• Journal of the Korean Society of Safety
• /
• v.34 no.6
• /
• pp.50-57
• /
• 2019

In this study, various types of deep learning models that have been proposed recently are classified according to data input / output types and analyzed to find the deep learning model suitable for constructing a crack detection model. First the deep learning models are classified into image classification model, object segmentation model, object detection model, and instance segmentation model. ResNet-101, DeepLab V2, Faster R-CNN, and Mask R-CNN were selected as representative deep learning model of each type. For the comparison, ResNet-101 was implemented for all the types of deep learning model as a backbone network which serves as a main feature extractor. The four types of deep learning models were trained with 500 crack images taken from real concrete structures and collected from the Internet. The four types of deep learning models showed high accuracy above 94% during the training. Comparative evaluation was conducted using 40 images taken from real concrete structures. The performance of each type of deep learning model was measured using precision and recall. In the experimental result, Mask R-CNN, an instance segmentation deep learning model showed the highest precision and recall on crack detection. Qualitative analysis also shows that Mask R-CNN could detect crack shapes most similarly to the real crack shapes.

• The Journal of the Institute of Internet, Broadcasting and Communication
• /
• v.21 no.6
• /
• pp.77-82
• /
• 2021

There are three types of urinalysis: physical test, chemical test, and microscopic test. Among these, the chemical urinalysis is an easily accessible method of the general public to compare the chemical reaction of urinalysis strip with a standard colorimetric table by sight or purchase the portable urinalysis machine separately. Currently, with the popularization of smartphone, research on the urinalysis service using smartphone is increasing. The urinalysis screening application is one of the urinalysis services using a smartphone. However, the RGB values of the urinalysis pad taken by the urinalysis screening application have large deviations due to the effect of lighting. Deviation of RGB value debases the accuracy of urinalysis discrimination. Therefore, in this paper, the accuracy of urinaylsis pad image discrimination is improved through CNN after classifying urinalysis strips taken by the urinalysis screening application based on smartphone by urinalysis pad items. Urinalysis strip was taken from various backgrounds to generate CNN image, and urinalysis discrimination was analyzed using the ResNet-50 CNN model.

• Journal of the Korea Society of Computer and Information
• /
• v.25 no.12
• /
• pp.45-53
• /
• 2020

The purpose of this paper is to reduce the ratio of the patient accidents that may occur in nursing hospitals. In other words, it determines whether the person approaching the dangerous area is a elderly (patient uniform) group or a practitioner(Casual Clothing) group, based on the clothing displayed by CCTV. We collected the basic learning data from web crawling techniques and nursing hospitals. Then model training data was created with Image Generator and Labeling program. Due to the limited performance of CCTV, it is difficult to create a good model with both high accuracy and speed. Therefore, we implemented the ResNet model with relatively excellent accuracy and the YOLO3 model with relatively excellent speed. Then we wanted to allow nursing hospitals to choose a model that they wanted. As a result of the study, we implemented a model that can distinguish patient and casual clothes with appropriate accuracy. Therefore, it is believed that it will contribute to the reduction of safety accidents in nursing hospitals by preventing the elderly from accessing the danger zone.

• Applied Chemistry for Engineering
• /
• v.22 no.6
• /
• pp.658-663
• /
• 2011

The SNCR-SCR (selective non-catalytic reduction-selective catalytic reduction) hybrid system is an economical NOx removal system. In this study, the effect of the operating parameters of the SNCR-SCR hybrid system on NOx removal efficiency was investigated. When the SNCR reactor was operated at a temperature lower than the optimum temperature ($900{\sim}950^{\circ}C$), an additional NO removal is obtained basesd on the utilization of $NH_3$ slip. On the other hand, the SNCR reactor operated above the temperature resulted in no additional NO removal of SCR due to decomposition of $NH_3$. Therefore, the SNCR process should be operated at optimum temperature to obtain high NO removal efficiency and low $NH_3$ slip. Thus, it is important to adjust NSR (normalized stoichiometric ratio) so that $SR_{RES}$ can be maintained at an appropriate level.

• Journal of Korea Society of Industrial Information Systems
• /
• v.27 no.1
• /
• pp.1-9
• /
• 2022

The goal of this paper is to create a deep learning model based on triplet loss for generating similar child drawing selection algorithms. To assess the similarity of children's drawings, the distance between feature vectors belonging to the same class should be close, and the distance between feature vectors belonging to different classes should be greater. Therefore, a similar child drawing selection algorithm was developed in this study by building a deep learning model combining Triplet Loss and residual network(ResNet), which has an advantage in measuring image similarity regardless of the number of classes. Finally, using this model's similar child drawing selection algorithm, the similarity between the target child drawing and the other drawings can be measured and drawings with a high similarity can be chosen.

• Journal of Microbiology and Biotechnology
• /
• v.33 no.4
• /
• pp.441-448
• /
• 2023

Brucellosis is a contagious zoonotic disease that infects millions of people annually with hundreds of millions more being exposed. It is caused by Brucella, a highly infectious bacterial species capable of infecting humans with an estimated dose of 10-100 organisms. Sirtuin 1 (SIRT1) has been reported to contribute to prevention of viral diseases as well as a chronic infection caused by Mycobacterium bovis. Here, we investigated the role of SIRT1 in the establishment of Brucella abortus infection in both in vitro and in vivo systems using the reported SIRT1 activators resveratrol (RES), piceatannol (PIC), and ginsenoside Rg3 (Rg3). In RAW264.7 cells, SIRT1 activators did not alter the adherence of Brucella or Salmonella Typhimurium. However, reduced uptake of Brucella was observed in cells treated with PIC and Rg3, and survival of Brucella within the cells was only observed to decrease in cells that were treated with Rg3, while PIC treatment reduced the intracellular survival of Salmonella. SIRT1 treatment in mice via oral route resulted in augmented Brucella resistance for PIC and Rg3, but not RES. PIC treatment favors Th2 immune response despite reduced serum pro-inflammatory cytokine production, while Rg3-treated mice displayed high IL-12 and IFN-γ serum production. Overall, our findings encourage further investigation into the complete mechanisms of action of the different SIRT1 activators used as well as their potential benefit as an effective alternative approach against intracellular and extracellular pathogens.

• Journal of radiological science and technology
• /
• v.46 no.6
• /
• pp.535-542
• /
• 2023

This study aimed to implement a deep learning model that can perform quantitative quality control through ACTS software used for quantitative evaluation of ACR phantom in CT quality control and evaluate its usefulness. By changing the scanning conditions, images of three modules of the ACR phantom's slice thickness (ST), low contrast resolution (LC), and high contrast resolution (HC) were obtained and classified as ACTS software. The deep learning model used ResNet18, implementing three models in which ST, HC, and LC were learned with epoch 50 and an integrated model in which three modules were learned with Epoch 10, 30, and 50 at once. The performance of each model was evaluated through Accuracy and Loss. When comparing and evaluating the accuracy and loss function values of the deep learning models by ST, LC, and HC modules, the Accuracy and Loss of the HC model were the best with 100% and 0.0081, and in the integrated model according to the Epoch value, Accuracy and Loss with epoch 50 were the best with 96.29% and 0.1856. This paper showed that quantitative quality control is possible through a deep learning model, and it can be used as a basis and evidence for applying deep learning to the CT quality control.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.16 no.3
• /
• pp.757-770
• /
• 2022

Palmprint recognition has drawn increasingly attentions in the past decade due to its uniqueness and reliability. Traditional palmprint recognition methods usually use high-resolution images as the identification basis so that they can achieve relatively high precision. However, high-resolution images mean more computation cost in the recognition process, which usually cannot be guaranteed in mobile computing. Therefore, this paper proposes an improved low-resolution palmprint image recognition method based on residual networks. The main contributions include: 1) We introduce a channel attention mechanism to refactor the extracted feature maps, which can pay more attention to the informative feature maps and suppress the useless ones. 2) The ResStage group structure proposed by us divides the original residual block into three stages, and we stabilize the signal characteristics before each stage by means of BN normalization operation to enhance the feature channel. Comparison experiments are conducted on a public dataset provided by the Hong Kong Polytechnic University. Experimental results show that the proposed method achieve a rank-1 accuracy of 98.17% when tested on low-resolution images with the size of 12dpi, which outperforms all the compared methods obviously.

• KIPS Transactions on Software and Data Engineering
• /
• v.9 no.5
• /
• pp.169-176
• /
• 2020

Recently, deep neural networks (DNNs) have brought revolutions to many mobile and embedded devices by providing human-level machine intelligence for various applications. However, high inference accuracy of such DNNs comes at high computational costs, and, hence, there have been significant efforts to reduce computational overheads of DNNs either by compressing off-the-shelf models or by designing a new small footprint DNN architecture tailored to resource constrained devices. One notable recent paradigm in designing small footprint DNN models is sharing parameters in several layers. However, in previous approaches, the parameter-sharing techniques have been applied to large deep networks, such as ResNet, that are known to have high redundancy. In this paper, we propose a parameter-sharing method for already parameter-efficient small networks such as ShuffleNetV2. In our approach, small templates are combined with small layer-specific parameters to generate weights. Our experiment results on ImageNet and CIFAR100 datasets show that our approach can reduce the size of parameters by 15%-35% of ShuffleNetV2 while achieving smaller drops in accuracies compared to previous parameter-sharing and pruning approaches. We further show that the proposed approach is efficient in terms of latency and energy consumption on modern embedded devices.

• Proceedings of the Korean Radioactive Waste Society Conference
• /
• 2004.06a
• /
• pp.319-329
• /
• 2004

The electrorefining experiments with an anode composed of U, Y, Gd, Nd and Ce (or U, Gd, Dy and Ce) were carried out in the KC1-LiCl eutectic melt at $500^{\circ}C$, Uranium was the major component in the cathode deposits at the high initial uranium concentration, and the separation factors of the uranium with respect to the rare earths (REs) were calculated according to the applied voltage and the uranium concentration in the molten salt. The current efficiency was inversely in proportion to the applied voltage in the range of 1.0 V to 1, 9 V (vs. STS304L). The dependency of the applied voltage on the current efficiency as well as the deposition rate was discussed in terms of the microstructural feature and crystal structure of the deposit.