• Proceedings of the Korean Society of Broadcast Engineers Conference
• /
• fall
• /
• pp.290-293
• /
• 2021

In this paper, we implement a learning efficiency verification system to inspire learning motivation and help improve concentration by detecting the situation of the user studying. To this aim, data on learning attitude and concentration are measured by extracting the movement of the user's face or body through a real-time camera. The Jetson board was used to implement the real-time embedded system, and a convolutional neural network (CNN) was implemented for image recognition. After detecting the feature part of the object using a CNN, motion detection is performed. The captured image is shown in a GUI written in PYQT5, and data is collected by sending push messages when each of the actions is obstructed. In addition, each function can be executed on the main screen made with the GUI, and functions such as a statistical graph that calculates the collected data, To do list, and white noise are performed. Through learning efficiency checking system, various functions including data collection and analysis of targets were provided to users.

• The Bulletin of The Korean Astronomical Society
• /
• v.46 no.1
• /
• pp.42.1-42.1
• /
• 2021

In this study, we present a visual explanation of a deep learning solar flare forecast model and its relationship to physical parameters of solar active regions (ARs). For this, we use full-disk magnetograms at 00:00 UT from the Solar and Heliospheric Observatory/Michelson Doppler Imager and the Solar Dynamics Observatory/Helioseismic and Magnetic Imager, physical parameters from the Space-weather HMI Active Region Patch (SHARP), and Geostationary Operational Environmental Satellite X-ray flare data. Our deep learning flare forecast model based on the Convolutional Neural Network (CNN) predicts "Yes" or "No" for the daily occurrence of C-, M-, and X-class flares. We interpret the model using two CNN attribution methods (guided backpropagation and Gradient-weighted Class Activation Mapping [Grad-CAM]) that provide quantitative information on explaining the model. We find that our deep learning flare forecasting model is intimately related to AR physical properties that have also been distinguished in previous studies as holding significant predictive ability. Major results of this study are as follows. First, we successfully apply our deep learning models to the forecast of daily solar flare occurrence with TSS = 0.65, without any preprocessing to extract features from data. Second, using the attribution methods, we find that the polarity inversion line is an important feature for the deep learning flare forecasting model. Third, the ARs with high Grad-CAM values produce more flares than those with low Grad-CAM values. Fourth, nine SHARP parameters such as total unsigned vertical current, total unsigned current helicity, total unsigned flux, and total photospheric magnetic free energy density are well correlated with Grad-CAM values.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.17 no.5
• /
• pp.1396-1412
• /
• 2023

Conversation modeling is an important and challenging task in the field of natural language processing because it is a key component promoting the development of automated humanmachine conversation. Most recent research concerning conversation modeling focuses only on the current utterance (considered as the current question) to generate a response, and thus fails to capture the conversation's logic from its beginning. Some studies concatenate the current question with previous conversation sentences and use it as input for response generation. Another approach is to use an encoder to store all previous utterances. Each time a new question is encountered, the encoder is updated and used to generate the response. Our approach in this paper differs from previous studies in that we explicitly separate the encoding of the question from the encoding of its context. This results in different encoding models for the question and the context, capturing the specificity of each. In this way, we have access to the entire context when generating the response. To this end, we propose a deep neural network-based model, called the Context Model, to encode previous utterances' information and combine it with the current question. This approach satisfies the need for context information while keeping the different roles of the current question and its context separate while generating a response. We investigate two approaches for representing the context: Long short-term memory and Convolutional neural network. Experiments show that our Context Model outperforms a baseline model on both ConvAI2 Dataset and a collected dataset of conversational English.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2021.05a
• /
• pp.374-376
• /
• 2021

For the diagnosis of cancer-related diseases in clinical practice, pathological examination using biopsy is essential after basic diagnosis using imaging equipment. In order to proceed with such a biopsy, the assistance of an oncologist, clinical pathologist, etc. with specialized knowledge and the minimum required time are essential for confirmation. In recent years, research related to the establishment of a system capable of automatic classification of cancer cells using artificial intelligence is being actively conducted. However, previous studies show limitations in the type and accuracy of cells based on a limited algorithm. In this study, we propose a method to identify a total of 4 cancer cells through a convolutional neural network, a kind of deep learning. The optical images obtained through cell culture were learned through EfficientNet after performing pre-processing such as identification of the location of cells and image segmentation using OpenCV. The model used various hyper parameters based on EfficientNet, and trained InceptionV3 to compare and analyze the performance. As a result, cells were classified with a high accuracy of 96.8%, and this analysis method is expected to be helpful in confirming cancer.

• Proceedings of the Korea Water Resources Association Conference
• /
• 2021.06a
• /
• pp.200-200
• /
• 2021

하천은 용수공급, 관개, 친수활동, 산업활동 등 인간의 활동에 중요한 역할을 한다. 이에 따라 수질관리는 필수적이며 유기물, 중금속, 화학물질 등의 용존물질들은 수질문제에 직접적으로 영향을 미친다. 따라서 하천에서의 용존물질의 혼합 거동을 파악하기 위한 연구가 지난 수십년간 이루어지고 있다. 하천 흐름에 따른 오염물질의 이동 및 확산 거동을 예측하기 위하여 1차원 추적모형이 활용되는데, 그 중 하천저장대 모형(Transient Storage Model, TSM)은 자연하천의 복잡하고 불규칙한 수리·지형적인 특성을 단순하게 반영할 수 있다는 장점때문에 가장 많이 사용된다. 하지만 TSM은 매개변수에 대한 의존성과 불확도가 크며, TSM의 저장대에서의 농도분포에 대한 지수함수형태의 모델링이 하상간극수역(Hyporheic zone)에서의 저장대 특성을 반영하기에 구조적으로 부정확하다는 단점이 제기되고 있다. 최근 이러한 TSM의 단점을 보완하고 하천에서의 저장대 메커니즘을 보다 정확하게 구현하고자 체류시간분포(residence time distribution)를 이용한 확률론적 저장대 모델링 프레임워크가 등장하고 있다. 본 연구에서는 본류대와 저장대에서의 오염물질의 체류시간분포를 분리하여 해석하고 이를 전달함수(transfer function)를 이용한 합성곱으로 결합한 형태의 프레임워크를 적용하여 모델링하였다. 상기의 모형을 검증하기 위하여 2019년 감천의 4.85km 구간에서 추적자 실험을 실시하였다. 실험 당시 유량은 12.9 m³/s로 풍수기에 해당되며 평균 유속은 약 0.6 m/s로 측정되었다. 모형의 매개변수는 추적자 실험으로부터 최적화 기법을 통해 역모델링기법으로 결정하였다. 제안된 모형에 의한 모의 결과를 추적자 실험에서의 농도측정자료와 비교한 결과, 평균 0.988의 결정계수를 보여 매우 높은 정확도를 보이고 있음을 알 수 있었다. 저장대특성을 나타내는 농도곡선의 꼬리부에 대하여 같은 조건에서 1차원 이송-분산(ADE) 모형, TSM의 모의결과와도 비교한 결과 본 모형은 추적자 실험 농도측정 결과와 평균 0.195의 오차율을 보이며, 이는 ADE 모형과 TSM의 오차율인 14.03과 1.866에 비해 매우 정확한 것으로 나타났다.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2021.05a
• /
• pp.607-610
• /
• 2021

When learning a model in supervised learning, input data and the label of the data are required. However, labeling is high cost task and if automated, there is no guarantee that the label will always be correct. In the case of supervised learning in such a noisy labels environment, the accuracy of the model increases at the initial stage of learning, but decrease significantly after a certain period of time. There are various methods to solve the noisy label problem. But in most cases, the probability predicted by the model is used as the pseudo label. So, we proposed a method to predict the true label more quickly by refining the probabilities predicted by the model. Result of experiments on the same environment and dataset, it was confirmed that the performance improved and converged faster. Through this, it can be applied to methods that use the probability distribution predicted by the model among existing studies. And it is possible to reduce the time required for learning because it can converge faster in the same environment.

• Proceedings of the Korea Water Resources Association Conference
• /
• 2021.06a
• /
• pp.118-118
• /
• 2021

최근 지구 온난화 현상으로 인한 기후변화로 이상기후 현상이 발생하고 있으며 이로 인해 장기적으로 폭염의 빈도 및 강도 상승에 따른 가뭄 피해 우려가 증가하고 있다. 농업 가뭄은 강수량 부족, 토양 수분 부족, 저수량 부족 등 농업분야에 영향을 주는 인자들과 관련되어 있어 농작물 생육 및 수확량 감소를 야기한다. 우리나라는 논농사가 주를 이루고 있어 국내 농업 가뭄은 주수원공인 농업용 저수지의 가용저수용량으로 판단 가능하다. 따라서 안정적인 농업용수 공급을 위해 수리시설물의 모니터링, 공급량 등의 분석이 이루어져야 하며, 농업 가뭄에 대비하기 위해 농업용 저수지의 가용저수용량 파악이 필요하다. 수자원 분야에서 지점자료의 시·공간적 한계점을 보완하기 위해 인공위성 자료를 활용한 연구가 활발히 이루어지고 있으며, 본 연구에서는 위성영상 자료 및 딥러닝 기반 알고리즘을 적용하여 농업용 저수지 수표면 탐지 및 시계열 분석을 목적으로 한다. 위성영상 자료는 5일 주기 및 10 m 공간해상도를 가진 Sentinel-2 위성영상 자료를 활용하고자 하였으며, 딥러닝에 적용하기 위하여 100장 이상의 영상 이미지를 구축하였다. 딥러닝 기반 알고리즘으로는 Convolutional Neural Network (CNN)을 활용하였으며, CNN은 주로 이미지 분류나 객체 검출 문제를 해결하기 위해 제안된 모델로 최근 픽셀 단위로 분류가 가능한 알고리즘이 개발되어 높은 정확도의 수표면 탐지가 가능할 것으로 판단된다. 따라서 본 연구에서는 CNN 기반 수표면 탐지 알고리즘을 개발하여 Sentinel-2 영상 기준 경기도 안성시를 대상으로 소규모 농업용 저수지의 수표면적에 대한 시계열 데이터를 분석하고자 한다.

• Proceedings of the Korea Water Resources Association Conference
• /
• 2021.06a
• /
• pp.131-131
• /
• 2021

Deep learning models, especially those based on long short-term memory (LSTM), have presented their superiority in addressing time series data issues recently. This study aims to comprehensively evaluate the performance of deep learning models that belong to the supervised learning category in streamflow prediction. Therefore, six deep learning models-standard LSTM, standard gated recurrent unit (GRU), stacked LSTM, bidirectional LSTM (BiLSTM), feed-forward neural network (FFNN), and convolutional neural network (CNN) models-were of interest in this study. The Red River system, one of the largest river basins in Vietnam, was adopted as a case study. In addition, deep learning models were designed to forecast flowrate for one- and two-day ahead at Son Tay hydrological station on the Red River using a series of observed flowrate data at seven hydrological stations on three major river branches of the Red River system-Thao River, Da River, and Lo River-as the input data for training, validation, and testing. The comparison results have indicated that the four LSTM-based models exhibit significantly better performance and maintain stability than the FFNN and CNN models. Moreover, LSTM-based models may reach impressive predictions even in the presence of upstream reservoirs and dams. In the case of the stacked LSTM and BiLSTM models, the complexity of these models is not accompanied by performance improvement because their respective performance is not higher than the two standard models (LSTM and GRU). As a result, we realized that in the context of hydrological forecasting problems, simple architectural models such as LSTM and GRU (with one hidden layer) are sufficient to produce highly reliable forecasts while minimizing computation time because of the sequential data nature.

• Annual Conference of KIPS
• /
• 2021.11a
• /
• pp.681-684
• /
• 2021

In this paper, we study image clustering without labeling using machine learning techniques. We proposed an unsupervised machine learning technique to design an image clustering model that automatically categorizes images into groups. Our experiment focused on inception convolutional neural networks (inception V3) with k-mean methods to cluster images. For this, we collect the public datasets containing Food-K5, Flowers, Handwritten Digit, Cats-dogs, and our dataset Rice Germination, and the owner dataset Palm print. Our experiment can expand into three-part; First, format all the images to un-label and move to whole datasets. Second, load dataset into the inception V3 extraction image features and transferred to the k-mean cluster group hold on six classes. Lastly, evaluate modeling accuracy using the confusion matrix base on precision, recall, F1 to analyze. In this our methods, we can get the results as 1) Handwritten Digit (precision = 1.000, recall = 1.000, F1 = 1.00), 2) Food-K5 (precision = 0.975, recall = 0.945, F1 = 0.96), 3) Palm print (precision = 1.000, recall = 0.999, F1 = 1.00), 4) Cats-dogs (precision = 0.997, recall = 0.475, F1 = 0.64), 5) Flowers (precision = 0.610, recall = 0.982, F1 = 0.75), and our dataset 6) Rice Germination (precision = 0.997, recall = 0.943, F1 = 0.97). Our experiment showed that modeling could get an accuracy rate of 0.8908; the outcomes state that the proposed model is strongest enough to differentiate the different images and classify them into clusters.

• Tuberculosis and Respiratory Diseases
• /
• v.86 no.3
• /
• pp.226-233
• /
• 2023

Background: Inactive or old, healed tuberculosis (TB) on chest radiograph (CR) is often found in high TB incidence countries, and to avoid unnecessary evaluation and medication, differentiation from active TB is important. This study develops a deep learning (DL) model to estimate activity in a single chest radiographic analysis. Methods: A total of 3,824 active TB CRs from 511 individuals and 2,277 inactive TB CRs from 558 individuals were retrospectively collected. A pretrained convolutional neural network was fine-tuned to classify active and inactive TB. The model was pretrained with 8,964 pneumonia and 8,525 normal cases from the National Institute of Health (NIH) dataset. During the pretraining phase, the DL model learns the following tasks: pneumonia vs. normal, pneumonia vs. active TB, and active TB vs. normal. The performance of the DL model was validated using three external datasets. Receiver operating characteristic analyses were performed to evaluate the diagnostic performance to determine active TB by DL model and radiologists. Sensitivities and specificities for determining active TB were evaluated for both the DL model and radiologists. Results: The performance of the DL model showed area under the curve (AUC) values of 0.980 in internal validation, and 0.815 and 0.887 in external validation. The AUC values for the DL model, thoracic radiologist, and general radiologist, evaluated using one of the external validation datasets, were 0.815, 0.871, and 0.811, respectively. Conclusion: This DL-based algorithm showed potential as an effective diagnostic tool to identify TB activity, and could be useful for the follow-up of patients with inactive TB in high TB burden countries.