• Annual Conference of KIPS
• /
• 2021.11a
• /
• pp.618-621
• /
• 2021

수박의 이미지와 수박의 무게 데이터를 활용해 수박의 당도를 예측하고 모델의 정확도를 측정한다. 과피가 얇고, 부피가 작은 과일의 경우 휴대용 비파괴 당도 측정기를 통해 비교적 간편하게 당도 측정이 가능하다. 하지만 수박은 과피도 두껍고, 부피도 크기 때문에 넓은 장소와 비용을 부담해야 하는 선별장에만 당도를 측정할 수 있는 실정이다. 본 논문에서는 줄무늬가 끊어지지 않고, 원형이 아닌 타원형이 맛있는 수박이라는 속설에 부합하는 수박이 실제로 맛있는 수박인지를 확인하고자 수박 이미지를 수집하여 당도에 따라 이미지를 분류한 다음, CNN 을 적용하여 수박 당도 예측을 실시하였다. 실험 결과 타원형 수박은 당도가 높은 것으로 나타났으나 줄무늬가 끊어진 수박과 끊어지지 않은 수박 간의 당도 차이는 없는 것으로 나타났다. 향후 수박의 당도에 영향을 미칠 수 있는 다양한 변수를 활용하여 정확도를 높인다면 현재 사용되고 있는 비파괴 당도 측정기를 보완할 수 있을 것으로 기대된다.

• Proceedings of the Korea Water Resources Association Conference
• /
• 2021.06a
• /
• pp.194-194
• /
• 2021

최근 지구온난화 및 이상기후 현상으로 인하여 집중호우의 빈도와 강도가 급증하고 있다. 그리고 급격한 도시화로 불투수 면적이 증가하여 도시지역에 침수피해가 빈번하게 발생하고 있는 실정이다. 이러한 침수피해를 방지하기 위하여 침수위험지구, 재해위험지구를 선정하여 집중호우에 대하여 집중관리를 하고 있지만 위험지구이외의 곳에서 침수가 발생할 경우 신속하게 대처하지 못하는 문제가 발생하고 있다. 또한, 하천이 범람하여 발생하는 외수침수의 경우 수위를 실시간으로 확인할 수 있어 미리 대응이 가능하지만, 내수침수의 경우 지하에 매설되어 있는 관로의 상태를 확인할 수 없기 때문에 순간적으로 발생하는 침수에 대하여 신속하게 대처를 해야 한다. 현재 침수 피해를 신속하게 대처하기 위하여 CCTV를 활용해 침수의 발생여부를 모니터링 하고 있지만 CCTV설치 지역에 비하여 적은 인력으로 모든 CCTV를 확인하지 못하여 침수피해를 신속하게 대처하지 못하고 있는 실정이다. 본 연구에서는 침수사진 자료를 CNN(Convolutional Neural Network)기법을 이용하여 학습시켜 침수의 발생여부를 판단하는 모델을 제안하였다. 딥러닝 기법의 CNN은 이미지의 특징을 추출하여 학습하는 과정을 가지게 되는데 학습이 완료된 모델은 침수사진의 특징을 파악하여 침수가 발생하였는지에 대한 여부를 자동적으로 판단하게 된다. 본 연구결과를 CCTV관재센터 혹은 지자체와의 연계를 통하여 침수의 발생여부를 자동적으로 판단해주는 시스템이 개발된다면 신속한 침수피해 대처가 이루어 질 수 있을 것이라 판단된다.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2021.10a
• /
• pp.348-350
• /
• 2021

When predicting the concentration of fine dust using deep learning, there is a problem that the characteristics of a high concentration of 81㎍/m³ or more are not well reflected in the prediction model. In this paper, a comparison through predictive performance was conducted to confirm the results of reflecting the characteristics of fine dust in the high concentration area according to the deep learning algorithm. As a result of performance evaluation, overall, similar levels of results were shown, but the RNN model showed higher accuracy than other models at concentrations of "very bad" based on AQI. This confirmed that the RNN algorithm reflected the characteristics of the high concentration better than the DNN and LSTM algorithms.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2021.10a
• /
• pp.521-523
• /
• 2021

Data augmentation are techniques used to increase the amount of data by using small amount of existing data. With the spread of the Internet, we can easily obtain data. However, there are still certain industries, like medicine, where it is difficult to obtain data. The same is true for image data in which a black screen is displayed on video wall controller. Because it is rare that a black screen is displayed during operation, it is not easy to obtain an image with a black screen. We propose a DCGAN based architecture that generate dataset using a small amount of black screen image.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2021.10a
• /
• pp.524-526
• /
• 2021

As the video wall controller market is growing rapidly, issues that have not been addressed so far are raised. One of them is a phenomenon in which a black screen is displayed on a multi-screen. Black screen is displayed due to an error in the video being displayed in the video wall controller. Human intervention is inevitable to recognize and solve the black screen. However, it is impossible for the operator to monitor the multi-screen 24 hours a day. In this paper, we propose a model that detects the black screen being displayed on the video wall controller. We propose a CNN based architecture to detect a black screen.

• Smart Structures and Systems
• /
• v.32 no.1
• /
• pp.61-81
• /
• 2023

This study presents an ensemble learning based Bayesian model updating approach for structural damage diagnosis. In the developed framework, the structure is initially decomposed into a set of substructures. The autoregressive moving average (ARMAX) model is established first for structural damage localization based structural motion equation. The wavelet packet decomposition is utilized to extract the damage-sensitive node energy in different frequency bands for constructing structural surrogate models. Four methods, including Kriging predictor (KRG), radial basis function neural network (RBFNN), support vector regression (SVR), and multivariate adaptive regression splines (MARS), are selected as candidate structural surrogate models. These models are then resampled by bootstrapping and combined to obtain an ensemble model by probabilistic ensemble. Meanwhile, the maximum entropy principal is adopted to search for new design points for sample space updating, yielding a more robust ensemble model. Through the iterations, a framework of surrogate ensemble learning based model updating with high model construction efficiency and accuracy is proposed. The specificities of the method are discussed and investigated in a case study.

• Journal of Preventive Medicine and Public Health
• /
• v.56 no.4
• /
• pp.327-337
• /
• 2023

Objectives: Considering the importance of social determinants of health (SDHs) in promoting the health of residents of informal settlements and their diversity, abundance, and breadth, this study aimed to identify, measure, and rank SDHs for health promotion interventions targeting informal settlement residents in a metropolitan area in Iran. Methods: Using a hybrid method, this study was conducted in 3 phases from 2019 to 2020. SDHs were identified by reviewing studies and using the Delphi method. To examine the SDHs among informal settlement residents, a cross-sectional analysis was conducted using researcher-made questionnaires. Multilayer perceptron analysis using an artificial neural network was used to rank the SDHs by priority. Results: Of the 96 determinants identified in the first phase of the study, 43 were examined, and 15 were identified as high-priority SDHs for use in health-promotion interventions for informal settlement residents in the study area. They included individual health literacy, nutrition, occupational factors, housing-related factors, and access to public resources. Conclusions: Since identifying and addressing SDHs could improve health justice and mitigate the poor health status of settlement residents, ranking these determinants by priority using artificial intelligence will enable policymakers to improve the health of settlement residents through interventions targeting the most important SDHs.

• Smart Media Journal
• /
• v.12 no.1
• /
• pp.32-46
• /
• 2023

Natural Language Processing (NLP) has grown tremendously in recent years. Typically, bilingual, and multilingual translation models have been deployed widely in machine translation and gained vast attention from the research community. On the contrary, few studies have focused on translating between spoken and sign languages, especially non-English languages. Prior works on Sign Language Translation (SLT) have shown that a mid-level sign gloss representation enhances translation performance. Therefore, this study presents a new large-scale Korean sign language dataset, the Museum-Commentary Korean Sign Gloss (MCKSG) dataset, including 3828 pairs of Korean sentences and their corresponding sign glosses used in Museum-Commentary contexts. In addition, we propose a translation framework based on self-supervised learning, where the pretext task is a text-to-text from a Korean sentence to its back-translation versions, then the pre-trained network will be fine-tuned on the MCKSG dataset. Using self-supervised learning help to overcome the drawback of a shortage of sign language data. Through experimental results, our proposed model outperforms a baseline BERT model by 6.22%.

• Earthquakes and Structures
• /
• v.24 no.2
• /
• pp.127-140
• /
• 2023

The vibration of microtubule in human cells is the source of electrical field around it and inside cell structure. The induction of electrical field is a direct result of the existence of dipoles on the surface of the microtubules. Measuring the electrical fields could be performed using nano-scale sensors and the data could be transformed to other computers using internet of things (IoT) technology. Processing these data is feasible by artificial intelligence-based methods. However, the first step in analyzing the vibrational behavior is to study the mechanics of microtubules. In this regard, the vibrational behavior of the microtubules is investigated in the present study. A shell model is utilized to represent the microtubules' structure. The displacement field is assumed to obey first order shear deformation theory and classical theory of elasticity for anisotropic homogenous materials is utilized. The governing equations obtained by Hamilton's principle are further solved using analytical method engaging Navier's solution procedure. The results of the analytical solution are used to train, validate and test of the deep neural network. The results of the present study are validated by comparing to other results in the literature. The results indicate that several geometrical and material factors affect the vibrational behavior of microtubules.

• Structural Engineering and Mechanics
• /
• v.87 no.3
• /
• pp.231-242
• /
• 2023

A novel combinatorial type-2 adaptive neuro-fuzzy inference system (T2-ANFIS) and robust proportional integral derivative (PID) control framework for intelligent vibration mitigation of uncertain structural system is introduced. The fuzzy logic controllers (FLCs), are designed independently of the mathematical model of the system. The type-1 FLCs, have a limited ability to reduce the effect of uncertainty, due to their fuzzy sets with a crisp degree of membership. In real applications, the consequent part of the fuzzy rules is uncertain. The type-2 FLCs, are robust to the fuzzy rules and the process parameters due to the fuzzy degree of membership functions and footprint of uncertainty (FOU). The adaptivity of the proposed method is provided with the optimum tuning of the parameters using the neural network training algorithms. In our approach, the PID control force is obtained using the generalized type-2 neuro-fuzzy in such a way that the stability and robustness of the controller are guaranteed. The robust performance and stability of the presented framework are demonstrated in a numerical study for an eleven-story seismically-excited building structure combined with an active tuned mass damper (ATMD). The results indicate that the introduced type-2 neuro-fuzzy PID control scheme is effective to attenuate plant states in the presence of the structured and unstructured uncertainties, compared to the conventional, type-1 FLC, type-2 FLC, and type-1 neuro-fuzzy PID controllers.