• Journal of Environmental Impact Assessment
• /
• v.31 no.4
• /
• pp.214-225
• /
• 2022

This study emphasized the soil of environmental impact assessment to devise measures to minimize the negative impact of project implementation on the environment. As a series of efforts for impact assessment procedures, a national inventory-based database was established for urban development projects, and three machine learning model performance evaluation as well as soil pollutant concentration distribution mapping were conducted. Here, nine soil pollutants were mapped to the metropolitan area of South Korea using the Random Forest model, which showed the best performance. The results of this study found that concentrations of Zn, F, and Cd were relatively concerned in Seoul, where urbanization is the most active. In addition, in the case of Hg and Cr6+, concentrations were detected below the standard, which was derived from a lack of pollutants such as industrial and industrial complexes that affect contents of heavy metals. A significant correlation between land cover and pollutants was inferred through the spatial distribution mapping of soil pollutants. Through this, it is expected that efficient soil management measures for minimizing soil pollution and planning decisions regarding the location of the project site can be established.

• Journal of the Korea Institute of Information Security & Cryptology
• /
• v.33 no.2
• /
• pp.295-303
• /
• 2023

As the cloud computing market grows, a variety of cloud services are now reliably delivered. Administrative agencies and public institutions of South Korea are transferring all their information systems to cloud systems. It is essential to develop security solutions in advance in order to safely operate cloud services, as protecting cloud services from misuse and malicious access by insiders and outsiders over the Internet is challenging. In this paper, we propose a zero trust model for cloud storage services that store sensitive data. We then verify the effectiveness of the proposed model by operating a cloud storage service. Memory, web, and network forensics are also performed to track access and usage of cloud users depending on the adoption of the zero trust model. As a cloud storage service, we use Amazon S3(Simple Storage Service) and deploy zero trust techniques such as access control lists and key management systems. In order to consider the different types of access to S3, furthermore, we generate service requests inside and outside AWS(Amazon Web Services) and then analyze the results of the zero trust techniques depending on the location of the service request.

• Land and Housing Review
• /
• v.14 no.2
• /
• pp.125-135
• /
• 2023

Drones are used in various fields, including land survey, transportation, forestry/agriculture, marine, environment, disaster prevention, water resources, cultural assets, and construction, as their industrial importance and market size have increased. In this study, image data for deep learning was collected using a mavic3 drone capturing images at a shooting altitude was 20 m with ×7 magnification. Swin Transformer and UperNet were employed as the backbone and architecture of the deep learning model. About 800 sheets of labeled data were augmented to increase the amount of data. The learning process encompassed three rounds. The Cross-Entropy loss function was used in the first and second learning; the Tversky loss function was used in the third learning. In the future, when the crack detection model is advanced through convergence with the Internet of Things (IoT) through additional research, it will be possible to detect patching or potholes. In addition, it is expected that real-time detection tasks of drones can quickly secure the detection of pavement maintenance sections.

• Korean Journal of Remote Sensing
• /
• v.39 no.5_4
• /
• pp.1085-1095
• /
• 2023

Time-series data generated from satellite data are crucial resources for change detection and monitoring across various fields. Existing research in time-series data generation primarily relies on single-image analysis to maintain data uniformity, with ongoing efforts to enhance spatial and temporal resolutions by utilizing diverse image sources. Despite the emphasized significance of time-series data, there is a notable absence of automated data collection and preprocessing for research purposes. In this paper, to address this limitation, we propose a system that automates the collection of satellite information in user-specified areas to generate time-series data. This research aims to collect data from various satellite sources in a specific region and convert them into time-series data, developing an automatic satellite image collection system for this purpose. By utilizing this system, users can collect and extract data for their specific regions of interest, making the data immediately usable. Experimental results have shown the feasibility of automatically acquiring freely available Landsat and Sentinel images from the web and incorporating manually inputted high-resolution satellite images. Comparisons between automatically collected and edited images based on high-resolution satellite data demonstrated minimal discrepancies, with no significant errors in the generated output.

• Composites Research
• /
• v.36 no.6
• /
• pp.429-434
• /
• 2023

In this study, aimed at improving the existing acoustic emission sensor for real time monitoring, a macro-fiber composite (MFC) transducer was employed as the acoustic emission sensor in the gas leak detection system. Prior to implementation, structural analysis was conducted to optimize the MFC's design. Consequently, the flexibility of the MFC facilitated excellent adherence to curved pipes, enabling the reception of acoustic emission (AE) signals without complications. Analysis of AE signals revealed substantial variations in parameter values for both high-pressure and low-pressure leaks. Notably, in the parameters of the Fast Fourier Transform (FFT) graph, the change amounted to 120% to 626% for high-pressure leaks compared to the case without leaks, and approximately 9% to 22% for low-pressure leaks. Furthermore, depending on the distance from the leak site, the magnitude of change in parameters tended to decrease as the distance increased. As the results, in the future, not only will it be possible to detect a leak by detecting the amount of parameter change in the future, but it will also be possible to identify the location of the leak from the amount of change.

• Journal of Internet of Things and Convergence
• /
• v.10 no.2
• /
• pp.1-15
• /
• 2024

The construction of smart communities is a new method and important measure to ensure the security of residential areas. In order to solve the problem of low accuracy in face recognition caused by distorting facial features due to monitoring camera angles and other external factors, this paper proposes the following optimization strategies in designing a face recognition network: firstly, a global graph convolution module is designed to encode facial features as graph nodes, and a multi-scale feature enhancement residual module is designed to extract facial keypoint features in conjunction with the global graph convolution module. Secondly, after obtaining facial keypoints, they are constructed as a directed graph structure, and graph attention mechanisms are used to enhance the representation power of graph features. Finally, tensor computations are performed on the graph features of two faces, and the aggregated features are extracted and discriminated by a fully connected layer to determine whether the individuals' identities are the same. Through various experimental tests, the network designed in this paper achieves an AUC index of 85.65% for facial keypoint localization on the 300W public dataset and 88.92% on a self-built dataset. In terms of face recognition accuracy, the proposed network achieves an accuracy of 83.41% on the IBUG public dataset and 96.74% on a self-built dataset. Experimental results demonstrate that the network designed in this paper exhibits high detection and recognition accuracy for faces in surveillance videos.

• Journal of the Korea Society of Computer and Information
• /
• v.29 no.4
• /
• pp.63-72
• /
• 2024

The size of the home interior market is growing rapidly as interest in interiors and self-interiors has increased due to the recent rapid increase in single-person households. Most of the Bathroom remodeling is done by requesting an offline interior contractor and then visually checking it in the final completion stage, so it is not easy to re-construct even if the customer is dissatisfied. Therefore, this study developed an immersive virtual reality-based Bathroom self-remodeling system that can visually check space and design from the design stage to the final stage by incorporating VR technology to feel realistic in virtual reality. The bathroom structure may select the basic structure of the predefined bathroom space or freely set the size of the bathroom space. For furniture items, you can choose the washbasin, toilet, bathtub, shower booth that are essential elements of the bathroom and others(mounting rack, trash bin, handle). The tile texture may change the texture of the selected tile at a desired wall position by selecting various textures. In particular,, the texture of the wall tile can be varied by selecting horizontal, vertical and inclined directions at the desired angle according to the needs of the user and the contractor. As a future research project, this system will enable indoor sharing between users who are physically far away in real-time, allowing direct experience in virtual spaces, connection to purchases, and request estimates.

• Journal of the Korea Computer Graphics Society
• /
• v.30 no.3
• /
• pp.149-158
• /
• 2024

Photo booths have traditionally provided a fun and easy way to capture and print photos to cherish memories. These booths allow individuals to capture their desired poses and props, sharing memories with friends and family. To enable diverse expressions, generative AI-powered photo booths have emerged. However, existing AI photo booths face challenges such as difficulty in taking group photos, inability to accurately reflect user's poses, and the challenge of applying different concepts to individual subjects. To tackle these issues, we present CINEMAPIC, a photo booth system that allows users to freely choose poses, positions, and concepts for their photos. The system workflow includes three main steps: pre-processing, generation, and post-processing to apply individualized concepts. To produce high-quality group photos, the system generates a transparent image for each character and enhances the backdrop-composited image through a small number of denoising steps. The workflow is accelerated by applying an optimized diffusion model and GPU parallelization. The system was implemented as a prototype, and its effectiveness was validated through a user study and a large-scale pilot operation involving approximately 400 users. The results showed a significant preference for the proposed system over existing methods, confirming its potential for real-world photo booth applications. The proposed CINEMAPIC photo booth is expected to lead the way in a more creative and differentiated market, with potential for widespread application in various fields.

• Journal of the Korea Computer Graphics Society
• /
• v.30 no.3
• /
• pp.171-178
• /
• 2024

Exercises using a Balance Board (BB) are effective in developing balance, strengthening core muscles, and improving physical fitness and concentration. In particular, the Smart Balance Board (SBB), which integrates with various digital content, provides appropriate feedback compared to traditional balance boards, maximizing the effectiveness of the exercise. However, most systems only offer visual and auditory feedback, failing to evaluate the impact on user engagement, interest, and the accuracy of exercise postures. This study proposes an Immersive Smart Balance Board (I-SBB) that utilizes multiple sensors to enable training with various feedback mechanisms and precise postures. The proposed system, based on Arduino, consists of a gyro sensor for measuring the board's posture, a communication module for wired/wireless communication, an infrared sensor to guide the user's foot placement, and a vibration motor for tactile feedback. The board's posture measurements are smoothly corrected using a Kalman Filter, and the multi-sensor data is processed in real-time using FreeRTOS. The proposed I-SBB is shown to be effective in enhancing user concentration and engagement, as well as generating interest, by integrating with diverse content.

• Journal of the Korea Institute of Information Security & Cryptology
• /
• v.34 no.4
• /
• pp.747-761
• /
• 2024

Recently, modern vehicles have been controlled by Electronic Control Units (ECUs), by which the safety and convenience of drivers are highly improved. It is known that a luxury vehicle has more than 100 ECUs to electronically control its function. However, the modern vehicles are getting targeted by cyber attacks because of this computer-based automotive system. To address the cyber attacks, automotive manufacturers have been developing some methods for securing their vehicles, such as automotive Intrusion Detection System (IDS). This development is only allowed to the automotive manufacturers because they have databases for their in-vehicle network (i.e., DBC Format File) which are highly confidential. This confidentiality poses a significant challenge to external researchers who attempt to conduct automotive security researches. To handle this restricted information, in this paper, we propose a method to partially understand the DBC Format File by analyzing in-vehicle network traffics. Our method is designed to analyze Controller Area Network (CAN) traffics so that checksum signals are identified in CAN Frame Data Field. Also, our method creates a Lookup Set by which a checksum signal is correctly estimated for a given message. We validate our method with the publicly accessible dataset as well as one from a real vehicle.