이슈논문

Data Report>이슈논문

이슈 키워드 논문 TOP10

이슈논문
| | | | 다운로드

1CMIP5 기반 하천유량 예측을 위한 딥러닝 LSTM 모형의 최적 학습기간 산정

저자 : 천범석 ( Chun Beom-Seok ) , 이태화 ( Lee Tae-hwa ) , 김상우 ( Kim Sang-woo ) , 임경재 ( Lim Kyoung-jae ) , 정영훈 ( Jung Young-hun ) , 도종원 ( Do Jong-won ) , 신용철 ( Shin Yong-chul )

발행기관 : 한국농공학회 간행물 : 한국농공학회논문집 64권 1호 발행 연도 : 2022 페이지 : pp. 39-50 (12 pages)

다운로드

(기관인증 필요)

초록보기

In this study, we suggested the optimal training period for predicting the streamflow using the LSTM (Long Short-Term Memory) model based on the deep learning and CMIP5 (The fifth phase of the Couple Model Intercomparison Project) future climate scenarios. To validate the model performance of LSTM, the Jinan-gun (Seongsan-ri) site was selected in this study. We comfirmed that the LSTM-based streamflow was highly comparable to the measurements during the calibration (2000 to 2002/2014 to 2015) and validation (2003 to 2005/2016 to 2017) periods. Additionally, we compared the LSTM-based streamflow to the SWAT-based output during the calibration (2000∼2015) and validation (2016∼2019) periods. The results supported that the LSTM model also performed well in simulating streamflow during the long-term period, although small uncertainties exist. Then the SWAT-based daily streamflow was forecasted using the CMIP5 climate scenario forcing data in 2011∼2100. We tested and determined the optimal training period for the LSTM model by comparing the LSTM-/SWAT-based streamflow with various scenarios. Note that the SWAT-based streamflow values were assumed as the observation because of no measurements in future (2011∼2100). Our results showed that the LSTM-based streamflow was similar to the SWAT-based streamflow when the training data over the 30 years were used. These findings indicated that training periods more than 30 years were required to obtain LSTM-based reliable streamflow forecasts using climate change scenarios.

2딥러닝 기반 모션인식 기술을 적용한 에너지 소모량 추정 알고리즘 개발

저자 : 이용국 ( Yong-kook Lee ) , 박재현 ( Jae-hyeon Park )

발행기관 : 한국체육학회 간행물 : 한국체육학회지 61권 2호 발행 연도 : 2022 페이지 : pp. 79-95 (17 pages)

다운로드

(기관인증 필요)

초록보기

이 연구는 딥러닝기반 모션인식기술(Deep Learning Motion Recognize : DLMR)에 적용할 성인들의 걷기 속도별 에너지 소비량을 산출하여, DLMR 기술에 적용할 성인들의 성별과 연령대별 에너지 소비량 추정공식(회귀식)을 산출하여 적용하는 데 목적이 있다. 첫째, 성별과 연령대별 에너지 소비량 추정공식을 산출을 위해 20-60대로 구성된 총 300명을 연구대상으로 선정하였으며, 간접연량추정 운동부하검사 측정기계를 이용하여 가스 교환율을 근거로 에너지소비량(kcal)을 산출하였다. 둘째, 웨어러블과 DLMR의 신체활동량(거리, 속도) 타당도 및 에너지소비량 추정치 비교를 위해서 20-60대로 구성된 총30명의 연구대상을 선정하였으며, DLMR이 부착된 룸에서 직사각형(6m)바닥을 웨어러블(미밴드, S사 기어, S사 헬스)기기를 착용하여 측정하였다. 모든 자료처리는 SPSS Ver 21.0과 MS-Excel을 사용하였으며, 교차타당도를 위해서 모형생성과 검증데이터 비율은 7:3을 적용하였다. 모든 통계적 유의수준은 0.05로 설정하였다. 각 연령대(20대-60대), 성별(남성, 여성)에 따른 에너지 소비량 다중회귀분석 결과 에너지소비량 총 변화량을 걷기속도의 점수로 설명할 수 있었다. 또한 DLMR에서 산출된 에너지 소비량 추정치와 웨어러블로 산출된 에너지소비량의 비교시 DLMR의 타당도가 타 웨어러블 검사도구보다 높게 산출되었으며, DLMR로 측정한 신체활동 거리와 실제 신체활동 거리는 보통의 정적상관과 높은 급내상관이 나타났다(p<0.05). 성별, 연령대, 걷기 속도별에 따른 추정된 에너지 소비량의 신뢰성 및 타당성은 검증이 되었다.


The purpose of this study was to develop a walking energy expenditure estimation equation accroding to sex and age in adults, using walking speeds from 0.1 km/h to 6.4km/h, to be applied to Deep Learning Motion Recognition (DLMR). First, a total of 300participants (150 men) between 20-60 years of age were selected in Seoul, South Korea. Energy expenditure (kcal) was calculated based on gas exchange ration by using indirect calorimetry (graded exercise test; GXT, Quark b2, Cosmed, Italy). Next, 30 participants(15 men) between 20-60 years of age were selected for determining the valdity of physical activity amount assessment (distance, pace) using a wearble device (MIband, S brand gear, S brand healthm Seoul, South Korea) and DLMR. A comparison of energy expenditure measurement was mad between the two divices while walking along a 6-meter rectangle of the floor. All data were processed using SPSS ver 21.0 (IBM corp., Armonk, MY, USA) and Excel in Microsoft® Office 2016(Microsoft Corporation, Redmond, WA, USA). The ration of model markup and validation data for cross valdity was set to 7:3 The level of statistical significance was set at 0.05 DLMR mmeasurement of energy according to sex, age, and walking pace was shown to be val id and reilable, indicating that the application of DLMR for this purpose is feasible, Furthemore, the energy expenditure regression equation after adding of body mass index and age as independent varialbes contributed to maxmize reliability.

KCI등재

3딥러닝을 이용한 연안 소형 어선 주요 치수 추정 연구

저자 : 장민성 ( Min Sung Jang ) , 김동준 ( Dong-joon Kim ) , 자오양 ( Yang Zhao )

발행기관 : 한국수산해양기술학회 간행물 : 수산해양기술연구(구 한국어업기술학회지) 58권 3호 발행 연도 : 2022 페이지 : pp. 272-280 (9 pages)

다운로드

(기관인증 필요)

초록보기

The first step is to determine the principal dimensions of the design ship, such as length between perpendiculars, beam, draft and depth when accomplishing the design of a new vessel. To make this process easier, a database with a large amount of existing ship data and a regression analysis technique are needed. Recently, deep learning, a branch of artificial intelligence (AI) has been used in regression analysis. In this paper, deep learning neural networks are used for regression analysis to find the regression function between the input and output data. To find the neural network structure with the highest accuracy, the errors of neural network structures with varying the number of the layers and the nodes are compared. In this paper, Python TensorFlow Keras API and MATLAB Deep Learning Toolbox are used to build deep learning neural networks. Constructed DNN (deep neural networks) makes helpful in determining the principal dimension of the ship and saves much time in the ship design process.

KCI등재SCOUPUS

4딥러닝을 통한 하이엔드 패션 브랜드 감성 학습

저자 : 장세윤 ( Seyoon Jang ) , 김하연 ( Ha Youn Kim ) , 이유리 ( Yuri Lee ) , 설진석 ( Jinseok Seol ) , 김성재 ( Seongjae Kim ) , 이상구 ( Sang-goo Lee )

발행기관 : 한국의류학회 간행물 : 한국의류학회지 46권 1호 발행 연도 : 2022 페이지 : pp. 165-181 (17 pages)

다운로드

(기관인증 필요)

초록보기

The fashion industry is creating innovative business models using artificial intelligence. To efficiently utilize artificial intelligence (AI), fashion data must be classified. Until now, such data have been classified focusing only on the objective properties of fashion products. Their subjective attributes, such as fashion brand sensibilities, are holistic and heuristic intuitions created by a combination of design elements. This study aims to improve the performance of collaborative filtering in the fashion industry by extracting fashion brand sensibility using computer vision technology. The image data set of fashion brand sensibility consists of high-end fashion brand photos that share sensibilities and communicate well in fashion. About 26,000 fashion photos of 11 high-end fashion brand sensibility labels have been collected from the 16FW to 21SS runway and 50 years of US Vogue magazines beginning from 1971. We use EfficientNet-B1 to establish the main architecture and fine-tune the network with ImageNet-ILSVRC. After training fashion brand sensibilities through deep learning, the proposed model achieved an F-1 score of 74% on accuracy tests. Furthermore, as a result of comparing AI machine and human experts, the proposed model is expected to be expanded to mass fashion brands.

KCI등재

5딥러닝 언어모형의 평가와 언어학

저자 : 송상헌 ( Song Sanghoun )

발행기관 : 서강대학교 언어정보연구소 간행물 : 언어와 정보사회 45권 0호 발행 연도 : 2022 페이지 : pp. 169-191 (23 pages)

다운로드

(기관인증 필요)

초록보기

This article addresses how the deep learning-based language models can be evaluated with respect to linguistic knowledge. Building upon the overlook, this article discusses how linguistics can make a substantial contribution to the development of the artificial intelligence systems. As many transformer-based models have been competitively implemented for the last few years, it is required to evaluate the multiple models in a common and reliable way. For this purpose, a wide range of linguistic evaluation metrics have been designed and constructed. The evaluation datasets involve the concepts used in theoretical linguistics, such as syntax, semantics, and pragmatics. The evaluation process follows the guideline used in psycholinguistic experiments. As such, the linguistic knowledge enhances interpretability of the deep leaning-based natural language processing techniques. It is contended that linguistics will play a pivotal role in evaluating and improving the language models in further research.

KCI등재

6머신러닝과 딥러닝을 이용한 영산강의 Chlorophyll-a 예측 성능 비교 및 변화 요인 분석

저자 : 심선희 ( Sun-hee Shim ) , 김유흔 ( Yu-heun Kim ) , 이혜원 ( Hye Won Lee ) , 김민 ( Min Kim ) , 최정현 ( Jung Hyun Choi )

발행기관 : 한국물환경학회 간행물 : 한국물환경학회지 38권 6호 발행 연도 : 2022 페이지 : pp. 292-305 (14 pages)

다운로드

(기관인증 필요)

초록보기

The Yeongsan River, one of the four largest rivers in South Korea, has been facing difficulties with water quality management with respect to algal bloom. The algal bloom menace has become bigger, especially after the construction of two weirs in the mainstream of the Yeongsan River. Therefore, the prediction and factor analysis of Chlorophyll-a (Chl-a) concentration is needed for effective water quality management. In this study, Chl-a prediction model was developed, and the performance evaluated using machine and deep learning methods, such as Deep Neural Network (DNN), Random Forest (RF), and eXtreme Gradient Boosting (XGBoost). Moreover, the correlation analysis and the feature importance results were compared to identify the major factors affecting the concentration of Chl-a. All models showed high prediction performance with an R2 value of 0.9 or higher. In particular, XGBoost showed the highest prediction accuracy of 0.95 in the test data. The results of feature importance suggested that Ammonia (NH3-N) and Phosphate (PO4-P) were common major factors for the three models to manage Chl-a concentration. From the results, it was confirmed that three machine learning methods, DNN, RF, and XGBoost are powerful methods for predicting water quality parameters. Also, the comparison between feature importance and correlation analysis would present a more accurate assessment of the important major factors.

KCI등재

7광학 위성 영상 기반 선박탐지의 정확도 개선을 위한 딥러닝 초해상화 기술의 영향 분석

저자 : 박성욱 ( Seongwook Park ) , 김영호 ( Yeongho Kim ) , 김민식 ( Minsik Kim )

발행기관 : 대한원격탐사학회 간행물 : 대한원격탐사학회지 38권 5호 발행 연도 : 2022 페이지 : pp. 559-570 (12 pages)

다운로드

(기관인증 필요)

초록보기

광학 위성 영상의 공간해상도가 낮게 되면 크기가 작은 객체들의 경우 객체 탐지의 어려움이 따른다. 따라서 본 연구에서는 위성 영상의 공간해상도를 향상시키는 초해상화(Super-resolution) 기술이 객체 탐지 정확도 향상에 대한 영향이 유의미한지 알아보고자 하였다. 쌍을 이루지 않는(unpaired) 초해상화 알고리즘을 이용하여 Sentinel-2 영상의 공간해상도를 3.2 m로 향상시켰으며, 객체 탐지 모델인 Faster-RCNN, RetinaNet, FCOS, S2ANet을 활용하여 초해상화 적용 유무에 따른 선박 탐지 정확도 변화를 확인했다. 그 결과 선박 탐지 모델의 성능 평가에서 초해상화가 적용된 영상으로 학습된 선박 탐지 모델들에서 Average Precision (AP)가 최소 12.3%, 최대 33.3% 향상됨을 확인하였고, 초해상화가 적용되지 않은 모델에 비해 미탐지 및 과탐지가 줄어듦을 보였다. 이는 초해상화 기술이 객체 탐지에서 중요한 전처리 단계가 될 수 있다는 것을 의미하고, 객체 탐지와 더불어 영상 기반의 다른 딥러닝 기술의 정확도 향상에도 크게 기여할 수 있을 것으로 기대된다.


When a satellite image has low spatial resolution, it is difficult to detect small objects. In this research, we aim to check the effect of super resolution on object detection. Super resolution is a software method that increases the resolution of an image. Unpaired super resolution network is used to improve Sentinel-2’s spatial resolution from 10 m to 3.2 m. Faster-RCNN, RetinaNet, FCOS, and S2ANet were used to detect vessels in the Sentinel-2 images. We experimented the change in vessel detection performance when super resolution is applied. As a result, the Average Precision (AP) improved by at least 12.3% and up to 33.3% in the ship detection models trained with the super-resolution image. False positive and false negative cases also decreased. This implies that super resolution can be an important pre-processing step in object detection, and it is expected to greatly contribute to improving the accuracy of other image-based deep learning technologies along with object detection.

KCI등재

8딥러닝 기반 객체 인식을 통한 철계 열처리 부품의 인지에 관한 연구

저자 : 박혜정 ( Hye-jung Park ) , 황창하 ( Chang-ha Hwang ) , 김상권 ( Sang-gwon Kim ) , 여국현 ( Kuk-hyun Yeo ) , 서상우 ( Sang-woo Seo )

발행기관 : 한국열처리공학회 간행물 : 열처리공학회지 35권 6호 발행 연도 : 2022 페이지 : pp. 327-336 (10 pages)

다운로드

(기관인증 필요)

초록보기

In this study, a model for automatically recognizing several steel parts through a camera before charging materials was developed under the assumption that the temperature distribution in the pre-air atmosphere was known. For model development, datasets were collected in random environments and factories. In this study, the YOLO-v5 model, which is a YOLO model with strengths in real-time detection in the field of object detection, was used, and the disadvantages of taking a lot of time to collect images and learning models was solved through the transfer learning methods. The performance evaluation results of the derived model showed excellent performance of 0.927 based on mAP 0.5. The derived model will be applied to the model development study, which uses the model to accurately recognize the material and then match it with the temperature distribution in the atmosphere to determine whether the material layout is suitable before charging materials. (Received November 8, 2022; Revised November 11, 2022; Accepted November 21, 2022)

KCI등재

9금융 실현변동성을 위한 내재변동성과 인터넷 검색량을 활용한 딥러닝

저자 : 신지원 ( Jiwon Shin ) , 신동완 ( Dong Wan Shin )

발행기관 : 한국통계학회 간행물 : 응용통계연구 35권 1호 발행 연도 : 2022 페이지 : pp. 93-104 (12 pages)

다운로드

(기관인증 필요)

초록보기

S&P 500과 RUSSELL 2000, DJIA, Nasdaq 100 4가지 미국 주가지수의 실현변동성(realized volatility, RV)을 예측하는데 있어서 사람들의 관심 지표로 삼을 수 있는 인터넷 검색량(search volume, SV) 지수와 내재변동성(implied volatility, IV)를 이용하여 LSTM 딥러닝(deep learning) 방법으로 RV의 예측력을 높이고자하였다. SV을 이용한 LSTM 방법의 실현변동성 예측력이 기존의 기본적인 vector autoregressive (VAR) 모형, vector error correction (VEC)보다 우수하였다. 또한, 최근 제안된 RV와 IV의 공적분 관계를 이용한 vector error correction heterogeneous autoregressive (VECHAR) 모형보다도 전반적으로 예측력이 더 높음을 확인하였다.


In forecasting realized volatility of the major US stock price indexes (S&P 500, Russell 2000, DJIA, Nasdaq 100), internet search volume reflecting investor’s interests and implied volatility are used to improve forecast via a deep learning method of the LSTM. The LSTM method combined with search volume index produces better forecasts than existing standard methods of the vector autoregressive (VAR) and the vector error correction (VEC) models. It also beats the recently proposed vector error correction heterogeneous autoregressive (VECHAR) model which takes advantage of the cointegration relation between realized volatility and implied volatility.

KCI등재

10딥러닝 기반 광섬유 분포 음향·진동 계측기술을 활용한 장거리 외곽 침입감지 시스템 개발

저자 : 김희운 ( Huioon Kim ) , 이주영 ( Joo-young Lee ) , 정효영 ( Hyoyoung Jung ) , 김영호 ( Young Ho Kim ) , 권준혁 ( Jun Hyuk Kwon ) , 기송도 ( Song Do Ki ) , 김명진 ( Myoung Jin Kim )

발행기관 : 한국센서학회 간행물 : 센서학회지 31권 1호 발행 연도 : 2022 페이지 : pp. 24-30 (7 pages)

다운로드

(기관인증 필요)

초록보기

Distributed fiber-optic acoustic·vibration sensing technology is becoming increasingly popular in many industrial and academic areas such as in securing large edifices, exploring underground seismic activity, monitoring oil well/reservoir, etc. Long-range perimeter intrusion detection exemplifies an application that not only detects intrusion, but also pinpoints where it happens and recognizes kinds of threats made along the perimeter where a single fiber cable was installed. In this study, we developed a distributed fiber-optic sensing device that measures a distributed acoustic·vibration signature (pattern) for intrusion detection. In addition, we demontrate the proposed deep learning algorithm and how it classifies various intrusion events. We evaluated the sensing device and deep learning algorithm in a practical testbed setup. The evaluation results confirm that the developed system is a promising intrusion detection system for long-distance and seamless recognition requirements.

내가 찾은 최근 검색어

최근 열람 자료

맞춤 논문

보관함

내 보관함
공유한 보관함

1:1문의

닫기