors were vehicle, road and environment, and human in order based on types of injured driver’s death. 4) Based on the law violation types total hitting and correlation rates of the model were 53.57% and 0.97853. Based on the types of injured driver’s death total hitting and correlation rates of the model were also 71.4% and 0.59583.

교통사고 영상기록장치(DVD: Driving Video Recorder) 자료를 활용한 다중충돌사고 해석기법의 개발

고상선,정호교 대한교통학회 2012 대한교통학회 학술대회지 Vol.67 No.-

고배속 CD-ROM용 비대칭형 광픽업 미세구동기의 구동특성

고상선,류제하,박기환,정호섭 한국소음진동공학회 1998 소음 진동 Vol.8 No.2

This paper presents actuating characteristics of an asymmetric high-speed optical pick-up fine actuator that can be installed in a small area such as a notebook personal computer. In the asymmetric actuator four points (mass center, actuation center, supporting point of wire suspension on a bobbin, and optical axis) are not coincident so that the proposed actuator suspension reveals undesirable suspension resonance in the pitch and yaw direction. Lumped parameter dynamic model in each direction is used to investigate the driving characteristics with respect to relative location of the four points. Some of desired design directions toward reducing resonance peaks are suggested by using sensitivity information. In order to avoid undesirable resonance, at least supporting point on the obbin must be located in the middle of the mass and actuation center of the asymmetric pick-up actuator.

주목 메커니즘 기반의 심층신경망을 이용한 음성 감정인식

고상선,조혜승,김형국,Ko, Sang-Sun,Cho, Hye-Seung,Kim, Hyoung-Gook 한국음향학회 2017 韓國音響學會誌 Vol.36 No.6

본 논문에서는 주목 메커니즘 기반의 심층 신경망을 사용한 음성 감정인식 방법을 제안한다. 제안하는 방식은 CNN(Convolution Neural Networks), GRU(Gated Recurrent Unit), DNN(Deep Neural Networks)의 결합으로 이루어진 심층 신경망 구조와 주목 메커니즘으로 구성된다. 음성의 스펙트로그램에는 감정에 따른 특징적인 패턴이 포함되어 있으므로 제안하는 방식에서는 일반적인 CNN에서 컨벌루션 필터를 tuned Gabor 필터로 사용하는 GCNN(Gabor CNN)을 사용하여 패턴을 효과적으로 모델링한다. 또한 CNN과 FC(Fully-Connected)레이어 기반의 주목 메커니즘을 적용하여 추출된 특징의 맥락 정보를 고려한 주목 가중치를 구해 감정인식에 사용한다. 본 논문에서 제안하는 방식의 검증을 위해 6가지 감정에 대해 인식 실험을 진행하였다. 실험 결과, 제안한 방식이 음성 감정인식에서 기존의 방식보다 더 높은 성능을 보였다. In this paper, we propose a speech emotion recognition method using a deep neural network based on the attention mechanism. The proposed method consists of a combination of CNN (Convolution Neural Networks), GRU (Gated Recurrent Unit), DNN (Deep Neural Networks) and attention mechanism. The spectrogram of the speech signal contains characteristic patterns according to the emotion. Therefore, we modeled characteristic patterns according to the emotion by applying the tuned Gabor filters as convolutional filter of typical CNN. In addition, we applied the attention mechanism with CNN and FC (Fully-Connected) layer to obtain the attention weight by considering context information of extracted features and used it for emotion recognition. To verify the proposed method, we conducted emotion recognition experiments on six emotions. The experimental results show that the proposed method achieves higher performance in speech emotion recognition than the conventional methods.

ALSCAL法에 의한 交通事故 原因認知 및 減少對策 選好모델 構築에 관한 硏究

오석기,고상선 東亞大學校 大學院 1996 大學院論文集 Vol.21 No.-

Such social changes as the change of industrial structure, the expansion of economic scale, the elevationof national life level and rapid motorizaiton bring about social problems that are traffic accident, traffic accident is related with hyman life. Accordingly, this study is essentially for the establishment of acknowledgement model of the cause and preference model of the decrease counterplan of traffic accident using ALSCAL.

主成分分析法에 의한 우리나라 都市의 類型化에 관한 硏究

尹時雲,高祥善 東亞大學校 海洋資源硏究所 1988 硏究論文集 Vol.1 No.1

Korea has been in the situation of re-organizing the metropolitan governing system and unbalanced development of its territory. Considering the studies of urban planning so far, relatively no studies have been done on locations of cities and towns and dependence connection among them. Accordingly, the purpose of this study is defining remarkable tendency of cities and the typical patterns of urban structures as well, throughout the principal component analysis based on industry. The following are the results of the study in brief. 1. Classified into 4 patterns of urban. 1) Highly-accumulated-city 2) Advanced-city 3) Industrial-city 4) Underdeveloped-city 2. Urbanized and industrialized cities based on old central cities of region have developed into highly-accumulated cities and advanced cities and also newly industrialized cities have become industrial cities. 3. Conditions of development:Re-organization of transportation.