• Proceedings of the Korea Information Processing Society Conference
• /
• 2022.05a
• /
• pp.528-530
• /
• 2022

심층강화학습에 완전 연결 신경망과 합성곱 신경망은 잘 활용되는 것에 반해 순환 신경망은 잘 활용되지 않는다. 이는 강화학습이 마르코프 속성을 전제로 하기 때문이다. 지금까지의 강화학습은 환경이 마르코프 속성을 만족하도록 사전 작업이 필요했다, 본 논문에서는 마르코프 속성을 따르지 않는 환경에서 이러한 사전 작업 없이도 순환 신경망의 은닉 상태를 통해 마르코프 속성을 학습함으로써 학습 성능을 개선할 수 있다는 것을 소개한다.

• International journal of advanced smart convergence
• /
• v.10 no.4
• /
• pp.278-288
• /
• 2021

Recently, IoT systems are cloud-based, so that continuous and large amounts of data collected from sensor nodes are processed in the data server through the cloud. However, in the centralized configuration of large-scale cloud computing, computational processing must be performed at a physical location where data collection and processing take place, and the need for edge computers to reduce the network load of the cloud system is gradually expanding. In this paper, a cluster system consisting of 6 inexpensive Raspberry Pi boards was constructed to perform fast data processing. And we propose "Kubernetes cluster system(KCS)" for processing large data collection and analysis by model distribution and data pipeline method. To compare the performance of this study, an ensemble model of deep learning was built, and the accuracy, processing performance, and processing time through the proposed KCS system and model distribution were compared and analyzed. As a result, the ensemble model was excellent in accuracy, but the KCS implemented as a data pipeline proved to be superior in processing speed..

• Journal of Korea Multimedia Society
• /
• v.25 no.3
• /
• pp.451-460
• /
• 2022

Recently, several deep learning studies have been reported to automatically identify the location of diagnostic devices using endoscopic data. In previous studies, there was no design to determine whether the configuration of the dataset resulted in differences in the accuracy in which artificial intelligence models perform image classification. Studies that are based on large amounts of data are likely to have different results depending on the composition of the dataset or its proportion. In this study, we intended to determine the existence and extent of accuracy according to the composition of the dataset by compiling it into three main types using larynx, esophagus, gastroscopy, and laryngeal endoscopy images.

• Nuclear Engineering and Technology
• /
• v.54 no.8
• /
• pp.3027-3033
• /
• 2022

Background: In this study, various types of deep-learning models for predicting in vitro radiosensitivity from gene-expression profiling were compared. Methods: The clonogenic surviving fractions at 2 Gy from previous publications and microarray gene-expression data from the National Cancer Institute-60 cell lines were used to measure the radiosensitivity. Seven different prediction models including three distinct multi-layered perceptrons (MLP), four different convolutional neural networks (CNN) were compared. Folded cross-validation was applied to train and evaluate model performance. The criteria for correct prediction were absolute error < 0.02 or relative error < 10%. The models were compared in terms of prediction accuracy, training time per epoch, training fluctuations, and required calculation resources. Results: The strength of MLP-based models was their fast initial convergence and short training time per epoch. They represented significantly different prediction accuracy depending on the model configuration. The CNN-based models showed relatively high prediction accuracy, low training fluctuations, and a relatively small increase in the memory requirement as the model deepens. Conclusion: Our findings suggest that a CNN-based model with moderate depth would be appropriate when the prediction accuracy is important, and a shallow MLP-based model can be recommended when either the training resources or time are limited.

• Imaging Science in Dentistry
• /
• v.54 no.1
• /
• pp.81-91
• /
• 2024

Purpose: The objective of this study was to propose a deep-learning model for the detection of the mandibular canal on dental panoramic radiographs. Materials and Methods: A total of 2,100 panoramic radiographs (PANs) were collected from 3 different machines: RAYSCAN Alpha (n=700, PAN A), OP-100 (n=700, PAN B), and CS8100 (n=700, PAN C). Initially, an oral and maxillofacial radiologist coarsely annotated the mandibular canals. For deep learning analysis, convolutional neural networks (CNNs) utilizing U-Net architecture were employed for automated canal segmentation. Seven independent networks were trained using training sets representing all possible combinations of the 3 groups. These networks were then assessed using a hold-out test dataset. Results: Among the 7 networks evaluated, the network trained with all 3 available groups achieved an average precision of 90.6%, a recall of 87.4%, and a Dice similarity coefficient (DSC) of 88.9%. The 3 networks trained using each of the 3 possible 2-group combinations also demonstrated reliable performance for mandibular canal segmentation, as follows: 1) PAN A and B exhibited a mean DSC of 87.9%, 2) PAN A and C displayed a mean DSC of 87.8%, and 3) PAN B and C demonstrated a mean DSC of 88.4%. Conclusion: This multi-device study indicated that the examined CNN-based deep learning approach can achieve excellent canal segmentation performance, with a DSC exceeding 88%. Furthermore, the study highlighted the importance of considering the characteristics of panoramic radiographs when developing a robust deep-learning network, rather than depending solely on the size of the dataset.

• Journal of Digital Convergence
• /
• v.19 no.6
• /
• pp.251-258
• /
• 2021

In this paper, we proposed an encoder-decoder model utilizing residual learning to improve the accuracy of the U-Net-based semantic segmentation method. U-Net is a deep learning-based semantic segmentation method and is mainly used in applications such as autonomous vehicles and medical image analysis. The conventional U-Net occurs loss in feature compression process due to the shallow structure of the encoder. The loss of features causes a lack of context information necessary for classifying objects and has a problem of reducing segmentation accuracy. To improve this, The proposed method efficiently extracted context information through an encoder using residual learning, which is effective in preventing feature loss and gradient vanishing problems in the conventional U-Net. Furthermore, we reduced down-sampling operations in the encoder to reduce the loss of spatial information included in the feature maps. The proposed method showed an improved segmentation result of about 12% compared to the conventional U-Net in the Cityscapes dataset experiment.

• Journal of the Korea Convergence Society
• /
• v.10 no.11
• /
• pp.1-7
• /
• 2019

Recently, various approaches to detect network attacks using machine learning have been studied and are being applied to detect new attacks and to increase precision. However, the machine learning method is dependent on feature extraction and takes a long time and complexity. It also has limitation of performace due to learning data imbalance. In this study, we propose a method to solve the degradation of classification performance due to imbalance of learning data among the limit points of detection system. To do this, we generate data using Generative Adversarial Networks (GANs) and propose a classification method using Convolutional Neural Networks (CNNs). Through this approach, we can confirm that the accuracy is improved when applied to the NSL-KDD and UNSW-NB15 datasets.

• The Journal of the Convergence on Culture Technology
• /
• v.10 no.2
• /
• pp.505-510
• /
• 2024

Since rolling contact fatigue cracks can always occur on the rail surface, which is the contact surface between wheels and rails, railway rails require thorough inspection and diagnosis to thoroughly inspect the condition of the cracks and prevent breakage. Recent detailed guidelines on the performance evaluation of track facilities present the requirements for methods and procedures for track performance evaluation. However, diagnosing and grading rail surface damage mainly relies on external inspection (visual inspection), which inevitably relies on qualitative evaluation based on the subjective judgment of the inspector. Therefore, in this study, we conducted a deep learning model study for rail surface defect detection using Fast R-CNN. After building a dataset of rail surface defect images, the model was tested. The performance evaluation results of the deep learning model showed that mAP was 94.9%. Because Fast R-CNN has a high crack detection effect, it is believed that using this model can efficiently identify rail surface defects.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2023.11a
• /
• pp.540-542
• /
• 2023

Due to the serious issues posed by facial manipulation technologies, many researchers are becoming increasingly interested in the identification of face forgeries. The majority of existing face forgery detection methods leverage powerful data adaptation ability of neural network to derive distinguishing traits. These deep learning-based detection methods frequently treat the detection of fake faces as a binary classification problem and employ softmax loss to track CNN network training. However, acquired traits observed by softmax loss are insufficient for discriminating. To get over these limitations, in this study, we introduce a novel discriminative feature learning based on Vision Transformer architecture. Additionally, a separation-center loss is created to simply compress intra-class variation of original faces while enhancing inter-class differences in the embedding space.

• The Journal of the Institute of Internet, Broadcasting and Communication
• /
• v.22 no.4
• /
• pp.171-176
• /
• 2022

CNN is a type of deep learning and is a neural network used to process images or image data. The filter traverses the image and extracts features of the image to distinguish the image. Deep learning has the characteristic that the more data, the better models can be made, and CNN uses a method of artificially increasing the amount of data by means of data augmentation such as rotation, zoom, shift, and flip to compensate for the weakness of less data. When learning CNN, we would like to check whether outline image learning is helpful in improving performance compared to conventional data augmentation techniques.