• Journal of Korean Tunnelling and Underground Space Association
• /
• v.24 no.6
• /
• pp.513-524
• /
• 2022

Recently, detecting damages of civil infrastructures from digital images using deep learning technology became a very popular research topic. In order to adapt those methodologies to the field, it is essential to explain robustness of deep learning models. Our research points out that the existing pixel-based deep learning model evaluation metrics are not sufficient for detecting cracks since cracks have linear appearance, and proposes a new evaluation methodology to explain crack segmentation deep learning model more rationally. Specifically, we design, implement and validate a methodology to generate tolerance buffer alongside skeletonized ground truth data and prediction results to consider overall similarity of topology of the ground truth and the prediction rather than pixel-wise accuracy. We could overcome over-estimation or under-estimation problem of crack segmentation model evaluation through using our methodology, and we expect that our methodology can explain crack segmentation deep learning models better.

• Journal of the Korea Convergence Society
• /
• v.12 no.10
• /
• pp.55-61
• /
• 2021

In this paper, we proposed a DeepLabv3+ based encoder-decoder model utilizing an attention mechanism for precise semantic segmentation. The DeepLabv3+ is a semantic segmentation method based on deep learning and is mainly used in applications such as autonomous vehicles, and infrared image analysis. In the conventional DeepLabv3+, there is little use of the encoder's intermediate feature map in the decoder part, resulting in loss in restoration process. Such restoration loss causes a problem of reducing segmentation accuracy. Therefore, the proposed method firstly minimized the restoration loss by additionally using one intermediate feature map. Furthermore, we fused hierarchically from small feature map in order to effectively utilize this. Finally, we applied an attention mechanism to the decoder to maximize the decoder's ability to converge intermediate feature maps. We evaluated the proposed method on the Cityscapes dataset, which is commonly used for street scene image segmentation research. Experiment results showed that our proposed method improved segmentation results compared to the conventional DeepLabv3+. The proposed method can be used in applications that require high accuracy.

• Journal of the Computational Structural Engineering Institute of Korea
• /
• v.34 no.5
• /
• pp.263-270
• /
• 2021

In this study, a reliability-based design optimization of a 130-m class fixed-type offshore platform, to be installed in the North Sea, was carried out, while considering environmental, material, and manufacturing uncertainties to enhance its structural safety and economic aspects. For the reliability analysis, and reliability-based design optimization of the structural integrity, unity check values (defined as the ratio between working and allowable stress, for axial, bending, and shear stresses), of the members of the offshore platform were considered as constraints. Weight of the supporting jacket structure was minimized to reduce the manufacturing cost of the offshore platform. Statistical characteristics of uncertainties were defined based on observed and measured data references. Reliability analysis and reliability-based design optimization of a jacket-type offshore structure were computationally burdensome due to the large number of members; therefore, we suggested a method for variable screening, based on the importance of their output responses, to reduce the dimension of the problem. Furthermore, a deterministic design optimization was carried out prior to the reliability-based design optimization, to improve overall computational efficiency. Finally, the optimal design obtained was compared with the conventional rule-based offshore platform design in terms of safety and cost.

• Korean Journal of Psychosomatic Medicine
• /
• v.29 no.2
• /
• pp.121-127
• /
• 2021

Objectives : The characteristics of depressive symptoms in patients with cancer is different from those in control group. However, few research has focused on the association between depressive symptoms in cancer patients. The aim of this study was to compare the network structure of depressive symptoms between patients with cancer and normal control. Methods : This study was based on cohort data from Korea National Health and Nutrition Examination Survey in 2016-2018. The Patient health Quetionnaire-9 (PHQ-9) was used to assess depressive symptoms in 599 patients with cancer and 599 age-sex matched controls. We estimated network structure of depressive symptom using Isingfit model. Results : There was no significant difference of each PHQ-9 item score. There were strong associations between symptoms were concentration problem-psychomotor activity, anhedonia-depressed mood, and depressed mood-suicidal ideation in both groups. Strength centrality of worthlessness was significantly higher in patients with cancer. Conclusions : These results suggest that worthless is associated with other depressive symptoms more tightly in patients with cancer. Worthless can serve as important treatment targets for intervention of depression in patients with cancer.

• The Journal of the Korea Contents Association
• /
• v.19 no.7
• /
• pp.199-210
• /
• 2019

Virtual Advertising, which was introduced exclusively in sports casting programs in 2010, has enlarged its scope to terrestrial TV networks' sports news, entertainment shows, and dramas by 2015. Such advertising deregulation allows broadcasting business operators to insert more various virtual advertising methods into TV programs. Despite recent evaluation that virtual advertising was deregulated to a large degree, it is still inadequate compared to foreign state of affairs and has a lot of room for growth. Therefore, this research explores a literature review of virtual advertising in other countries and considers possible ways for virtual advertising in Korea to move forward. Additionally, through in-depth interview with seven virtual advertising experts, the research unravels positive and negative impacts of virtual advertising as well as its current state of affairs and struggles. This research also analyses the regulation of virtual advertising and finally explores possible revitalization strategies. The results of the research show that it is necessary to first improve the viewers' favorable concerning virtual advertising in order to revitalize virtual advertising. Revitalization will also require a clarification of regulation as well as a more unified and consistent content review and rating system. Furthermore, it is imperative that data of advertising impact will be accessible to advertisers and that advertising regulation will loosen. Revitalization will also require a clarification of regulation as well as a more unified and consistent content review and rating system. Furthermore, it is imperative that data of advertising impact will be accessible to advertisers and that advertising regulation will loosen. It is necessary to further develop new techniques and creators of virtual advertising. The research suggests strategies and alternative paths for the growth and revitalization of the virtual advertising market in light of recently revised law.

• Industry Promotion Research
• /
• v.4 no.1
• /
• pp.79-86
• /
• 2019

The quality of service delivery by governments to their citizens is a subject of concern in the contemporary society. E-governance is a critical aspect that is transforming government operation and service delivery to citizens and other bodies through application of information technology. This research explores the state of e-governing focusing on nations that are developing with special attention to Pakistan. The difficulties and benefits encountered are identified. The results are vital for implementers of e-governing systems in these countries. The methodology applied entails a secondary research that involves analysis and synthesis of literature relating the research topic. The results reveal that Pakistan has made incredible steps in setting up e-governance systems with growth in internet use and access of data from a digital platform. The findings reveal that e-government is associated with multiple benefits including enhanced quality of services, cost efficiency in service provision, enhance transparency and elimination of corruption, provide the basis for eradication of poverty, boost economic stability of a country, and provide room for direct democracy. The research also found that developing countries experience challenges in form of financial constraints, poor ICT infrastructure, illiteracy on e-government, political consensus constraints, legal obstacles, social and cultural constraints. E-government has the capability to transform the quality of governance provided by governments, and policymakers and implementers should address the constraints that act as a hindrance to its implementation.

• Korean Journal of Psychosomatic Medicine
• /
• v.28 no.2
• /
• pp.135-142
• /
• 2020

Objectives : To identify the relationship between somatization, stress, depression, anxiety, and psychological symptoms risk for nurses working in the intensive care unit. Create clinical evidence of psychosomatic medicine research and complement the meaning of somatization. Methods : Seventy of the mental health checkups conducted by the National Mental Health Center among the nurses using tools including Perceived Stress scale, Fatigue Severity Scale, Patient Health questionnaire-15, Korean Beck Depression Inventory, Korean Beck Anxiety Inventory, and Symptom Checklist-90-Revision. Results : 12.9% of the patients experienced more than moderate somatization. There was no statistical relationship between somatization and psychological stress perception, but feeling of anxiety and decreased self-confidence were related to the level of somatization. The group with severe somatization experienced more depression and anxiety. The group with high physical fatigue also had no statistical relationship with psychological stress perception, but had an effect on the feeling of tension, stress, or decreased control. Physical fatigue level was increased by experience of depression, not by anxiety. For psychological symptoms the higher the level of somatization, the higher the obsession and hostility was explored. In the linear regression model, stress, depression, and anxiety accounted for 39.3% of somatization and 16.1% of physical fatigue symptoms. Conclusions : We can estimate the decrease in stress cognitive symptoms, accompanying depression and anxiety, compulsion and hostility as characteristics of somatization. The causal relationship between somatization and psychological symptoms cannot be confirmed in this study, but the interrelationships are observed, can be referred to mediation strategies.

• Korean Journal of Remote Sensing
• /
• v.38 no.6_4
• /
• pp.1925-1934
• /
• 2022

Monitoring nuclear activity for inaccessible areas using remote sensing technology is essential for nuclear non-proliferation. In recent years, deep learning has been actively used to detect nuclear-activity-related small objects. However, high-resolution satellite imagery containing small objects can result in class imbalance. As a result, there is a performance degradation problem in detecting small objects. Therefore, this study aims to improve detection accuracy by analyzing the effect of the ratio of small objects related to nuclear activity in the input data for the performance of the deep learning model. To this end, six case datasets with different ratios of small object pixels were generated and a U-Net model was trained for each case. Following that, each trained model was evaluated quantitatively and qualitatively using a test dataset containing various types of small object classes. The results of this study confirm that when the ratio of object pixels in the input image is adjusted, small objects related to nuclear activity can be detected efficiently. This study suggests that the performance of deep learning can be improved by adjusting the object pixel ratio of input data in the training dataset.

• Korean Journal of Psychosomatic Medicine
• /
• v.30 no.2
• /
• pp.172-178
• /
• 2022

Objectives : Although subclinical depression symptoms are associated with suicidal idea, most research have focused on clinical depression such as major depressive disorder or dysthymia. The aim of this study is to investigate network structure of depressive symptom and to reveal which symptoms are associated with suicidal ideation. Methods : We used part of data from the seventh Korea National Health and Nutrition Examination Survey. Participants were between 19 and 65 years of age (N=8,741). Network analysis with Isingfit model is used to reveal network structure of depressive symptoms and most central symptom and edges assessed by patient health questionnaire (PHQ-9). Results : The most two central symptoms were psychomotor activity and suicidal ideation. The strongest edge was psychomotor activity-suicidal ideation. Suicidal ideation also has strong association with depressive mood and worthlessness. Conclusions : These results suggest that psychomotor activity and suicidal ideation can serve as treatment target for subclinical depression and psychomotor activity, worthlessness and depressed mood may be important factor for early intervention of suicidal ideation.

• Journal of Environmental Impact Assessment
• /
• v.31 no.6
• /
• pp.449-464
• /
• 2022

This study reviewed the development process of the consultation guideline for the environmental assessment of offshore wind power. Based on this, implications and improvement plans for a future revision of the guidelines. Domestic and foreign case studies reviewed the consulting cases on domestic offshore wind power development projects, environmental location consulting cases, and guidelines related to overseas offshore wind power and analyzed location characteristics and significant environmental issues by project. Major environmental issues related to offshore wind power include birds, noise and vibration, marine animals and plants, marine physics, marine water quality and sediments, marine landscapes, and other auxiliary facilities installed on land. Implications and improvements for revising the consultation guidelines for evaluating offshore wind environments require data and clear guidelines at the central government level to determine areas where offshore wind projects can be located. In a situation where the importance of cumulative impact assessment is emphasized, guidelines for cumulative impact assessment methodologies for each item that reflect the domestic situation should be prepared for a cumulative impact assessment on offshore wind power environmental issues. In addition, when revising the consultation guidelines, empirical research cases should be reflected through the accumulation of environmental surveys and monitoring data of offshore wind farms by sea area.