• 대한원격탐사학회지
• /
• 제37권2호
• /
• pp.275-290
• /
• 2021

대기 중의 이산화황(SO2)은 주로 인위적 배출원에 의해 발생하며 화학 반응을 통해 (초)미세먼지를 형성하여 직간접적으로 주변 환경 및 인체 건강에 해로운 영향을 주는 물질이다. 특히 지상에서의 농도는 인간 활동과 밀접한 관련이 있어 모니터링의 필요성이 매우 크다. 따라서, 본 연구에서는 TROPOMI SO2 연직 컬럼 농도 산출물 및 타 위성 산물과 모델 산출물 등을 융합 활용하여 기계학습 기법에 적용하여 SO2 지상 농도 추정모델을 개발하였다. 기계학습 기법으로는 널리 활용되고 있는 RF(Random Forest)에 잔차 보정 과정을 결합한 2-step 잔차 보정 RF를 적용하였다. 개발된 모델은 무작위, 공간 및 시간별 10-fold 교차 검증을 통하여 검증하였으며, 기울기(slope) 값이 1.14-1.25, 상관계수(R) 값이 0.55-0.65, rRMSE 값이 약 58-63% 정도로 나타났다. 이는 잔차 보정이 적용되지 않은 기존의 RF 대비 slope의 경우 약 10%, R과 rRMSE의 경우 약 3% 가량 향상된 결과를 보인다. 국가별로 나누어 분석하였을 때에는 샘플 수가 적고 SO2의 전반적인 농도가 낮은 일본 지역에서의 공간별 10-fold 교차검증 성능이 소폭 감소하는 것으로 나타났다. SO2 지상농도 분포를 계절별로 표출하였을 때, 일본의 경우 다른 지역 대비 연중 저농도가 관찰되며 높은 결측 값 비율로 인하여 관측소 농도 대비 2-step 잔차 보정 RF 모델에서 과대 모의하는 경향이 관찰되었다. 대표적 고농도 발생지인 중국의 YRD(Yangtze River Delta) 와 한국의 SMA(Seoul Metropolitan Area)의 계절적 분포 변화를 추가적으로 분석하였을 때, 연료 연소로 인한 겨울철 농도 증가 패턴이 나타났다. 이는 인위적 배출원의 영향을 크게 받는 SO2의 시공간적인 분포 특성을 잘 반영하고 있는 결과이다. 따라서, 본 연구를 통하여 제안한 모델은 장기적으로 SO2 지상 농도의 시공간적 분포를 파악하는 데에 활용될 수 있을 것으로 기대된다.

• Journal of Electrical Engineering and Technology
• /
• 제11권1호
• /
• pp.29-37
• /
• 2016

Fast and accurate fault diagnosis of the position sensor is of great significance to ensure the reliability as well as sensor fault tolerant operation of the Switched Reluctance Wind Generator (SRWG). This paper presents a fault diagnostic scheme for a SRWG based on the residual between the estimated rotor position and the actual output of the position sensor. Extreme Learning Machine (ELM), which could build a nonlinear mapping among flux linkage, current and rotor position, is utilized to design an assembled estimator for the rotor position detection. The data for building the ELM based assembled position estimator is derived from the magnetization curves which are obtained from Finite Element Analysis (FEA) of an SRWG with the structure of 8 stator poles and 6 rotor poles. The effectiveness and accuracy of the proposed fault diagnosis method are verified by simulation at various operating conditions. The results provide a feasible theoretical and technical basis for the effective condition monitoring and predictive maintenance of SRWG.

• 대한전기학회논문지
• /
• 제40권8호
• /
• pp.782-791
• /
• 1991

In the fields of pattern recognition such as speech recognition, several new techniques using Artifical Neural network Models have been proposed and implemented. In particular, the Multilayer Perception Model has been shown to be effective in static speech pattern recognition. But speech has dynamic or temporal characteristics and the most important point in implementing speech recognition systems using Artificial Neural Network Models for continuous speech is the learning of dynamic characteristics and the distributed cues and contextual effects that result from temporal characteristics. But Recurrent Multilayer Perceptron Model is known to be able to learn sequence of pattern. In this paper, the results of applying the Recurrent Model which has possibilities of learning tedmporal characteristics of speech to phoneme recognition is presented. The test data consist of 144 Vowel+ Consonant + Vowel speech chains made up of 4 Korean monothongs and 9 Korean plosive consonants. The input parameters of Artificial Neural Network model used are the FFT coefficients, residual error and zero crossing rates. The Baseline model showed a recognition rate of 91% for volwels and 71% for plosive consonants of one male speaker. We obtained better recognition rates from various other experiments compared to the existing multilayer perceptron model, thus showed the recurrent model to be better suited to speech recognition. And the possibility of using Recurrent Models for speech recognition was experimented by changing the configuration of this baseline model.

• 인터넷정보학회논문지
• /
• 제20권6호
• /
• pp.29-36
• /
• 2019

When dealing with outdoor images in a variety of computer vision applications, the presence of shadow degrades performance. In order to understand the information occluded by shadow, it is essential to remove the shadow. To solve this problem, in many studies, involves a two-step process of shadow detection and removal. However, the field of shadow detection based on CNN has greatly improved, but the field of shadow removal has been difficult because it needs to be restored after removing the shadow. In this paper, it is assumed that shadow is detected, and shadow-less image is generated by using original image and shadow mask. In previous methods, based on CGAN, the image created by the generator was learned from only the aspect of the image patch in the adversarial learning through the discriminator. In the contrast, we propose a novel method using a discriminator that judges both the whole image and the local patch at the same time. We not only use the residual generator to produce high quality images, but we also use joint loss, which combines reconstruction loss and GAN loss for training stability. To evaluate our approach, we used an ISTD datasets consisting of a single image. The images generated by our approach show sharp and restored detailed information compared to previous methods.

• 대한영양사협회학술지
• /
• 제25권1호
• /
• pp.44-58
• /
• 2019

Predicting the number of meals in a foodservice organization is an important decision-making process that is essential for successful food production, such as reducing the amount of residue, preventing menu quality deterioration, and preventing rising costs. Compared to other demand forecasts, the menu of dietary personnel includes diverse menus, and various dietary supplements include a range of side dishes. In addition to the menus, diverse subjects for prediction are very difficult problems. Therefore, the purpose of this study was to establish a method for predicting the number of meals including predictive modeling and considering various factors in addition to menus which are actually used in the field. For this purpose, 63 variables in eight categories such as the daily available number of people for the meals, the number of people in the time series, daily menu details, weekdays or seasons, days before or after holidays, weather and temperature, holidays or year-end, and events were identified as decision variables. An ensemble model using six prediction models was then constructed to predict the number of meals. As a result, the prediction error rate was reduced from 10%~11% to approximately 6~7%, which was expected to reduce the residual amount by approximately 40%.

• Journal of Information Processing Systems
• /
• 제19권4호
• /
• pp.427-438
• /
• 2023

Plenty of works have indicated that single image super-resolution (SISR) models relying on synthetic datasets are difficult to be applied to real scene text image super-resolution (STISR) for its more complex degradation. The up-to-date dataset for realistic STISR is called TextZoom, while the current methods trained on this dataset have not considered the effect of multi-scale features of text images. In this paper, a multi-scale and attention fusion model for realistic STISR is proposed. The multi-scale learning mechanism is introduced to acquire sophisticated feature representations of text images; The spatial and channel attentions are introduced to capture the local information and inter-channel interaction information of text images; At last, this paper designs a multi-scale residual attention module by skillfully fusing multi-scale learning and attention mechanisms. The experiments on TextZoom demonstrate that the model proposed increases scene text recognition's (ASTER) average recognition accuracy by 1.2% compared to text super-resolution network.

• 한국지능시스템학회논문지
• /
• 제19권4호
• /
• pp.574-579
• /
• 2009

복잡한 비선형 회귀문제들에서 최적의 신경망을 구축하기 위해서는 구조의 선정 및 노이즈에 의한 과잉학습(overtraining)등에 따른 많은 문제들이 있다. 본 논문에서는 flexible incremental 알고리즘을 이용하여 단계적으로 최적의 신경망을 구축하는 방법을 제안한다. Flexible incremental 알고리즘은 예측 잔류오차를 최소화하기 위해 단계적으로 추가되어지는 은닉노드 개수를 검증데이터를 이용하여 신축성 있게 조절하고, 빠른 학습을 위하여 ELM (Extreme Learning Machine)을 이용한다. 제안된 방법은 신경망의 구축과정에서 사용자의 어떠한 관여 없이도 빠른 학습과 적은 수의 은닉노드들에 의한 범용 근사화 (universal approximation)가 가능한 신경망의 구축이 가능한 장점을 가지고 있다. 다양한 종류의 벤치마크 데이터들을 이용한 실험 결과를 통하여 제안된 방법이 실제 회귀문제들에서 우수한 성능을 가짐을 확인하였다.

• 한국컴퓨터정보학회논문지
• /
• 제25권5호
• /
• pp.11-18
• /
• 2020

본 연구는 컨베이어 시스템에서 딥러닝을 이용한 배출구 막힘 판단 기술에 대하여 제안한다. 제안 방법은 산업 현장의 CCTV에서 수집한 영상을 이용하여 배출구 막힘 판단을 위한 다양한 CNN 모델들을 학습시키고, 성능이 가장 좋은 모델을 사용하여 실제 공정에 적용하는 것을 목적으로 한다. CNN 모델로는 잘 알려진 VGGNet, ResNet, DenseNet, 그리고 NASNet을 사용하였으며, 모델 학습과 성능 테스트를 위하여 CCTV에서 수집한 18,000장의 영상을 이용하였다. 다양한 모델에 대한 실험 결과, VGGNet은 99.89%의 정확도와 29.05ms의 처리 시간으로 가장 좋은 성능을 보였으며, 이로부터 배출구 막힘 판단 문제에 VGGNet이 가장 적합함을 확인하였다.

• 전기전자학회논문지
• /
• 제24권4호
• /
• pp.1122-1128
• /
• 2020

대역 확장(Bandwidth Extension)이란 채널 용량 부족 혹은 이동통신 기기에 탑재된 코덱의 특성으로 인해 부호화 및 복호화 과정에서 대역 제한(band limited)되거나 손상된 협대역 신호(NB, Narrow Band)를 복원, 확장하여 광대역 신호(WB, Wide Band)로 전환 시켜주는 것을 의미한다. 대역 확장 연구는 주로 음성 신호 위주로 대역 복제(SBR, Spectral Band Replication), IGF(Intelligent Gap Filling)과 같이 고대역을 주파수 영역으로 변환하여 복잡한 특징 추출 과정을 거쳐 이를 바탕으로 사라지거나 손상된 고대역을 복원한다. 본 논문에서는 딥러닝 모델 중 오토인코더(Autoencoder)를 바탕으로 1차원 합성곱 신경망(CNN, Convolutional Neural Network)들의 잔차 연결을 활용하여 복잡한 사전 전처리 과정 없이 일정한 길이의 시간 영역 신호를 입력시켜 대역 확장 시킨 음향 신호를 출력하는 모델을 제안한다. 또한 음성 영역에 제한되지 않는 음악을 포함한 여러 종류의 음원을 포함하는 데이터셋에 훈련시켜도 손상된 고대역을 복원할 수 있음을 확인하였다.

• Wind and Structures
• /
• 제36권6호
• /
• pp.405-421
• /
• 2023

Strong wind is the main factors of wind-damage of high-rise buildings, which often creates largely economical losses and casualties. Wind pressure plays a critical role in wind effects on buildings. To obtain the high-resolution wind pressure field, it often requires massive pressure taps. In this study, two traditional methods, including bilinear and bicubic interpolation, and two deep learning techniques including Residual Networks (ResNet) and Generative Adversarial Networks (GANs), are employed to reconstruct wind pressure filed from limited pressure taps on the surface of an ideal building from TPU database. It was found that the GANs model exhibits the best performance in reconstructing the wind pressure field. Meanwhile, it was confirmed that k-means clustering based retained pressure taps as model input can significantly improve the reconstruction ability of GANs model. Finally, the generalization ability of k-means clustering based GANs model in reconstructing wind pressure field is verified by an actual engineering structure. Importantly, the k-means clustering based GANs model can achieve satisfactory reconstruction in wind pressure field under the inputs processing by k-means clustering, even the 20% of pressure taps. Therefore, it is expected to save a huge number of pressure taps under the field reconstruction and achieve timely and accurately reconstruction of wind pressure field under k-means clustering based GANs model.