Hyun Yoo
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Previous Organizations / Institutes
The organizations / institutes from the scholar's other articles in the past ten years
Research Subjects
Hyun Yoo has a total of 10 co-authors. Here are the top co-authors by number of shared publications:
- Kyungyong Chung
Division of AI Computer Science and Engineering, Kyonggi University, Suwon-si 16227, Gyeonggi-do, Republic of Korea
12 shared publications - Seo-El Lee
Department of Public Safety Bigdata, Kyonggi University, Suwon-si, South Korea
3 shared publications - Soyoung Han
Department of Nursing, Yonsei University Wonju College of Medicine, Wonju, South Korea
3 shared publications - Geon Park1 shared publication
- Jaeyoon Ahn1 shared publication
- Jeonghyeon Chang
Contents Convergence Software Research Institute, Kyonggi University, Suwon, South Korea
1 shared publication
Latest Publications
- 31 August 2026
- Published by Korean Institute of Information Technology in The Journal of Korean Institute of Information Technology
- Vol. 24 (8) , 171-182
Tourist-Site Crowd Bottleneck Detection using Temporal Verification and VLM Reporting
- 31 August 2026
- Published by Korean Institute of Information Technology in The Journal of Korean Institute of Information Technology
- Vol. 24 (8) , 171-182
Tourist-Site Crowd Bottleneck Detection using Temporal Verification and VLM Reporting
- 29 July 2026
- Published by Elsevier in Information Processing & Management
- Vol. 64 (1)
Delta-Attention-Block-based time series forecasting model for automatic risk zone detection
- 29 July 2026
- Published by Elsevier in Information Processing & Management
- Vol. 64 (1)
Delta-Attention-Block-based time series forecasting model for automatic risk zone detection
- Open Access
- 21 August 2024
- Published by Springer Nature in Applied Intelligence
- Vol. 54 (22) , 12036
Correction to: Pose pattern mining using transformer for motion classification
- 21 August 2024
- Published by Springer Nature in Applied Intelligence
- Vol. 54 (22) , 12036
Open AccessCorrection to: Pose pattern mining using transformer for motion classification
- Open Access
- 12 August 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Access
- Vol. 12, 114626-114635
Feedback Based Evolutionary Spiral Learning Method for Reducing Class Ambiguity
In recent years, the deep learning based image classification has emerged as a crucial research topic in the fields of computer vision and artificial intelligence. However, the ambiguity inherent in image classification tasks causes the reduced accuracy in the classification process due to similarities between classes. This study proposes a feedback based evolutionary spiral learning method for reducing class ambiguity in the deep learning based image classification process. The proposed method consists of four stages: data collection, key image clustering, classification model training, and evaluation of classification results. If the evaluation results converge to specific measured values, the corresponding key image clusters are used as training data; otherwise, the process iterates through the previous stages in a spiral structure. Various experiments showed that the proposed method, compared to the traditional approach of manually labeling and generating training data, highly improved performance with an average of 82.38%. This reveals that the method can become a significant solution to address the issue of ambiguity that arises in deep learning based image classification tasks. In addition, it is expected to provide a faster and more optimized process in the fields of multi-class classification and detection.
- 12 August 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Access
- Vol. 12, 114626-114635
Open AccessFeedback Based Evolutionary Spiral Learning Method for Reducing Class Ambiguity
In recent years, the deep learning based image classification has emerged as a crucial research topic in the fields of computer vision and artificial intelligence. However, the ambiguity inherent in image classification tasks causes the reduced accuracy in the classification process due to similarities between classes. This study proposes a feedback based evolutionary spiral learning method for reducing class ambiguity in the deep learning based image classification process. The proposed method consists of four stages: data collection, key image clustering, classification model training, and evaluation of classification results. If the evaluation results converge to specific measured values, the corresponding key image clusters are used as training data; otherwise, the process iterates through the previous stages in a spiral structure. Various experiments showed that the proposed method, compared to the traditional approach of manually labeling and generating training data, highly improved performance with an average of 82.38%. This reveals that the method can become a significant solution to address the issue of ambiguity that arises in deep learning based image classification tasks. In addition, it is expected to provide a faster and more optimized process in the fields of multi-class classification and detection.
- 12 March 2024
- Published by Springer Nature in Applied Intelligence
- Vol. 54 (5) , 3841-3858
Pose pattern mining using transformer for motion classification
- 12 March 2024
- Published by Springer Nature in Applied Intelligence
- Vol. 54 (5) , 3841-3858
Pose pattern mining using transformer for motion classification
- 31 August 2023
- Published by Korean Institute of Information Technology in The Journal of Korean Institute of Information Technology
- Vol. 21 (8) , 11-17
Semantic Segmentation based Data Augmentation Techniques for Anomaly Detection of Concrete Structures
인공지능 및 딥러닝 기술의 발전으로 다양한 분야에서 혁신적인 성과가 이루어지고 있지만, 여전히 제한된 수량과 종류의 데이터셋을 이용하여 모델을 학습시키는 한계점이 존재한다. 본 논문에서는 콘크리트 구조물의 이상 탐지를 위한 시맨틱 분할 기반의 데이터 증강 기법을 제안한다. 제안하는 방법은 시맨틱 분할을 통해 결함 부분을 추출하고, 무작위 위치 선정과 알파 합성을 통해 객체와 배경을 합성하여 다양한 배경과 결함 조합을 포함하는 이미지 데이터셋을 생성한다. 이 방법은 Dice Score와 IoU Score를 사용하여 평가하였으며, 실험결과 이 방법은 기존의 방법보다 더 높은 성능을 보이고 있어 콘크리트 구조물 결함 탐지 분야에서 실용적으로 적용될 수 있다.
- 31 August 2023
- Published by Korean Institute of Information Technology in The Journal of Korean Institute of Information Technology
- Vol. 21 (8) , 11-17
Semantic Segmentation based Data Augmentation Techniques for Anomaly Detection of Concrete Structures
인공지능 및 딥러닝 기술의 발전으로 다양한 분야에서 혁신적인 성과가 이루어지고 있지만, 여전히 제한된 수량과 종류의 데이터셋을 이용하여 모델을 학습시키는 한계점이 존재한다. 본 논문에서는 콘크리트 구조물의 이상 탐지를 위한 시맨틱 분할 기반의 데이터 증강 기법을 제안한다. 제안하는 방법은 시맨틱 분할을 통해 결함 부분을 추출하고, 무작위 위치 선정과 알파 합성을 통해 객체와 배경을 합성하여 다양한 배경과 결함 조합을 포함하는 이미지 데이터셋을 생성한다. 이 방법은 Dice Score와 IoU Score를 사용하여 평가하였으며, 실험결과 이 방법은 기존의 방법보다 더 높은 성능을 보이고 있어 콘크리트 구조물 결함 탐지 분야에서 실용적으로 적용될 수 있다.
- 31 March 2021
- Published by Korean Society for Internet Information (KSII) in KSII Transactions on Internet and Information Systems
- Vol. 15 (3) , 974-992
IoT-Based Health Big-Data Process Technologies: A Survey
KSII Transactions on Internet and Information Systems TIIS Paper Details
- 31 March 2021
- Published by Korean Society for Internet Information (KSII) in KSII Transactions on Internet and Information Systems
- Vol. 15 (3) , 974-992
IoT-Based Health Big-Data Process Technologies: A Survey
KSII Transactions on Internet and Information Systems TIIS Paper Details
- Open Access
- 24 March 2021
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Access
- Vol. 9, 55802-55813
Diagnosis Support Model of Cardiomegaly Based on CNN Using ResNet and Explainable Feature Map
Medical expert support systems using medical data significantly contribute to the field of medicine by utilizing intelligent image analysis technology. Such medical systems must be validated and must have a transparent internal structure. Therefore, the technique of analyzing the results of the medical system is an important factor. This study proposes a diagnosis support model of cardiomegaly based on CNN using ResNet and explainable feature map. To configure the model, initially, a cardiomegaly diagnosis model is configured using ResNet, and a chest X-ray data set is used and learned. As the result changes resulting from the input changes of the model configured for model diagnosis support are detected, an explainable feature map for analysis is implemented. The input changes of the model are made as each pixel of the required image is reversed in sequence and the changes of the neural network output layer are saved on the feature map. As a result, the configured analysis method provides information inside the neural network through an accuracy result close to 80% and a visually expressed feature map. In particular, this interpretation method can be extended to a more general form in a medical support system that analyzes patterns.
- 24 March 2021
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Access
- Vol. 9, 55802-55813
Open AccessDiagnosis Support Model of Cardiomegaly Based on CNN Using ResNet and Explainable Feature Map
Medical expert support systems using medical data significantly contribute to the field of medicine by utilizing intelligent image analysis technology. Such medical systems must be validated and must have a transparent internal structure. Therefore, the technique of analyzing the results of the medical system is an important factor. This study proposes a diagnosis support model of cardiomegaly based on CNN using ResNet and explainable feature map. To configure the model, initially, a cardiomegaly diagnosis model is configured using ResNet, and a chest X-ray data set is used and learned. As the result changes resulting from the input changes of the model configured for model diagnosis support are detected, an explainable feature map for analysis is implemented. The input changes of the model are made as each pixel of the required image is reversed in sequence and the changes of the neural network output layer are saved on the feature map. As a result, the configured analysis method provides information inside the neural network through an accuracy result close to 80% and a visually expressed feature map. In particular, this interpretation method can be extended to a more general form in a medical support system that analyzes patterns.
- 3 August 2020
- Published by Springer Nature in Multimedia Tools and Applications
- Vol. 80 (26-27) , 34713-34729
Prediction of cardiac disease-causing pattern using multimedia extraction in health ontology
- 3 August 2020
- Published by Springer Nature in Multimedia Tools and Applications
- Vol. 80 (26-27) , 34713-34729
Prediction of cardiac disease-causing pattern using multimedia extraction in health ontology
- Open Access
- 26 July 2020
- Published by MDPI AG in Healthcare
- Vol. 8 (3) , 234
A Frequency Pattern Mining Model Based on Deep Neural Network for Real-Time Classification of Heart Conditions
Recently, a massive amount of big data of bioinformation is collected by sensor-based IoT devices. The collected data are also classified into different types of health big data in various techniques. A personalized analysis technique is a basis for judging the risk factors of personal cardiovascular disorders in real-time. The objective of this paper is to provide the model for the personalized heart condition classification in combination with the fast and effective preprocessing technique and deep neural network in order to process the real-time accumulated biosensor input data. The model can be useful to learn input data and develop an approximation function, and it can help users recognize risk situations. For the analysis of the pulse frequency, a fast Fourier transform is applied in preprocessing work. With the use of the frequency-by-frequency ratio data of the extracted power spectrum, data reduction is performed. To analyze the meanings of preprocessed data, a neural network algorithm is applied. In particular, a deep neural network is used to analyze and evaluate linear data. A deep neural network can make multiple layers and can establish an operation model of nodes with the use of gradient descent. The completed model was trained by classifying the ECG signals collected in advance into normal, control, and noise groups. Thereafter, the ECG signal input in real time through the trained deep neural network system was classified into normal, control, and noise. To evaluate the performance of the proposed model, this study utilized a ratio of data operation cost reduction and F-measure. As a result, with the use of fast Fourier transform and cumulative frequency percentage, the size of ECG reduced to 1:32. According to the analysis on the F-measure of the deep neural network, the model had 83.83% accuracy. Given the results, the modified deep neural network technique can reduce the size of big data in terms of computing work, and it is an effective system to reduce operation time.
- 26 July 2020
- Published by MDPI AG in Healthcare
- Vol. 8 (3) , 234
Open AccessA Frequency Pattern Mining Model Based on Deep Neural Network for Real-Time Classification of Heart Conditions
Recently, a massive amount of big data of bioinformation is collected by sensor-based IoT devices. The collected data are also classified into different types of health big data in various techniques. A personalized analysis technique is a basis for judging the risk factors of personal cardiovascular disorders in real-time. The objective of this paper is to provide the model for the personalized heart condition classification in combination with the fast and effective preprocessing technique and deep neural network in order to process the real-time accumulated biosensor input data. The model can be useful to learn input data and develop an approximation function, and it can help users recognize risk situations. For the analysis of the pulse frequency, a fast Fourier transform is applied in preprocessing work. With the use of the frequency-by-frequency ratio data of the extracted power spectrum, data reduction is performed. To analyze the meanings of preprocessed data, a neural network algorithm is applied. In particular, a deep neural network is used to analyze and evaluate linear data. A deep neural network can make multiple layers and can establish an operation model of nodes with the use of gradient descent. The completed model was trained by classifying the ECG signals collected in advance into normal, control, and noise groups. Thereafter, the ECG signal input in real time through the trained deep neural network system was classified into normal, control, and noise. To evaluate the performance of the proposed model, this study utilized a ratio of data operation cost reduction and F-measure. As a result, with the use of fast Fourier transform and cumulative frequency percentage, the size of ECG reduced to 1:32. According to the analysis on the F-measure of the deep neural network, the model had 83.83% accuracy. Given the results, the modified deep neural network technique can reduce the size of big data in terms of computing work, and it is an effective system to reduce operation time.