• Journal of Navigation and Port Research
• /
• v.46 no.4
• /
• pp.367-374
• /
• 2022

The purpose of this study was to propose a deep learning algorithm that applies to the fault diagnosis of fuel pumps and purifiers of autonomous ships. A deep learning algorithm reflecting the time dependence of the measured signal was configured, and the failure pattern was trained using the vibration signal, measured in the equipment's regular operation and failure state. Considering the sequential time-dependence of deterioration implied in the vibration signal, this study adopts Conv1D with sliding window computation for fault detection. The time dependence was also reflected, by transferring the measured signal from two-dimensional to three-dimensional. Additionally, the optimal values of the hyper-parameters of the Conv1D model were determined, using the grid search technique. Finally, the results show that the proposed data preprocessing method as well as the Conv1D model, can reflect the sequential dependency between the fault and its effect on the measured signal, and appropriately perform anomaly as well as failure detection, of the equipment chosen for application.

• The Journal of The Korea Institute of Intelligent Transport Systems
• /
• v.21 no.5
• /
• pp.171-182
• /
• 2022

The safe use of a structure requires it to be maintained in an undamaged state. Thus, a typical factor that determines the safety of a structure is a crack in it. In addition, cracks are caused by various reasons, damage the structure in various ways, and exist in different shapes. Making matters worse, if these cracks are unattended, the risk of structural failure increases and proceeds to a catastrophe. Hence, recently, methods of checking structural damage using deep learning and computer vision technology have been introduced. These methods usually have the premise that there should be a large amount of training image data. However, the amount of training image data is always insufficient. Particularly, this insufficiency negatively affects the performance of deep learning crack detection algorithms. Hence, in this study, a method of augmenting crack image data based on the image translation technique was developed. In particular, this method obtained the crack image data for training a deep learning neural network model by transforming a specific case of a asphalt crack image into a concrete crack image or vice versa . Eventually, this method expected that a robust crack detection algorithm could be developed by increasing the diversity of its training data.

• Journal of Aerospace System Engineering
• /
• v.17 no.4
• /
• pp.67-74
• /
• 2023

Recently, Urban Aircraft Mobility (UAM) has been attracting attention as a transportation system of the future, and small drones also play a role in various industries. The failure of various types of aviation systems can lead to crashes, which can result in significant property damage or loss of life. In the defense industry, where aviation systems are widely used, the failure of aviation systems can lead to mission failure. Therefore, this study proposes an anomaly detection model using deep learning technology to detect anomalies in aviation systems to improve the reliability of development and production, and prevent accidents during operation. As training and evaluating data sets, current data from aviation systems in an extremely low-temperature environment was utilized, and a deep learning network was implemented using the convolutional neural network, which is a deep learning technique that is commonly used for image recognition. In an extremely low-temperature environment, various types of failure occurred in the system's internal sensors and components, and singular points in current data were observed. As a result of training and evaluating the model using current data in the case of system failure and normal, it was confirmed that the abnormality was detected with a recall of 98 % or more.

• Journal of the Society of Disaster Information
• /
• v.19 no.3
• /
• pp.665-672
• /
• 2023

Purpose: Underground utility tunnel is facility that is jointly house infrastructure such as electricity, water and gas in city, causing condensation problems due to lack of airflow. This paper aims to prevent electricity leakage fires caused by condensation by detecting whether the control panel door in the underground utility tunnel is open using a deep learning model. Method: YOLO, a deep learning object recognition model, is trained to recognize the opening and closing of the control panel door using video data taken by a robot patrolling the underground utility tunnel. To improve the recognition rate, image augmentation is used. Result: Among the image enhancement techniques, we compared the performance of the YOLO model trained using mosaic with that of the YOLO model without mosaic, and found that the mosaic technique performed better. The mAP for all classes were 0.994, which is high evaluation result. Conclusion: It was able to detect the control panel even when there were lights off or other objects in the underground cavity. This allows you to effectively manage the underground utility tunnel and prevent disasters.

• Analytical Science and Technology
• /
• v.19 no.5
• /
• pp.400-406
• /
• 2006

Hydrogen analysis by elastic recoil detection has been performed utilizing polyimide film as a reference sample of known hydrogen content assuming the soundness of ion beam current integration. However beam current integration at higher incidence angle is not reliable. Scattering yield per unit fluence by current integration which is normalized per unit path length decreases as the sample tilt angle is getting higher. Moreover because beam current integration at high tilt angle is incomplete, hydrogen evaluation is very risky by direct comparison of sequentially collected recoil spectra between reference and target sample. In this study, primary ion beam dose is determined by backscattering spectrum that is collected simultaneously with recoil spectrum instead of ion beam current integration in order to reduce uncertainty arising in the process of current integration and to enhance the reliability of quantitative analysis. Three test samples are selected $-7.6{\mu}m$ polyimide film, hydrogen implanted silicondioxide and Au deposited carbon wafer- and analyzed by two methods and compared.

• Journal of the Computational Structural Engineering Institute of Korea
• /
• v.37 no.3
• /
• pp.173-178
• /
• 2024

This study presents a technique for assessing the dimensional quality of assembly parts in Prefabricated Steel Structures (PSS) using a stereo vision sensor. The stereo vision system captures images and point cloud data of the assembly area, followed by applying image processing algorithms such as fuzzy-based edge detection and Hough transform-based circular bolt hole detection to identify bolt hole locations. The 3D center positions of each bolt hole are determined by correlating 3D real-world position information from depth images with the extracted bolt hole positions. Principal Component Analysis (PCA) is then employed to calculate coordinate axes for precise measurement of distances between bolt holes, even when the sensor and structure orientations differ. Bolt holes are sorted based on their 2D positions, and the distances between sorted bolt holes are calculated to assess the assembly part's dimensional quality. Comparison with actual drawing data confirms measurement accuracy with an absolute error of 1mm and a relative error within 4% based on median criteria.

• Clean Technology
• /
• v.30 no.3
• /
• pp.175-187
• /
• 2024

Seaweeds are widely distributed along national coastlines around the world, and the biomass derived from them is an important marine biological organism. Seaweed is a crucial component of a healthy marine ecosystem. However, changes in marine environments have led to the occurrence of urchin barrens, and the damage caused by this phenomenon is steadily increasing. As a result, investigations into the distribution and spread of urchin barrens in the coastal areas of South Korea are being conducted regularly so efficient detection technologies are essential. One of the technologies that can swiftly and accurately analyze extensive areas is detection technology based on hyperspectral image information systems. This study aims to present the latest hyperspectral imaging technology for investigating the current status of urchin barrens and the methods for classifying this technology, including principles, preprocessing techniques, and correction methods. This study also proposes a classification technique for urchin barrens along the coast of Jeju Island that uses hyperspectral images and categorizes the urchin barrens into initial, intermediate, and advanced stages. The results showed that approximately 17.5% of the experimental areas were in the advanced stage. Based on this, various management and restoration methods tailored to different categories of urchin barren can be proposed.

• Journal of Sensor Science and Technology
• /
• v.33 no.5
• /
• pp.310-317
• /
• 2024

Formaldehyde (HCHO) is a major primary indoor air pollutant with various adverse effects on the human body, includingsuch as sick building syndrome, lung cancer, and nasal cancer. Therefore, gas sensors for effective HCHO detection detecting HCHO are crucial for maintaining a healthy indoor environments, and research is being conducted to develop high-performance sensors for this purpose. AnOne of the effective methods for enhancing the to enhance sensing properties is involves modifying the p-n heterojunction structure, which improves sensing through via electronic sensitization based on the expanded depletion region and chemical sensitization that dissociates specific gases. In this studyHerein, weWe fabricated NiO-decorated In₂O₃ NRs using an e-beam evaporator based on the glancing angle deposition technique by optimizing the NiO thickness (0, 1, 2, and 3 nm). When exposed to 50 ppm HCHO, NiO-decorated In₂O₃ NRs showed a 3.91%-fold enhancement in the gas response (Ra/Rg-1= 23.9) and a 41.47% faster response time (40.7 s) than-compared to bare In₂O₃ NRs with an extremely low theoretical detection limit of ≈approximately 9.3 ppb.

• Parasites, Hosts and Diseases
• /
• v.34 no.3
• /
• pp.177-184
• /
• 1996

In situ hybridization was performed to detect rat heumocwstis ca4nii in the lung sections. Rats were immunosuppressed by weekly subcutaneous injection of 10 mg/kg methylprednisolone. On the 6th, 8th and 9th week of immunosuppression, the lungs were removed and fled in 10% neutral formalin. A 22 base oligonucleotide probe complementary to p. carinii 5S ribosomal RMh was commercially synhesized and its 3' terminal was labeled wiH biotin. In situ hybridization was performed utilizing manual capillary action technolog)r on the Microprobe system. p. cnrinii were detected along the luminal surface of alveolar pneumocytes, in exudate of alveolar cavities, and also in secretory material of bronchioles. In the 6th week group, positive reaction was observed focally in the peripheral region of the lung sections, but the reaction was observed diffusely in the 8th or 9th week groups. In comparison with Grocott's methenamine silver stain, in situ hybridization technique can detect the organism rapidly, and can detect trophic forms very well. Furthermore, no nonspecific reaction with other pathogenic fungi and protozoa was recognized. Therefore, in situ hybridization can be a good technique to detect p. carinii in the lungs of infected rats.

• Journal of Korean Society of Transportation
• /
• v.18 no.1
• /
• pp.47-59
• /
• 2000

The purpose of this study is to develop travel-time estimation model using neural networks and prediction model using neural networks and kalman-filtering technique. The data used in this study are travel speed collected from inductive loop vehicle detection systems(VDS) and travel time collected from the toll collection system (TCS) between Seoul and Osan toll Plaza on the Seoul-Pusan Expressway. Two models, one for travel-time estimation and the other for travel-time Prediction were developed. Application cases of each model were divided into two cases, so-called, a single-region and a multiple-region. because of the different characteristics of travel behavior shown on each region. For the evaluation of the travel time estimation and Prediction models, two Parameters. i.e. mode and mean were compared using five-minute interval data sets. The test results show that mode was superior to mean in representing the relationship between speed and travel time. It is, however shown that mean value gives better results in case of insufficient data. It should be noted that the estimation and the Prediction of travel times based on the VDS data have been improved by using neural networks, because the waiting time at exit toll gates can be included for the estimation of travel time based on the VDS data by considering differences between VDS and TCS travel time Patterns in the models. In conclusion, the results show that the developed models decrease estimation and prediction errors. As a result of comparing the developed model with the existing model using the observed data, the equality coefficients of the developed model was average 88% and the existing model was average 68%. Thus, the developed model was improved minimum 17% and maximum 23% rather then existing model .