• Journal of Ginseng Research
• /
• v.31 no.2
• /
• pp.69-73
• /
• 2007

Since chemical structures of ginsenoside as active ingredient of Panax ginseng are known, accumulating evidence have shown that ginsenoside is one of bio-active ligands through the diverse physiological and pharmacological evaluations. Chemical structures of ginsenoside could be divided into three parts depending on diol or triol ginsenoside: Steroid- or cholesterol-like backbone structure, carbohydrate portions, which are attached at the carbon-3, -6 or -20, and aliphatic side chain coupled to the backbone structure at the carbon-20. Ginsenosides also exist as stereoisomer at the carbon-20. Bioactive ligands usually exhibit the their structure-function relationships. In ginsenosides, there is little known about the relationship of chemical structure and biological activity. Recent reports have shown that ginsenoside $Rg_3$, one of active ginsenosides, exhibits its differential physiological or pharmacological actions depending on its chemical structure. This review will show how ginsenoside $Rg_3$, as a model compound, is functionally coupled to voltage-gated ion channel or ligand-gated ion channel regulations in related with its chemical structure.

• Smart Structures and Systems
• /
• v.29 no.1
• /
• pp.153-166
• /
• 2022

Identifying fine cracks in steel bridge facilities is a challenging task of structural health monitoring (SHM). This study proposed an end-to-end crack image segmentation framework based on a one-step Convolutional Neural Network (CNN) for pixel-level object recognition with high accuracy. To particularly address the challenges arising from small object detection in complex background, efforts were made in loss function selection aiming at sample imbalance and module modification in order to improve the generalization ability on complicated images. Specifically, loss functions were compared among alternatives including the Binary Cross Entropy (BCE), Focal, Tversky and Dice loss, with the last three specialized for biased sample distribution. Structural modifications with dilated convolution, Spatial Pyramid Pooling (SPP) and Feature Pyramid Network (FPN) were also performed to form a new backbone termed CrackDet. Models of various loss functions and feature extraction modules were trained on crack images and tested on full-scale images collected on steel box girders. The CNN model incorporated the classic U-Net as its backbone, and Dice loss as its loss function achieved the highest mean Intersection-over-Union (mIoU) of 0.7571 on full-scale pictures. In contrast, the best performance on cropped crack images was achieved by integrating CrackDet with Dice loss at a mIoU of 0.7670.

• Journal of the Korean Institute of Telematics and Electronics B
• /
• v.32B no.1
• /
• pp.1-10
• /
• 1995

A graphic user interface for effectively browsing object-oriented databases in complex applications is designed and implemented. The rationale behind our design lies in enabling the users of various levels and needs to investigate the database according to their respective interests and desired depths. A novel idea in our design is the introduction of zooming techniques from a logical view, which vidualize the backbone abstraction concepts of the object-oriented data model. These objectives are verified by evaluating the results of its implementation.

• Proceedings of the Korean Society for Bioinformatics Conference
• /
• 2000.11a
• /
• pp.28-31
• /
• 2000

We describe a hidden Markov model, HMMTIR, for general protein sequence based on the I-sites library of sequence-structure motifs. Unlike the linear HMMs used to model individual protein families, HMMSTR has a highly branched topology and captures recurrent local features of protein sequences and structures that transcend protein family boundaries. The model extends the I-sites library by describing the adjacencies of different sequence-structure motifs as observed in the database, and achieves a great reduction in parameters by representing overlapping motifs in a much more compact form. The HMM attributes a considerably higher probability to coding sequence than does an equivalent dipeptide model, predicts secondary structure with an accuracy of 74.6% and backbone torsion angles better than any previously reported method, and predicts the structural context of beta strands and turns with an accuracy that should be useful for tertiary structure prediction. HMMSTR has been incorporated into a public, fully-automated protein structure prediction server.

• Journal of the Korea Institute of Information Security & Cryptology
• /
• v.16 no.3
• /
• pp.165-172
• /
• 2006

There are various threats as side effects against the growth of information technology, and malicious codes such as Internet worms may bring about confusions to upset a national backbone network. In this paper, we examine the existed spreading models and propose a new worm spreading model on Internet environment. We also predict and analyze the spreading effects of high-speed Internet worms. The proposed model leads to a better prediction of the worm spreading since various factors are considered.

• Journal of IKEEE
• /
• v.27 no.3
• /
• pp.345-349
• /
• 2023

In this paper, we propose a study on the development of deep learning structure for defective pixel detection of next-generation smart LED display board using imaging device. In this research, a technique utilizing imaging devices and deep learning is introduced to automatically detect defects in outdoor LED billboards. Through this approach, the effective management of LED billboards and the resolution of various errors and issues are aimed. The research process consists of three stages. Firstly, the planarized image data of the billboard is processed through calibration to completely remove the background and undergo necessary preprocessing to generate a training dataset. Secondly, the generated dataset is employed to train an object recognition network. This network is composed of a Backbone and a Head. The Backbone employs CSP-Darknet to extract feature maps, while the Head utilizes extracted feature maps as the basis for object detection. Throughout this process, the network is adjusted to align the Confidence score and Intersection over Union (IoU) error, sustaining continuous learning. In the third stage, the created model is employed to automatically detect defective pixels on actual outdoor LED billboards. The proposed method, applied in this paper, yielded results from accredited measurement experiments that achieved 100% detection of defective pixels on real LED billboards. This confirms the improved efficiency in managing and maintaining LED billboards. Such research findings are anticipated to bring about a revolutionary advancement in the management of LED billboards.

• Journal of the Korean Society of Radiology
• /
• v.16 no.1
• /
• pp.25-34
• /
• 2022

Brain computed tomography (CT) is useful for brain lesion diagnosis, such as brain hemorrhage, due to non-invasive methodology, 3-dimensional image provision, low radiation dose. However, there has been numerous misdiagnosis owing to a lack of radiologist and heavy workload. Recently, object detection technologies based on artificial intelligence have been developed in order to overcome the limitations of traditional diagnosis. In this study, the applicability of a deep learning-based YOLOv5s model was evaluated for brain hemorrhage detection using brain CT images. Also, the effect of hyperparameters in the trained YOLOv5s model was analyzed. The YOLOv5s model consisted of backbone, neck and output modules. The trained model was able to detect a region of brain hemorrhage and provide the information of the region. The YOLOv5s model was trained with various activation functions, optimizer functions, loss functions and epochs, and the performance of the trained model was evaluated in terms of brain hemorrhage detection accuracy and training time. The results showed that the trained YOLOv5s model is able to provide a bounding box for a region of brain hemorrhage and the accuracy of the corresponding box. The performance of the YOLOv5s model was improved by using the mish activation function, the stochastic gradient descent (SGD) optimizer function and the completed intersection over union (CIoU) loss function. Also, the accuracy and training time of the YOLOv5s model increased with the number of epochs. Therefore, the YOLOv5s model is suitable for brain hemorrhage detection using brain CT images, and the performance of the model can be maximized by using appropriate hyperparameters.

• International journal of advanced smart convergence
• /
• v.8 no.4
• /
• pp.75-81
• /
• 2019

Though recent advancement of deep learning methods have provided satisfactory results from large data domain, somehow yield poor performance on few-shot classification tasks. In order to train a model with strong performance, i.e. deep convolutional neural network, it depends heavily on huge dataset and the labeled classes of the dataset can be extremely humongous. The cost of human annotation and scarcity of the data among the classes have drastically limited the capability of current image classification model. On the contrary, humans are excellent in terms of learning or recognizing new unseen classes with merely small set of labeled examples. Few-shot learning aims to train a classification model with limited labeled samples to recognize new classes that have neverseen during training process. In this paper, we increase the backbone depth of the embedding network in orderto learn the variation between the intra-class. By increasing the network depth of the embedding module, we are able to achieve competitive performance due to the minimized intra-class variation.

• Publications of The Korean Astronomical Society
• /
• v.30 no.2
• /
• pp.473-474
• /
• 2015

Halo merger trees are the essential backbone of semi-analytic models for galaxy formation and evolution. Srisawat et al. (2013) show that different tree building algorithms can build different halo merger histories from a numerical simulation for structure formation. In order to understand the differences induced by various tree building algorithms, we investigate the impact of halo merger trees on a semi-analytic model. We find that galaxy properties in our models show differences between trees when using a common parameter set. The models independently calibrated for each tree can reduce the discrepancies between global galaxy properties at z=0. Conversely, with regard to the evolutionary features of galaxies, the calibration slightly increases the differences between trees. Therefore, halo merger trees extracted from a common numerical simulation using different, but reliable, algorithms can result in different galaxy properties in the semi-analytic model. Considering the uncertainties in baryonic physics governing galaxy formation and evolution, however, these differences may not necessarily be significant.

• Structural Engineering and Mechanics
• /
• v.76 no.3
• /
• pp.355-365
• /
• 2020

In this study, considering the hydroelastic response represented by the springing and whipping phenomena, we propose a method of estimating the fatigue damage in the longitudinal connections of ships. First, we screened the design sea states using a load transfer function based on the frequency domain. We then conducted a time domain fluid-structure interaction (FSI) analysis using WISH-FLEX, an in-house code based on the weakly nonlinear approach. To obtain an effective and robust analytical result of the hydroelastic response, we also conducted an experimental model test with a 1/50-scale backbone-based model of a ship, and compared the experimental results with those obtained from the FSI analysis. Then, by combining the results obtained from the hydroelastic response with those obtained from the numerical fatigue analysis, we developed a fatigue damage estimation method. Finally, to demonstrate the effectiveness of the developed method, we evaluated the fatigue strength for the longitudinal connections of the real ship and compared it with the results obtained from the model tests.