• The Journal of The Korea Institute of Intelligent Transport Systems
• /
• v.22 no.5
• /
• pp.137-147
• /
• 2023

LiDAR is used in autonomous driving and various industrial fields to measure the size and distance of an object. In addition, the sensor also provides intensity images based on the amount of reflected light. This has a positive effect on sensor data processing by providing information on the shape of the object. LiDAR guarantees higher performance as the resolution increases but at an increased cost. These conditions also apply to LiDAR intensity images. Expensive equipment is essential to acquire high-resolution LiDAR intensity images. This study developed artificial intelligence to improve low-resolution LiDAR intensity images into high-resolution ones. Therefore, this study performed parameter analysis for the optimal super-resolution neural network model. The super-resolution algorithm was trained and verified using 2,500 LiDAR intensity images. As a result, the resolution of the intensity images were improved. These results can be applied to the autonomous driving field and help improve driving environment recognition and obstacle detection performance

• Journal of the Institute of Convergence Signal Processing
• /
• v.24 no.1
• /
• pp.15-20
• /
• 2023

In general, maintaining an incorrect sitting posture for a long time is widely known to adversely affect the spine. Recently, several researchers have been interested in the causal relationship between incorrect sitting posture and spinal diseases, and have been studying methods to precisely measure changes in sitting or standing posture to prevent spinal diseases. In previous studies, we have developed a sensor device capable of measuring real-time posture change, applied a momentum calculation algorithm to improve the accuracy of real-time posture change measurement, and verified the accuracy of the postural change measurement sensor. In this study, we developed a posture measurement and analysis device that considers changes in the center of body pressure through the developed sitting pressure measurement, and it confirmed the sensor as an auxiliary tool to increase the accuracy of posture correction training with improving the user's visual feedback.

• Journal of the Economic Geographical Society of Korea
• /
• v.26 no.3
• /
• pp.310-323
• /
• 2023

Urbanization and income distribution issues are global interest, and the results of studies on the impact of urbanization on income inequality are different for each country and period. This study analyzes the impact of urbanization on income inequality using regional data from 2000-2021. In particular, serial correlation, spatial dependence, and common factor effects of the Gini coefficient are confirmed and analyzed through a dynamic spatial panel regression model. As a result, urbanization has a positive effect on reducing income inequality. Therefore, it is necessary to continuously promote regional urbanization to improve the income distribution problem. Areas with already high urbanization rates should reduce income inequality by narrowing the wage gap by expanding training opportunities for low-skilled workers, and need to come up with measures to prevent counter-urbanization.

• Convergence Security Journal
• /
• v.24 no.2
• /
• pp.201-207
• /
• 2024

Voice is a critical element of human communication, and the development of speech recognition models is one of the significant achievements in artificial intelligence, which has recently been applied in various aspects of human life. The application of speech recognition models in the military field is also inevitable. However, before artificial intelligence models can be applied in the military, it is necessary to research their vulnerabilities. In this study, we evaluates the military applicability of the multilingual speech recognition model "Whisper" by examining its vulnerabilities to battlefield noise, white noise, and adversarial attacks. In experiments involving battlefield noise, Whisper showed significant performance degradation with an average Character Error Rate (CER) of 72.4%, indicating difficulties in military applications. In experiments with white noise, Whisper was robust to low-intensity noise but showed performance degradation under high-intensity noise. Adversarial attack experiments revealed vulnerabilities at specific epsilon values. Therefore, the Whisper model requires improvements through fine-tuning, adversarial training, and other methods.

• Journal of Advanced Navigation Technology
• /
• v.28 no.1
• /
• pp.60-67
• /
• 2024

The majority of civil aviation accidents are caused by human factors, and especially for rotary-wing aircraft, accidents often occur in situations where pilots unexpectedly or unintentionally enter into instrument meteorological conditions (IIMC). This research analyzed the error rates of rotary-wing aircraft pilots under low visibility conditions from various angles to gain insights into flight characteristics and to explore measures to reduce accidents in IIMC situations. The occurrence rate of errors by pilots under low visibility conditions was examined using a flight simulator equipped with motion, with 65 pilots participating in the experiment. Flight data obtained through the experiment were used to aggregate and analyze the number of errors under various conditions, such as reductions in flight visibility, the presence or absence of spatial disorientation, and the pilot's qualifications. The analysis revealed peculiarities in flight characteristics under various conditions, and significant differences were found in the rate of error occurrence according to the pilot's qualification level, possession of instrument flight rules (IFR) qualifications, and during different phases of flight. The results of this research are expected to contribute significantly to the prevention of aircraft accidents in IIMC situations by improving pilot education and training programs.

• The Journal of the Korea institute of electronic communication sciences
• /
• v.19 no.1
• /
• pp.301-307
• /
• 2024

This study introduces a novel cascade fusion architecture aimed at enhancing thermal images across various scale conditions. The processing of thermal images at multiple scales has been challenging due to the limitations of existing methods that are designed for specific scales. To overcome these limitations, this paper proposes a unified framework that utilizes cascade feature fusion to effectively learn multi-scale representations. Confidence maps from different image scales are fused in a cascaded manner, enabling scale-invariant learning. The architecture comprises end-to-end trained convolutional neural networks to enhance image quality by reinforcing mutual scale dependencies. Experimental results indicate that the proposed technique outperforms existing methods in multi-scale thermal image enhancement. Performance evaluation results are provided, demonstrating consistent improvements in image quality metrics. The cascade fusion design facilitates robust generalization across scales and efficient learning of cross-scale representations.

• The Journal of the Korea institute of electronic communication sciences
• /
• v.19 no.3
• /
• pp.563-570
• /
• 2024

This paper proposes a method to recognize and track drivable lane areas to assist the driver. The main topic is designing a deep-based network that predicts drivable road areas using computer vision and deep learning technology based on images acquired in real time through a camera installed in the center of the windshield inside the vehicle. This study aims to develop a new model trained with data directly obtained from cameras using the YOLO algorithm. It is expected to play a role in assisting the driver's driving by visualizing the exact location of the vehicle on the actual road consistent with the actual image and displaying and tracking the drivable lane area. As a result of the experiment, it was possible to track the drivable road area in most cases, but in bad weather such as heavy rain at night, there were cases where lanes were not accurately recognized, so improvement in model performance is needed to solve this problem.

• The Journal of the Convergence on Culture Technology
• /
• v.10 no.1
• /
• pp.81-87
• /
• 2024

This study aimed to develop an Interprofessinal Education(IPE) program for third-year healthcare students to provide patient safety-oriented services and demonstrate professionalism, and to determine the effects of applying the program for five days on patient safety knowledge and patient safety performance confidence. Key topics included understanding job roles by profession, training in patient risk prediction, scenario-based patient experience, and strategies for identifying improvement. As a result of the study, after the application of the IPE program, the patient safety knowledge decreased statistically significantly from 39 points to 37 points(p=.007). The patient safety performance confidence increased statistically significantly from 6.71 pints to 7.50 points(p<.001). In addition, students who experienced clinical practice had higher patient safety knowledge after applying the IPE program, but there was no difference in patient safety performance. Repeated studies are recommended to prove the effectiveness of the IPE program, and specific measures should be taken to expand and continuously manage the IPE program.

• Journal of the Korean Institute of Gas
• /
• v.28 no.2
• /
• pp.84-92
• /
• 2024

This study applied the User-Centered Design(UCD) to develop effective VR safety training content for specific users. The UCD-based design was tailored to the VR, facilitating efficient design activities. The UCD process comprises key activities: deriving design concepts from user needs, designing with VR features, developing prototypes, conducting comprehensive evaluations with experts and users, and completing the finals. Unlike traditional UCD, this flexible approach allows iterative cycles, enhancing the quality and user satisfaction of VR safety training contents.

• Journal of the Korea Society of Computer and Information
• /
• v.29 no.11
• /
• pp.113-118
• /
• 2024

This paper proposes an optimized fruit set prediction model for cucumbers using an AI-based automatic growth control system. Based on data collected from experimental farms at Sunchon National University and Suncheon Bay cucumber farms, we constructed and compared the performance of models using three machine learning algorithms: Random Forest, XGBoost, and LightGBM. The models were trained using 19 environmental and growth-related variables, including temperature, humidity, and CO2 concentration. The LightGBM model showed the best performance (RMSE: 1.9803, R-squared: 0.5891). However, all models had R-squared values below 0.6, indicating limitations in capturing data nonlinearity and temporal dependencies. The study identified key factors influencing cucumber fruit set prediction through feature importance analysis. Future research should focus on collecting additional data, applying complex feature engineering, introducing time series analysis techniques, and considering data augmentation and normalization to improve model performance. This study contributes to the practical application of smart farm technology and the development of data-driven agricultural decision support systems.