• 한국인터넷방송통신학회논문지
• /
• 제21권5호
• /
• pp.203-208
• /
• 2021

제조현장에서 작업자는 작업 지시서에 따라 제조 공정에 소재를 투입하고 투입 기록을 남기는 방식으로 운영해왔으나, 누락하는 경우가 많아 제품 LOT 추적이 안되는 경우가 발생하고 있었으며, 최근 스마트공장 구축으로 RFID-Tag를 활용하여 소재 투입 정보를 자동입력 하는 시스템으로 진행되고 있다. 특히, 생산라인에 투입되는 RACK에 부착된 TAG 정보를 수신하여 RACK(TAG) ID와 RACK 투입시간 데이터 분석을 통한 투입정보를 자동으로 생성토록 하여 초기 자동인식률이 97%로 양호하였으나 멀티소재 사용 RACK, TAG분실, 신규 제품 투입 이슈 등이 발생하면서 자동인식률이 계속 낮아지는 상황이다. 인공지능형 스마트공장 데이터셋 구축 방법은 자동인식률 향상과 실시간 모니터링이 가능해지므로 생산 공정의 전반에 있어 속도와 수율(정상제품 비율)을 높이는데 기여할 것으로 기대한다.

• 방송공학회논문지
• /
• 제26권4호
• /
• pp.419-428
• /
• 2021

그림자 제거는 객체 추적 및 검출 등 영상처리 기술의 핵심 전처리 요소이다. 최근 심층 합성곱 신경망 (Deep Convolutional Neural Network) 기반의 영상 인식 기술이 발전함에 따라 심층 학습을 이용한 그림자 제거 연구들이 활발히 진행되고 있다. 본 논문에서는 자기 주의 증류(Self Attention Distillation)를 이용하여 심층 특징을 추출하는 새로운 그림자 제거 방법을 제안한다. 제안된 방법은 각 층에서 추출된 그림자 검출 결과를 하향식 증류를 통해 점진적으로 정제한다. 특히, 그림자 검출 결과에 대한 정답을 이용하지 않고 그림자 제거를 위한 문맥적 정보를 형성함으로써 효율적인 심층 신경망 학습을 수행한다. 그림자 제거를 위한 다양한 데이터 셋에 대한 실험 결과를 통해 제안하는 방법이 실제 환경에서 발생한 그림자 제거에 효과적임을 보인다.

• 한국컴퓨터정보학회논문지
• /
• 제26권9호
• /
• pp.37-47
• /
• 2021

본 논문에서는 U-네트에서 수용 영역을 다양화하여 기존의 모델보다 정확도가 개선된 새로운 혀 영역 분할을 위한 딥러닝 모델을 제안한다. 수용 영역 다양화를 위하여 병렬 컨볼루션, 팽창된 컨볼루션, 상수 채널 증가 등의 방법을 사용하였다. 제안된 딥러닝 모델에 대하여, 학습 영상과 테스트 영상이 유사한 TestSet1과 그렇지 않은 TestSet2의 두 가지 테스트 데이터에 대해 혀 영역검출 실험을 진행하였다. 수용 영역이 다양화됨에 따라 혀 영역 분할 성능이 향상되는 것을 실험결과에서 확인할 수 있었다. 제안한 방법의 mIoU 값은 TestSet1의 경우 98.14%, TestSet2의 경우 91.90%로 U-net, DeepTongue, TongueNet 등 기존 모델의 결과보다 높았다.

• 대한원격탐사학회지
• /
• 제38권1호
• /
• pp.73-82
• /
• 2022

Marine fisheries resources face major anthropogenic threat from unregulated fishing activities; thus require precise detection for protection through marine surveillance. Korea developed an efficient land-based small fishing vessel monitoring system using real-time V-Pass data. However, those data directly do not provide information on fishing activities, thus further efforts are necessary to differentiate their activity status. In Korea, especially in Busan, longlining is practiced by many small fishing vessels to catch several types of fishes that need to be identified for proper monitoring. Therefore, in this study we have improved the existing fishing status classification method by applying Hidden Markov Model (HMM) on V-Pass data in order to further classify their fishing status into three groups, viz. non-fishing, longlining and other types of fishing. Data from 206 fishing vessels at Busan on 05 February, 2021 were used for this purpose. Two tiered HMM was applied that first differentiates non-fishing status from the fishing status, and finally classifies that fishing status into longlining and other types of fishing. Data from 193 and 13 ships were used as training and test datasets, respectively. Using this model 90.45% accuracy in classifying into fishing and non-fishing status and 88.23% overall accuracy in classifying all into three types of fishing statuses were achieved. Thus, this method is recommended for monitoring the activities of small fishing vessels equipped with V-Pass, especially for detecting longlining.

• 스마트미디어저널
• /
• 제11권4호
• /
• pp.30-37
• /
• 2022

Disease threatens plant growth and recognizing the type of disease is essential to making a remedy. In recent years, deep learning has witnessed a significant improvement for this task, however, a large volume of labeled images is one of the requirements to get decent performance. But annotated images are difficult and expensive to obtain in the agricultural field. Therefore, designing an efficient and effective strategy is one of the challenges in this area with few labeled data. Transfer learning, assuming taking knowledge from a source domain to a target domain, is borrowed to address this issue and observed comparable results. However, current transfer learning strategies can be regarded as a supervised method as it hypothesizes that there are many labeled images in a source domain. In contrast, unsupervised transfer learning, using only images in a source domain, gives more convenience as collecting images is much easier than annotating. In this paper, we leverage unsupervised transfer learning to perform plant disease recognition, by which we achieve a better performance than supervised transfer learning in many cases. Besides, a vision transformer with a bigger model capacity than convolution is utilized to have a better-pretrained feature space. With the vision transformer-based unsupervised transfer learning, we achieve better results than current works in two datasets. Especially, we obtain 97.3% accuracy with only 30 training images for each class in the Plant Village dataset. We hope that our work can encourage the community to pay attention to vision transformer-based unsupervised transfer learning in the agricultural field when with few labeled images.

• Nuclear Engineering and Technology
• /
• 제54권2호
• /
• pp.608-616
• /
• 2022

Power peaking factors (PPF) is an important parameter for safe and efficient reactor operation. There are several methods to calculate the PPF at TRIGA research reactors such as MCNP and TRIGLAV codes. However, these methods are time-consuming and required high specifications of a computer system. To overcome these limitations, artificial intelligence was introduced for parameter prediction. Previous studies applied the neural network method to predict the PPF, but the publications using the ANFIS method are not well developed yet. In this paper, the prediction of PPF using the ANFIS was conducted. Two input variables, control rod position, and neutron flux were collected while the PPF was calculated using TRIGLAV code as the data output. These input-output datasets were used for ANFIS model generation, training, and testing. In this study, four ANFIS model with two types of input space partitioning methods shows good predictive performances with R² values in the range of 96%-97%, reveals the strong relationship between the predicted and actual PPF values. The RMSE calculated also near zero. From this statistical analysis, it is proven that the ANFIS could predict the PPF accurately and can be used as an alternative method to develop a real-time monitoring system at TRIGA research reactors.

• 대한원격탐사학회지
• /
• 제38권1호
• /
• pp.111-126
• /
• 2022

Land use and land cover (LULC) mapping is an important factor in geospatial analysis. Although highly precise ground-based LULC monitoring is possible, it is time consuming and costly. Conversely, because the synthetic aperture radar (SAR) sensor is an all-weather sensor with high resolution, it could replace field-based LULC monitoring systems with low cost and less time requirement. Thus, LULC is one of the major areas in SAR applications. We developed a LULC model using only KOMPSAT-5 single co-polarized data and digital elevation model (DEM) data. Twelve HH-polarized images and 18 VV-polarized images were collected, and two HH-polarized images and four VV-polarized images were selected for the model testing. To train the LULC model, we applied the conditional generative adversarial network (cGAN) method. We used U-Net combined with the residual unit (ResUNet) model to generate the cGAN method. When analyzing the training history at 1732 epochs, the ResUNet model showed a maximum overall accuracy (OA) of 93.89 and a Kappa coefficient of 0.91. The model exhibited high performance in the test datasets with an OA greater than 90. The model accurately distinguished water body areas and showed lower accuracy in wetlands than in the other LULC types. The effect of the DEM on the accuracy of LULC was analyzed. When assessing the accuracy with respect to the incidence angle, owing to the radar shadow caused by the side-looking system of the SAR sensor, the OA tended to decrease as the incidence angle increased. This study is the first to use only KOMPSAT-5 single co-polarized data and deep learning methods to demonstrate the possibility of high-performance LULC monitoring. This study contributes to Earth surface monitoring and the development of deep learning approaches using the KOMPSAT-5 data.

• Composite Materials and Engineering
• /
• 제3권3호
• /
• pp.201-220
• /
• 2021

In this study, potential of three machine learning techniques i.e., M5P, Support vector machines and Gaussian processes were evaluated to find the best algorithm for the prediction of flexural strength of concrete mix with steel fibre. The study comprises the comparison of results obtained from above-said techniques for given dataset. The dataset consists of 124 observations from past research studies and this dataset is randomly divided into two subsets namely training and testing datasets with (70-30)% proportion by weight. Cement, fine aggregates, coarse aggregates, water, super plasticizer/ high-range water reducer, steel fibre, fibre length and curing days were taken as input parameters whereas flexural strength of the concrete mix was taken as the output parameter. Performance of the techniques was checked by statistic evaluation parameters. Results show that the Gaussian process technique works better than other techniques with its minimum error bandwidth. Statistical analysis shows that the Gaussian process predicts better results with higher coefficient of correlation value (0.9138) and minimum mean absolute error (1.2954) and Root mean square error value (1.9672). Sensitivity analysis proves that steel fibre is the significant parameter among other parameters to predict the flexural strength of concrete mix. According to the shape of the fibre, the mixed type performs better for this data than the hooked shape of the steel fibre, which has a higher CC of 0.9649, which shows that the shape of fibers do effect the flexural strength of the concrete. However, the intricacy of the mixed fibres needs further investigations. For future mixes, the most favorable range for the increase in flexural strength of concrete mix found to be (1-3)%.

• 대한안전경영과학회지
• /
• 제24권4호
• /
• pp.1-8
• /
• 2022

Elevators are the main means of transport in buildings. A malfunction of an elevator in operation may cause in convenience to users. Furthermore, fatal accidents, such as injuries and death, may occur to the passengers also. Therefore, it is important to prevent failure before accidents happen. In related studies, preventive measures are proposed through analyzing failures, and the lifespan of elevator components. However, these methods are limited to existing an elevator model and its surroundings, including operating conditions and installed environments. Vibration occurs when the elevator is operated. Experts have classified types of faults, which are symptoms for malfunctions (failures), via analyzing vibration. This study proposes an artificial intelligent model for classifying faults automatically with deep learning algorithms through elevator vibration data, hereby preventing failures before they occur. In this study, the vibration data of six elevators are collected. The proposed methodology in this paper removes "the measurement error data" with incorrect measurements and extracts operating sections from the input datasets for proceeding deep learning models. As a result of comparing the performance of training five deep learning models, the maximum performance indicates Accuracy 97% and F1 Score 97%, respectively. This paper presents an artificial intelligent model for detecting elevator fault automatically. The users' safety and convenience may increase by detecting fault prior to the fatal malfunctions. In addition, it is possible to reduce manpower and time by assisting experts who have previously classified faults.

• 정보보호학회논문지
• /
• 제32권5호
• /
• pp.817-826
• /
• 2022

AI 기술이 여러 분야에 성공적으로 도입되는 추세이며, 서비스로 환경에 배포된 모델들은 지적 재산권과 데이터를 보호하기 위해 모델의 정보를 노출시키지 않는 블랙 박스 상태로 배포된다. 블랙 박스 환경에서 공격자들은 모델 출력을 이용해 학습에 쓰인 데이터나 파라미터를 훔치려고 한다. 본 논문은 딥러닝 모델을 대상으로 모델 종류에 대한 정보를 추론하는 공격이 없다는 점에서 착안하여, 모델의 구성 레이어 정보를 직접 알아내기 위해 모델의 종류를 추론하는 공격 방법을 제안한다. MNIST 데이터셋으로 학습된 ResNet, VGGNet, AlexNet과 간단한 컨볼루션 신경망 모델까지 네 가지 모델의 그레이 박스 및 블랙 박스 환경에서의 출력값을 이용해 모델의 종류가 추론될 수 있다는 것을 보였다. 또한 본 논문이 제안하는 방식인 대소 관계 피쳐를 딥러닝 모델에 함께 학습시킨 경우 블랙 박스 환경에서 약 83%의 정확도로 모델의 종류를 추론했으며, 그 결과를 통해 공격자에게 확률 벡터가 아닌 제한된 정보만 제공되는 상황에서도 모델 종류가 추론될 수 있음을 보였다.