• Journal of the Korea Society of Computer and Information
• /
• v.29 no.2
• /
• pp.31-41
• /
• 2024

Deep learning models show excellent performance in tasks such as image classification and object detection in the field of computer vision, and are used in various ways in actual industrial sites. Recently, research on improving robustness has been actively conducted, along with pointing out that this deep learning model is vulnerable to hostile examples. A hostile example is an image in which small noise is added to induce misclassification, and can pose a significant threat when applying a deep learning model to a real environment. In this paper, we tried to confirm the robustness of the edge-learning classification model and the performance of the adversarial example detection model using it for adversarial examples of various algorithms. As a result of robustness experiments, the basic classification model showed about 17% accuracy for the FGSM algorithm, while the edge-learning models maintained accuracy in the 60-70% range, and the basic classification model showed accuracy in the 0-1% range for the PGD/DeepFool/CW algorithm, while the edge-learning models maintained accuracy in 80-90%. As a result of the adversarial example detection experiment, a high detection rate of 91-95% was confirmed for all algorithms of FGSM/PGD/DeepFool/CW. By presenting the possibility of defending against various hostile algorithms through this study, it is expected to improve the safety and reliability of deep learning models in various industries using computer vision.

• Journal of Internet Computing and Services
• /
• v.25 no.1
• /
• pp.91-98
• /
• 2024

With the advancement of ICT (Information and Communication Technology), searching for judgments through the internet has become increasingly convenient. However, predicting sentencing based on judgments remains a challenging task for individuals. This is because sentencing involves a complex process of applying aggravating and mitigating factors within the framework of legal provisions, and it often depends on the subjective judgment of the judge. Therefore, this research aimed to develop a model for predicting sentencing using artificial intelligence by focusing on structuring the data from judgments, making it suitable for AI applications. Through theoretical and statistical analysis of previous studies, we identified variables with high explanatory power for predicting sentencing. Additionally, by analyzing 50 legal judgments related to serious crimes that are publicly available, we presented a framework for extracting essential information from judgments. This framework encompasses basic case information, sentencing details, reasons for sentencing, the reasons for the determination of the sentence, as well as information about offenders, victims, and accomplices evident within the specific content of the judgments. This research is expected to contribute to the development of artificial intelligence technologies in the field of law in the future.

• Journal of the Korean Geotechnical Society
• /
• v.40 no.1
• /
• pp.15-28
• /
• 2024

This paper proposes a method to acquire image data inside tunnel structures and a method to structure the acquired image data. By improving the conditions by which image data are acquired inside the tunnel structure, high-quality image data can be obtained from area type tunnel scanning. To improve the data acquisition conditions, a longitudinal rail of the tunnel can be installed on the tunnel ceiling, and image data of the entire tunnel structure can be acquired by moving the installed rail. This study identified 0.5 mm cracked simulation lines under a distance condition of 20 m at resolutions of 3,840 × 2,160 and 720 × 480 pixels. In addition, the proposed image-data-structuring method could acquire image data in image tile units. Here, the image data of the tunnel can be structured by substituting the application factors (resolution of the acquired image and the tunnel size) into a relationship equation. In an experiment, the image data of a tunnel with a length of 1,000 m and a width of 20 m were obtained with a minimum overlap rate of 0.02% to 8.36% depending on resolution and precision, and the size of the local coordinate system was found to be (14 × 15) to (36 × 34) pixels.

• The Korea Journal of Herbology
• /
• v.39 no.3
• /
• pp.107-114
• /
• 2024

Objective : Cicadae Periostracum (CP), which is the discarded shell of the Cryptotympana atrata (Fabricius, 1775), is a recognized component of oriental medicine for treatment sore throat, itching, shock, sedation, edema. However, the safety and toxicity of CP have not yet been established. It has been reported that symptoms of addiction or side effects may occur in patients who take high doses of CP or who are hypersensitive to it. Therefore, we investigated the acute toxicity of an CP extracts in Sprague-Dawley (SD) rats. Methods : To study acute toxicity, five SD rats of each sex per group were treated with CP extracts at single doses of 0, 500, 1000, or 2000 mg/kg administrated by oral gavage, and body weight, clinical signs, and mortality were observed after dosing. At the end of 14-day observation period, all animals were sacrificed and complete hematological and macroscopic examinations were performed. Results : There were no dead animal and test article-related effects on body weight change or the gross finding. No toxicologically significant results were observed between control and treated groups in hematology. Although salivation related to stress at the highest dose was observed in clinical signs immediately after administration, it is considered to have no toxicological significance. Conclusion : As the results, we did not find any adverse effect at the dose levels of 500, 1000, or 2000 mg/kg in rats. The minimal lethal dose was considered to be over 2000 mg/kg body weight in rats.

• Korean Journal of Construction Engineering and Management
• /
• v.25 no.1
• /
• pp.62-72
• /
• 2024

Due to the distribution structure of domestic temporary equipment, quality control of temporary equipment is essential because more than 80% of temporary equipment is repeated and reused. Due to this importance, the Ministry of Land, Infrastructure and Transport has proposed quality management standards for temporary equipment for 10types of temporary equipments, including steel pipe support, but the overall quality of temporary eqipment cannot be confirmed because the quality is managed through sampling quality tests. In addition, although quality control standards exist for temporary material rental company, practical utilization was investigated and analyzed to be very low as they are mainly presented based on qualitative inspection standards by visual inspection. Therefore, the purpose of this study is to develop individual temporary material/quality/delivery management standards (Guides) for temporary equipment rental company to preemptively secure the quality of temporary equipment before bringing them into the construction site. If the standards developed through this study are applied to domestic temporary equipment rental companies, it is expected that high-quality temporary equipment with secured quality will be brought into the construction site as the quality of temporary equipment quality of domestic medium and small temporary equipment rental companies is improved safety accidents related to temporary structures.

• Smart Media Journal
• /
• v.13 no.3
• /
• pp.18-26
• /
• 2024

Road surface monitoring is essential for maintaining road environment safety through managing risk factors like rutting and crack detection. Using autonomous driving-based UGVs with high-performance 2D laser sensors enables more precise measurements. However, the increased energy consumption of these sensors is limited by constrained battery capacity. In this paper, we propose a fusion sensor system for efficient surface monitoring with UGVs. The proposed system combines color information from cameras and depth information from line laser sensors to accurately detect surface displacement. Furthermore, a dynamic sampling algorithm is applied to control the scanning frequency of line laser sensors based on the detection status of monitoring targets using camera sensors, reducing unnecessary energy consumption. A power consumption model of the fusion sensor system analyzes its energy efficiency considering various crack distributions and sensor characteristics in different mission environments. Performance analysis demonstrates that setting the power consumption of the line laser sensor to twice that of the saving state when in the active state increases power consumption efficiency by 13.3% compared to fixed sampling under the condition of λ=10, µ=10.

• Journal of the Computational Structural Engineering Institute of Korea
• /
• v.37 no.3
• /
• pp.163-171
• /
• 2024

Structural health monitoring for ships and offshore structures is important in various aspects. Ships and offshore structures are continuously exposed to various environmental conditions, such as waves, wind, and currents. In the event of an accident, immense economic losses, environmental pollution, and safety problems can occur, so it is necessary to detect structural damage or defects early. In this study, structural response data of multi-linked floating offshore structures under various wave load conditions was calculated by performing fluid-structure coupled analysis. Furthermore, the order reduction method with distortion base mode was applied to the structures for predicting the structural response by using the results of numerical analysis. The distortion base mode order reduction method can predict the structural response of a desired area with high accuracy, but prediction performance is affected by sensor arrangement. Optimization based on a genetic algorithm was performed to search for optimal sensor arrangement and improve the prediction performance of the distortion base mode-based reduced-order model. Consequently, a sensor arrangement that predicted the structural response with an error of about 84.0% less than the initial sensor arrangement was derived based on the root mean squared error, which is a prediction performance evaluation index. The computational cost was reduced by about 8 times compared to evaluating the prediction performance of reduced-order models for a total of 43,758 sensor arrangement combinations. and the expected performance was overturned to approximately 84.0% based on sensor placement, including the largest square root error.

• Food Science and Preservation
• /
• v.30 no.2
• /
• pp.262-271
• /
• 2023

The objective of this study is to compare and assess the effectiveness of real-time polymerase chain reaction (RT-PCR), loop-mediated isothermal amplification (LAMP), and the selective agar plate method for the detection of Salmonella spp. and Listeria monocytogenes in ready-to-eat (RTE) foods. In RTE foods, the detection performance of the three methods (RT-PCR [SureTect^TM kit and PowerChek^TM kit], LAMP [3M MDS], selective agar) were similar at 0-10, 10-50, 50-100, and 100- CFU/mL of Salmonella spp. and L. monocytogenes. We found that with RT-PCR, the Ct value of salad was significantly higher (p<0.05) than other RTE foods, indicating that fiber plays a critical role as an obstacle to the rapid detection of Salmonella spp. However, the Ct value displayed a mixed pattern according to the inoculation level of L. monocytogenes. The use of rapid detection kits and machines mostly depends on the user's choice, with accuracy, ease of use, and economy being the primary considerations. As an RT-PCR kit, SureTect^TM and PowerChek^TM showed high accuracy in detecting Salmonella spp. and L. monocytogenes in RTE foods, showing that they can replace the existing RT-PCR kits available. Additionally, LAMP also showed excellent detection performance, suggesting that it has the potential to be used as a food safety management tool.

• The Journal of the Institute of Internet, Broadcasting and Communication
• /
• v.24 no.1
• /
• pp.125-132
• /
• 2024

The purpose of this study is to discover a drone utilization model tailored to local characteristics, propose directions for building a drone demonstration city based on demand surveys for drone activation, and suggest ways to utilize and support a drone application system. First, according to the survey results, there was a high understanding of and necessity for drone demonstration projects, particularly in addressing urban issues, which were deemed to have a significant impact. Second, based on the analysis of priorities and short- and long-term approaches, disaster-related tasks were evaluated as a priority, requiring an approach through medium- to long-term strategies. Third, it was noted that budgetary considerations emerged as the most critical issue during project implementation. Practitioners and experts expressed willingness to actively introduce drone-based technologies into their work when budget and technology were ready. Budgetary constraints were identified as the most significant obstacle to proper implementation, emphasizing the need for resolution. Fourth, the necessity of demand surveys during project development was identified in certain areas. Demand surveys were deemed essential for drone-based demonstration city construction, and a survey indicated that public leadership in this regard was also necessary. Fifth, concerning approaches in specific areas, the field of safety and disaster management was highlighted as the most crucial for application.

• Steel and Composite Structures
• /
• v.50 no.3
• /
• pp.281-298
• /
• 2024

Stainless steel bolts (SSB) are increasingly utilized in bolted steel connections due to their good mechanical performance and excellent corrosion resistance. Fire accidents, which commonly occur in engineering scenarios, pose a significant threat to the safety of steel frames. The post-fire behavior of SSB has a significant influence on the structural integrity of steel frames, and neglecting the effect of temperature can lead to serious accidents in engineering. Therefore, it is important to evaluate the performance of SSB at elevated temperatures and their residual strength after a fire incident. To investigate the mechanical behavior of SSB after fire, 114 bolts with grades A4-70 and A4-80, manufactured from 316L austenitic stainless steel, were subjected to elevated temperatures ranging from 20℃ to 1200℃. Two different cooling methods commonly employed in engineering, namely cooling at ambient temperatures (air cooling) and cooling in water (water cooling), were used to cool the bolts. Tensile tests were performed to examine the influence of elevated temperatures and cooling methods on the mechanical behavior of SSB. The results indicate that the temperature does not significantly affect the Young's modulus and the ultimate strength of SSB. Up to 500℃, the yield strength increases with temperature, but this trend reverses when the temperature exceeds 500℃. In contrast, the ultimate strain shows the opposite trend. The strain hardening exponent is not significantly influenced by the temperature until it reaches 500℃. The cooling methods employed have an insignificant impact on the performance of SSB. When compared to high-strength bolts, 316L austenitic SSB demonstrate superior fire resistance. Design models for the post-fire mechanical behavior of 316L austenitic SSB, encompassing parameters such as the elasticity modulus, yield strength, ultimate strength, ultimate strain, and strain hardening exponent, are proposed, and a more precise stress-strain model is recommended to predict the mechanical behavior of 316L austenitic SSB after a fire incident.