• KSCE Journal of Civil and Environmental Engineering Research
• /
• v.30 no.4D
• /
• pp.351-360
• /
• 2010

In 2010, the number of registered vehicles reached almost at 17.48 millions in Korea. This dramatic increase of vehicles influenced to increase the number of traffic accidents which is one of the serious social problems and also to soar the personal and economic losses in Korea. Through this research, an enhanced intersection hazard prediction model by combining Genetic Algorithm and Artificial Neural Network will be developed in order to obtain the important data for developing the countermeasures of traffic accidents and eventually to reduce the traffic accidents in Korea. Firstly, this research has investigated the influencing factors of road geometric features on the traffic volume of each approaching for the intersections where traffic accidents and congestions frequently take place and, a linear regression model of traffic accidents and traffic conflicts were developed by examining the relationship between traffic accidents and traffic conflicts through the statistical significance tests. Secondly, this research also developed an intersection hazard prediction model by combining Genetic Algorithm and Artificial Neural Network through applying the intersection traffic volume, the road geometric features and the specific variables of traffic conflicts. Lastly, this research found out that the developed model is better than the existed forecasting models in terms of the reliability and accuracy by comparing the actual number of traffic accidents and the predicted number of accidents from the developed model. In conclusion, it is expect that the cost/effectiveness of any traffic safety improvement projects can be maximized if this developed intersection hazard prediction model by combining Genetic Algorithm and Artificial Neural Network use practically at field in the future.

• KSCE Journal of Civil and Environmental Engineering Research
• /
• v.30 no.6D
• /
• pp.587-597
• /
• 2010

When a car accident occurs, people who had an accident are not free from civil and criminal issues so that the accident investigator should reenact and analyze the accident situation accurately. In addition, the obtained documents through the analysis of such car accident occurrence and related factors have to be used to carry out the improvement of the areas that has numerous car accidents and complementary actions. The vehicle speed, accelerating force, braking power are currently known as the most affecting factors in accordance with many car accidents, traffic facilities, road design, etc. The vehicle's performance and rode friction coefficient road surface friction coefficient are affecting the most closely in this field. Especially, once the estimate of the speed of the accident moment relating to main eleven articles of Traffic Accident Exemption Law is very important and accuracy is required. However, currently the researches of these matters have not made exclusively yet in Korea. In this study by reflecting this current situation, until the sudden braking history is found from the car's sudden braking, it estimates accurately the transient brake time and rode friction coefficient by measuring a time of transient brake time through the precision speed detector (Vericom VC2000PC). The analysis of the experimental results calculated the transient brake time and friction coefficient to fit into the purpose of this study in the basis of different kind of various special purpose asphalt pavement and slip-prevention pavement and provided the fundamental data.

• Journal of the Korea Society of Computer and Information
• /
• v.28 no.10
• /
• pp.27-35
• /
• 2023

In this paper, we propose a process of increasing productivity by applying a deep learning-based defect detection and classification system to the prepreg fiber manufacturing process, which is in high demand in the field of producing composite materials. In order to apply it to toe prepreg manufacturing equipment that requires a solution due to the occurrence of a large amount of defects in various conditions, the optimal environment was first established by selecting cameras and lights necessary for defect detection and classification model production. In addition, data necessary for the production of multiple classification models were collected and labeled according to normal and defective conditions. The multi-classification model is made based on CNN and applies pre-learning models such as VGGNet, MobileNet, ResNet, etc. to compare performance and identify improvement directions with accuracy and loss graphs. Data augmentation and dropout techniques were applied to identify and improve overfitting problems as major problems. In order to evaluate the performance of the model, a performance evaluation was conducted using the confusion matrix as a performance indicator, and the performance of more than 99% was confirmed. In addition, it checks the classification results for images acquired in real time by applying them to the actual process to check whether the discrimination values are accurately derived.

• Geophysics and Geophysical Exploration
• /
• v.26 no.4
• /
• pp.171-184
• /
• 2023

Machine learning (ML)-based cavity detection using a large amount of survey data obtained from vehicle-mounted ground penetrating radar (GPR) has been actively studied to identify underground cavities. However, only simple image processing techniques have been used for preprocessing the ML input, and many conventional seismic and GPR data processing techniques, which have been used for decades, have not been fully exploited. In this study, based on the idea that a cavity can be identified using diffraction, we applied ML-based diffraction separation to GPR data to increase the accuracy of cavity detection using the YOLO v5 model. The original ML-based seismic diffraction separation technique was modified, and the separated diffraction image was used as the input to train the cavity detection model. The performance of the proposed method was verified using public GPR data released by the Seoul Metropolitan Government. Underground cavities and objects were more accurately detected using separated diffraction images. In the future, the proposed method can be useful in various fields in which GPR surveys are used.

• Korean Journal of Remote Sensing
• /
• v.39 no.6_1
• /
• pp.1497-1503
• /
• 2023

To address challenges in classifying clouds and snow cover when calculating ground reflectance in Near-UltraViolet (UV) wavelengths, this study introduces a methodology that combines cloud data from the Geostationary Environmental Monitoring Spectrometer (GEMS) and the Advanced Meteorological Imager (AMI)satellites for snow cover analysis. The proposed approach aims to enhance the quality of surface reflectance calculations, and combined cloud data were generated by integrating GEMS cloud data with AMI cloud detection data. When applied to compute GEMS surface reflectance, this fusion approach significantly mitigated underestimation issues compared to using only GEMS cloud data in snow-covered regions, resulting in an approximately 17% improvement across the entire observational area. The findings of this study highlight the potential to address persistent underestimation challenges in snow areas by employing fused cloud data, consequently enhancing the accuracy of other Level-2 products based on improved surface reflectivity.

• Journal of Korea Water Resources Association
• /
• v.56 no.11
• /
• pp.801-813
• /
• 2023

Recently, as studies about Artificial Neural Network (ANN) are actively progressing, studies for predicting water quality of rivers using ANN are being conducted. However, it is difficult to analyze the operation process inside ANN, because ANN is form of Black-box. Although eXplainable Artificial Intelligence (XAI) is used to analyze the computational process of ANN, research using XAI technology in the field of water resources is insufficient. This study analyzed Multi Layer Perceptron (MLP) to predict Water Temperature (WT), Dissolved Oxygen (DO), hydrogen ion concentration (pH) and Chlorophyll-a (Chl-a) at the Dasan water quality observatory in the Nakdong river using Layer-wise Relevance Propagation (LRP) among XAI technologies. The MLP that learned water quality was analyzed using LRP to select the optimal input data to predict water quality, and the prediction results of the MLP learned using the optimal input data were analyzed. As a result of selecting the optimal input data using LRP, the prediction accuracy of MLP, which learned the input data except daily precipitation in the surrounding area, was the highest. Looking at the analysis of MLP's DO prediction results, it was analyzed that the pH and DO a had large influence at the highest point, and the effect of WT was large at the lowest point.

• The Sea:JOURNAL OF THE KOREAN SOCIETY OF OCEANOGRAPHY
• /
• v.13 no.3
• /
• pp.280-285
• /
• 2008

Manganese nodule abundance estimated based on operation of a Free Fall Grab(FFG) needs to be corrected to make up for its incomplete recovery of nodule, because FFGs can not recover all the nodules distributed on seabed. The correction factor for nodule abundance was proposed as 1.29 and 1.13 in 1994 and 2002, respectively, mainly based on the analyses of seabed images. In this study we collected manganese nodules using both FFG and Box Corer(BC) at same stations to examine the accuracy of the previous correction factors. It was found that the nodule recovery of the BC was 1.4 times greater than that of the FFG at the same sampling station, suggesting the necessity of re-evaluation of the previously proposed correction factor for FFG. More extensive sampling and improvement of image analysis method are required to improve the precision of nodule abundance correction factor for FFG.

• Journal of Internet Computing and Services
• /
• v.25 no.1
• /
• pp.109-121
• /
• 2024

In various manufacturing processes such as textiles and automobiles, when equipment breaks down or stops, the machines do not work, which leads to time and financial losses for the company. Therefore, it is important to detect equipment abnormalities in advance so that equipment failures can be predicted and repaired before they occur. Most equipment failures are caused by bearing failures, which are essential parts of equipment, and detection bearing anomaly is the essence of PHM(Prognostics and Health Management) research. In this paper, we propose a preprocessing algorithm called SWT-SVD, which analyzes vibration signals from bearings and apply it to an anomaly transformer, one of the time series anomaly detection model networks, to implement bearing anomaly detection model. Vibration signals from the bearing manufacturing process contain noise due to the real-time generation of sensor values. To reduce noise in vibration signals, we use the Stationary Wavelet Transform to extract frequency components and perform preprocessing to extract meaningful features through the Singular Value Decomposition algorithm. For experimental validation of the proposed SWT-SVD preprocessing method in the bearing anomaly detection model, we utilize the PHM-2012-Challenge dataset provided by the IEEE PHM Conference. The experimental results demonstrate significant performance with an accuracy of 0.98 and an F1-Score of 0.97. Additionally, to substantiate performance improvement, we conduct a comparative analysis with previous studies, confirming that the proposed preprocessing method outperforms previous preprocessing methods in terms of performance.

• Journal of Internet Computing and Services
• /
• v.25 no.1
• /
• pp.57-67
• /
• 2024

This research investigates training methods for large language models to accurately identify sentiments and comprehend information about increasing and decreasing fluctuations in the financial domain. The main goal is to identify suitable datasets that enable these models to effectively understand expressions related to financial increases and decreases. For this purpose, we selected sentences from Wall Street Journal that included relevant financial terms and sentences generated by GPT-3.5-turbo-1106 for post-training. We assessed the impact of these datasets on language model performance using Financial PhraseBank, a benchmark dataset for financial sentiment analysis. Our findings demonstrate that post-training FinBERT, a model specialized in finance, outperformed the similarly post-trained BERT, a general domain model. Moreover, post-training with actual financial news proved to be more effective than using generated sentences, though in scenarios requiring higher generalization, models trained on generated sentences performed better. This suggests that aligning the model's domain with the domain of the area intended for improvement and choosing the right dataset are crucial for enhancing a language model's understanding and sentiment prediction accuracy. These results offer a methodology for optimizing language model performance in financial sentiment analysis tasks and suggest future research directions for more nuanced language understanding and sentiment analysis in finance. This research provides valuable insights not only for the financial sector but also for language model training across various domains.

• Convergence Security Journal
• /
• v.24 no.1
• /
• pp.59-68
• /
• 2024

With advancements in artificial intelligence technology, intelligent CCTV systems are being deployed across various environments, such as river bridges and construction sites. However, a conflict arises regarding the opening and closing of rooftop access points due to concerns over potential accidents and crime incidents and their role as emergency evacuation spaces. While the relevant law typically mandates the constant opening of designated rooftop access points, closures are often tacitly permitted in practice for security reasons, with a lack of appropriate legal measures. In this context, this study proposes a detection system utilizing intelligent CCTV to respond to emergencies that may occur on rooftops. We develop a system based on the YOLOv5 object detection model to detect assault and suicide attempts by jumping, introducing a new metric to assess them. Experimental results demonstrate that the proposed system rapidly detects assault and suicide attempts with high accuracy. Additionally, through a legal analysis of rooftop access point management, deficiencies in the legal framework regarding rooftop access and CCTV installation are identified, and improvement measures are proposed. With technological and legal improvements, we believe that crime and accident incidents in rooftop environments will decrease.