• Journal of the Korean Society of Surveying, Geodesy, Photogrammetry and Cartography
• /
• v.39 no.2
• /
• pp.93-101
• /
• 2021

Recently, with the urban redevelopment and the spread of the planned cities, there is increasing interest in the wind environment, which is related not only to design of buildings and landscaping but also to the comfortability of pedestrians. Numerical analysis for wind environment prediction is underway in many fields, such as dense areas of high-rise building or composition of the apartment complexes, a precisive 3D building model is essentially required in this process. Many studies conducted for wind environment analysis have typically used the method of creating a 3D model by utilizing the building layer included in the GIS (Geographic Information System) data. These data can easily and quickly observe the flow of atmosphere in a wide urban environment, but cannot be suitable for observing precisive flow of atmosphere, and in particular, the effect of a complicated structure of a single building on the flow of atmosphere cannot be calculated. Recently, drone photogrammetry has shown the advantage of being able to automatically perform building modeling based on a large number of images. In this study, we applied photogrammetry technology using a drone to evaluate the flow of atmosphere around two buildings located close to each other. Two 3D models were made into an automatic modeling technique and manual modeling technique. Auto-modeling technique is using an automatically generates a point cloud through photogrammetry and generating models through interpolation, and manual-modeling technique is a manually operated technique that individually generates 3D models based on point clouds. And then the flow of atmosphere for the two models was compared and analyzed. As a result, the wind environment of the two models showed a clear difference, and the model created by auto-modeling showed faster flow of atmosphere than the model created by manual modeling. Also in the case of the 3D mesh generated by auto-modeling showed the limitation of not proceeding an accurate analysis because the precise 3D shape was not reproduced in the closed area such as the porch of the building or the bridge between buildings.

• KIPS Transactions on Computer and Communication Systems
• /
• v.9 no.12
• /
• pp.291-306
• /
• 2020

Nowadays, Data-Network-AI (DNA)-based intelligent services and applications have become a reality to provide a new dimension of services that improve the quality of life and productivity of businesses. Artificial intelligence (AI) can enhance the value of IoT data (data collected by IoT devices). The internet of things (IoT) promotes the learning and intelligence capability of AI. To extract insights from massive volume IoT data in real-time using deep learning, processing capability needs to happen in the IoT end devices where data is generated. However, deep learning requires a significant number of computational resources that may not be available at the IoT end devices. Such problems have been addressed by transporting bulks of data from the IoT end devices to the cloud datacenters for processing. But transferring IoT big data to the cloud incurs prohibitively high transmission delay and privacy issues which are a major concern. Edge computing, where distributed computing nodes are placed close to the IoT end devices, is a viable solution to meet the high computation and low-latency requirements and to preserve the privacy of users. This paper provides a comprehensive review of the current state of leveraging deep learning within edge computing to unleash the potential of IoT big data generated from IoT end devices. We believe that the revision will have a contribution to the development of DNA-based intelligent services and applications. It describes the different distributed training and inference architectures of deep learning models across multiple nodes of the edge computing platform. It also provides the different privacy-preserving approaches of deep learning on the edge computing environment and the various application domains where deep learning on the network edge can be useful. Finally, it discusses open issues and challenges leveraging deep learning within edge computing.

• Journal of the Korea institute for structural maintenance and inspection
• /
• v.24 no.6
• /
• pp.10-16
• /
• 2020

Deterioration of reinforced concrete structure caused by chloride ingress is the main issue and regrading this, many studies have been investigated with both experiments and computational modelling. In addition to chloride diffusion, chloride sorption should be considered as a chloride transport mechanism in concrete structure and cracks formed in concrete structures are the main variable to evaluate the performance of the structures. In this study, after making two types of cracks width (0.1 and 0.3 mm) in reinforced concretes, chloride absorption tests were performed. Weight change and colour change using 0.1 AgNO3 solution from the samples were performed to measure chloride ingress. Image processing was also carried out to quantify range of colour change in carck face. From the result, it were confirmed that the amount of chloride absorption increases with exposure time and increasing crack width, and chlorides reached at steel depth within 1 hour. It would be possible that chloride can move through interface bewteen steel and concrete, thereby further study regarding this is required.

• Journal of Radiation Protection and Research
• /
• v.46 no.1
• /
• pp.14-23
• /
• 2021

Background: For radiological protection and control, the International Commission on Radiological Protection (ICRP) provides the nominal risk coefficients related to radiation exposure, which can be extrapolated using the excess relative risk and excess absolute risk obtained from the Life Span Study of atomic bomb survivors in Hiroshima and Nagasaki with the dose and dose-rate effectiveness factor (DDREF). Materials and Methods: Since it is impossible to directly estimate the radiation risk at doses less than approximately 100 mSv only from epidemiological knowledge and data, support from radiation biology is absolutely imperative, and thus, several national and international bodies have advocated the importance of bridging knowledge between biology and epidemiology. Because of the accident at the Tokyo Electric Power Company (TEPCO)'s Fukushima Daiichi Nuclear Power Station in 2011, the exposure of the public to radiation has become a major concern and it was considered that the estimation of radiation risk should be more realistic to cope with the prevailing radiation exposure situation. Results and Discussion: To discuss the issues from wide aspects related to radiological protection, and to realize bridging knowledge between biology and epidemiology, we have established a research group to develop low-dose and low-dose-rate radiation risk estimation methodology, with the permission of the Japan Health Physics Society. Conclusion: The aim of the research group was to clarify the current situation and issues related to the risk estimation of low-dose and low-dose-rate radiation exposure from the viewpoints of different research fields, such as epidemiology, biology, modeling, and dosimetry, to identify a future strategy and roadmap to elucidate a more realistic estimation of risk against low-dose and low-dose-rate radiation exposure.

• Journal of the Korean Recycled Construction Resources Institute
• /
• v.9 no.1
• /
• pp.85-91
• /
• 2021

Aggregate waste separator with rotating separating net was designed for applying classification process of construction waste. In order to evaluate the performance of the aggregate waste separator, according to the type of waste, standardized waste samples are prepared using acrylic. The appropriate operating point was evaluated by the classification efficiency and misclassification rate of recycled aggregate according to the control frequency of the blower operating and inlet position of the separating net. The classification efficiency at the operating point of the aggregate waste separator was evaluated through flow analysis assuming recycled aggregate and waste sample as particles. As a result of the performance test, when the distance. between the conveyor belt and the inlet was 0.2m, the classification efficiency was 95%, but the misclassification rate of recycled aggregate was 2% or more, which satisfies the classification efficiency and the misclassification rate of less than 2%. The operating point was shown at a control frequency of 58Hz at a suction distance of 0.254m. As a resu lt of flow analysis, there was no misclassification of recycled aggregate. In order to redu ce constru ction waste in the existing recycled aggregate production process, adsorption system using a rotating separating net that can be operated as an installation type was built.

• Journal of the Korea Convergence Society
• /
• v.12 no.3
• /
• pp.197-204
• /
• 2021

This study relates to a device that increases efficiency by reducing air bubbles in a hydraulic system used in hydraulic machinery. The reverse design and product production of the bubble reduction device, which is a commercial product overseas, was carried out. Overseas commercial products were set as the base model, a rotary rotor and an inclined rotor were added to increase the surface area of the fluid, and an annular equal distribution part with a slot in the lower part was additionally applied to distribute the fluid evenly. In addition, internal flow trends were analyzed and a system that evenly distributes the linear flow of fluid was selected as the first improvement model. Based on the first improvement model, a case where the angle of the inclined rotor is 45° was selected as the second improvement model. Based on this, as a result of setting the exit width of the annular equally distributed part as a variable, the bubble reduction efficiency was highest when the lower slot diameter of the annular part was 10mm. Finally, the system in which the average cross-sectional flow velocity decreased by 147% compared to the Base Model was derived as the final improved model.

• Smart Media Journal
• /
• v.9 no.4
• /
• pp.176-186
• /
• 2020

The size and shape of crops are diverse, and the growing environment is also different. Therefore, when one uses a drone to spray pesticides, the characteristics of each crop must be considered, and flight conditions such as the flight height and forwarding velocity of the drone should be changed. The droplet flow of pesticides is affected by various flight conditions, and a large change occurs in the sprayed area. As a result, an uneven distribution of liquid may be formed at the wake, and the transport efficiency will be decreased as well as there would be a risk of toxic scatter. Therefore, this paper analyzes the degree of distribution of pesticides to the crops through numerical analysis when pesticide is sprayed onto the selected three crops with different characteristics by using agricultural drones with different flight conditions. On the purpose of establishing a guideline for spraying pesticides using a drone in accordance with the characteristics of crops, this paper compares the amount of pesticides distributed in the crops at the wake of nozzle flow using the figure of merit, and the sum of transported liquid rate divided by the root mean square of the probability density function.

• Journal of Biomedical Engineering Research
• /
• v.42 no.4
• /
• pp.150-158
• /
• 2021

Intracranial hemorrhage (ICH) refers to acute bleeding inside the intracranial vault. Not only does this devastating disease record a very high mortality rate, but it can also cause serious chronic impairment of sensory, motor, and cognitive functions. Therefore, a prompt and professional diagnosis of the disease is highly critical. Noninvasive brain imaging data are essential for clinicians to efficiently diagnose the locus of brain lesion, volume of bleeding, and subsequent cortical damage, and to take clinical interventions. In particular, computed tomography (CT) images are used most often for the diagnosis of ICH. In order to diagnose ICH through CT images, not only medical specialists with a sufficient number of diagnosis experiences are required, but even when this condition is met, there are many cases where bleeding cannot be successfully detected due to factors such as low signal ratio and artifacts of the image itself. In addition, discrepancies between interpretations or even misinterpretations might exist causing critical clinical consequences. To resolve these clinical problems, we developed a diagnostic model predicting intracranial bleeding and its subtypes (intraparenchymal, intraventricular, subarachnoid, subdural, and epidural) by applying deep learning algorithms to CT images. We also constructed a visualization tool highlighting important regions in a CT image for predicting ICH. Specifically, 1) 27,758 CT brain images from RSNA were pre-processed to minimize the computational load. 2) Three different CNN-based models (ResNet, EfficientNet-B2, and EfficientNet-B7) were trained based on a training image data set. 3) Diagnosis performance of each of the three models was evaluated based on an independent test image data set: As a result of the model comparison, EfficientNet-B7's performance (classification accuracy = 91%) was a way greater than the other models. 4) Finally, based on the result of EfficientNet-B7, we visualized the lesions of internal bleeding using the Grad-CAM. Our research suggests that artificial intelligence-based diagnostic systems can help diagnose and treat brain diseases resolving various problems in clinical situations.

• The Journal of Korea Institute of Information, Electronics, and Communication Technology
• /
• v.15 no.1
• /
• pp.56-61
• /
• 2022

In this paper, we classify and present AI algorithms and natural language processing methods used in chatbots. A framework that can be used to implement a chatbot is also described. A chatbot is a system with a structure that interprets the input string by constructing the user interface in a conversational manner and selects an appropriate answer to the input string from the learned data and outputs it. However, training is required to generate an appropriate set of answers to a question and hardware with considerable computational power is required. Therefore, there is a limit to the practice of not only developing companies but also students learning AI development. Currently, chatbots are replacing the existing traditional tasks, and a practice course to understand and implement the system is required. RNN and Char-CNN are used to increase the accuracy of answering questions by learning unstructured data by applying technologies such as deep learning beyond the level of responding only to standardized data. In order to implement a chatbot, it is necessary to understand such a theory. In addition, the students presented examples of implementation of the entire system by utilizing the methods that can be used for coding education and the platform where existing developers and students can implement chatbots.

• Journal of the Korean Chemical Society
• /
• v.66 no.3
• /
• pp.185-193
• /
• 2022

The theoretical investigation has been performed to predict thermodynamic stability, density, detonation velocity, and detonation pressure of energetic salts produced by pairing of nitrogen-rich anions (tetrazine, oxadiazole etc.) and cations (NH₃OH⁺, NH₂NH₃⁺, CH₉N₆⁺, C₂H₆N₅⁺). All possible geometries and the binding energy for the trigger bond of energetic salts have been optimized at the B3LYP/cc-pVDZ level of theory. The detonation velocity and detonation pressure have been calculated using Kamlet-Jacobs equation, while enthalpy has been predicted at the G2MP2 level of theory. The predicted results reveal that the energetic salts including small sized NH₃OH⁺(1) and NH₂NH₃⁺(2) cations increase detonation property. And also the energetic salts including more amino group (-NH₂) such as CH₉N₆⁺(3) cation increase thermodynamic stability. These results provide basic information for the development the high energy density materials (HEDMs).