• Journal of the Korea Institute of Information Security & Cryptology
• /
• v.30 no.6
• /
• pp.1053-1065
• /
• 2020

With the recent development of hardware performance and artificial intelligence technology, sophisticated fake videos that are difficult to distinguish with the human's eye are increasing. Face synthesis technology using artificial intelligence is called Deepfake, and anyone with a little programming skill and deep learning knowledge can produce sophisticated fake videos using Deepfake. A number of indiscriminate fake videos has been increased significantly, which may lead to problems such as privacy violations, fake news and fraud. Therefore, it is necessary to detect fake video clips that cannot be discriminated by a human eyes. Thus, in this paper, we propose a deep-fake detection model applied with Bidirectional Convolution LSTM and Attention Module. Unlike LSTM, which considers only the forward sequential procedure, the model proposed in this paper uses the reverse order procedure. The Attention Module is used with a Convolutional neural network model to use the characteristics of each frame for extraction. Experiments have shown that the model proposed has 93.5% accuracy and AUC is up to 50% higher than the results of pre-existing studies.

• Journal of Trauma and Injury
• /
• v.35 no.3
• /
• pp.181-188
• /
• 2022

Purpose: Inferior vena cava (IVC) collapse is related to hypovolemia. Sonography has been used to measure the IVC diameter, but there is variation depending on the skill of the operator and it is difficult to obtain accurate measurements in patients who have a large amount of intestinal gas or are obese. As a modality to obtain accurate measurements, we measured the diameters of the IVC and aorta on trauma computed tomography scans and investigated the correlation between the IVC to aorta ratio and the shock index in blunt trauma patients. Methods: We retrospectively analyzed the medical records of 588 trauma patients who were transferred to the regional trauma center (level 1) of Wonkang University Hospital from March 2020 to February 2021. We included trauma patients 18 years or older who met the trauma activation criteria and underwent trauma computed tomography scans with intravenous contrast within 40 minutes of admission. The shock index was calculated from vital signs before trauma computed tomography scan, and measurements of the anteroposterior diameter of the IVC (AP), the transverse diameter of the IVC (T), and aorta were made 10 mm above the right renal vein in the venous phase. Results: Overall, 271 patients were included in this study, of whom 150 had a shock index ≤0.7 and 121 had a shock index >0.7. The T to AP ratio and AP to aorta ratio were significantly different between groups. Cutoffs were identified for the T to AP ratio and AP to aorta ratio (2.37 and 0.62, respectively) that produced clinically useful sensitivity and specificity for predicting a shock index >0.7, demonstrating moderate accuracy (T to AP ratio: area under the curve, 0.71; sensitivity, 59%; specificity, 87% and AP to aorta ratio: area under the curve, 0.70; sensitivity, 55%; specificity, 91%). Conclusions: The T to AP ratio and AP to aorta ratio are useful for predicting hemorrhagic shock in trauma patients.

• Proceedings of the Korea Water Resources Association Conference
• /
• 2023.05a
• /
• pp.246-246
• /
• 2023

Multi-purpose dams are operated accounting for both physical and socioeconomic factors. This study aims to evaluate the utility of a deep learning algorithm-based model for three multi-purpose dam operation (Seomjin River dam, Juam dam, and Juam Control dam) in Seomjin River. In this study, the Gated Recurrent Unit (GRU) algorithm is applied to predict hourly water level of the dam reservoirs over 2002-2021. The hyper-parameters are optimized by the Bayesian optimization algorithm to enhance the prediction skill of the GRU model. The GRU models are set by the following cases: single dam input - single dam output (S-S), multi-dam input - single dam output (M-S), and multi-dam input - multi-dam output (M-M). Results show that the S-S cases with the local dam information have the highest accuracy above 0.8 of NSE. Results from the M-S and M-M model cases confirm that upstream dam information can bring important information for downstream dam operation prediction. The S-S models are simulated with altered outflows (-40% to +40%) to generate the simulated water level of the dam reservoir as alternative dam operational scenarios. The alternative S-S model simulations show physically inconsistent results, indicating that our deep learning algorithm-based model is not explainable for multi-purpose dam operation patterns. To better understand this limitation, we further analyze the relationship between observed water level and outflow of each dam. Results show that complexity in outflow-water level relationship causes the limited predictability of the GRU algorithm-based model. This study highlights the importance of socioeconomic factors from hidden multi-purpose dam operation processes on not only physical processes-based modeling but also aritificial intelligence modeling.

• Journal of Internet Computing and Services
• /
• v.25 no.1
• /
• pp.17-27
• /
• 2024

Overfishing, climate change, and competitive fishing have led to a continuous decline in fishery production. To address these issues, the Total Allowable Catch (TAC) system has been established, which sets annual catch quotas for individual fish species and allows fishing only within those limits. As part of the TAC system, land-based investigators measure the length and height of fish species at auction markets to calculate the weight and TAC depletion. However, the accuracy of the acquired data varies depending on the skill level of the land-based investigators, and the labor-intensive nature of the work makes it unsustainable. To address these issues, this paper proposes a fish species recognition and length measurement system that automatically measures the length, height, and weight of eight TAC-managed fish species using the camera of a smart pad that can measure the distance to the water surface. This system can help to automate the current labor-intensive work, minimize data loss, and facilitate the establishment of the TAC system.

• The Journal of the Convergence on Culture Technology
• /
• v.10 no.3
• /
• pp.801-807
• /
• 2024

Linear motion robots are devices that perform functions such as transferring parts or positioning devices, and require high precision. In companies that develop linear robot application systems, human workers are in charge of quality control and fault diagnosis of linear robots, and the result and accuracy of a fault diagnosis varies depending on the skill level of the person in charge. Recently, there have been many attempts to utilize artificial intelligence to diagnose faults in industrial devices. In this paper, we present a system that automatically diagnoses linear rail and ball screw misalignment of a linear robot using transfer learning. In industrial systems, it is difficult to obtain a lot of learning data, and this causes a data imbalance problem. In this case, a transfer learning model configured by retraining an established model is widely used. The information obtained by using an acceleration sensor and torque sensor was used, and its usefulness was evaluated for each case. After converting the signal obtained from the sensor into a spectrogram image, the type of abnormality was diagnosed using an image recognition artificial intelligence classifier. It is expected that the proposed method can be used not only for linear robots but also for diagnosing other industrial robots.

• Korean Journal of Clinical Laboratory Science
• /
• v.56 no.3
• /
• pp.277-287
• /
• 2024

Transcranial Doppler (TCD) ultrasound is a crucial non-invasive tool for assessing cerebral blood flow and is widely used to diagnose and monitor cerebrovascular diseases. This paper reaffirms the importance of TCD, details examination methods and precautions, and provides a guide for practitioners. TCD evaluates the blood flow velocity to assess stenosis, occlusion, and hemodynamic changes. Distinguishing between increased blood flow volume and decreased vessel diameter based solely on velocity is challenging, necessitating a comprehensive approach to integrating clinical findings and hemodynamic changes. The reliability of TCD results depends on the skill of the examiner and requires standardized procedures and continuous training. Advances in automation and artificial intelligence promise enhanced accuracy and reliability. Future research should focus on validating and clinically applying these technologies. This paper is a review of the clinical significance of TCD, methods, and precautions, offering a valuable guide for practitioners and highlighting the potential benefits of ongoing advancements in TCD for the diagnosis and treatment of cerebrovascular diseases.

• Journal of Korean Society of Coastal and Ocean Engineers
• /
• v.30 no.5
• /
• pp.223-233
• /
• 2018

The predictability of winter storm waves using KMA's operational wind forecasts has been studied to predict wind waves and swells in the East coast of Korea using SWAN. The nested model were employed along the East coast of Korea to simulate the wave transformation in the coastal area and wave dissipation term of whitecapping is optimized to improve swell prediction accuracy. In this study, KMA's operational meteorological models (RDAPS and LDAPS) are used as input wind fields. In order to evaluate model accuracy, we also simulate wind waves and swells using ECMWF reanalysis and KIOST WRF wind and they are compared with the KMA's operational wave model and the wave measurement data on the offshore and onshore stations. As a result, it has the lowest RMSE and the highest correlation coefficient in the onshore when the input wind fields are KMA's operational meteorological forecasts. In the offshore, all of the simulate results shows good agreements with similar error statistics. It means that it is very feasible to use SWAN model with the modified whitecapping factor and KMA's operational meteorological forecasts for predicting the wind waves and swells in the East coast of Korea.

• Proceedings of the KSR Conference
• /
• 2007.11a
• /
• pp.1099-1104
• /
• 2007

The visual implementation of 3-dimensional national environment is focused by the requirement and importance in the fields such as, national development plan, telecommunication facility deployment plan, railway construction, construction engineering, spatial city development, safety and disaster prevention engineering. The currently used DEM system using contour lines, which embodies national geographic information based on the 2-D digital maps and facility information has limitation in implementation in reproducing the 3-D spatial city. Moreover, this method often neglects the altitude of the rail way infrastructure which has narrow width and long length. There it is needed to apply laser measurement technique in the spatial target object to obtain accuracy. Currently, the LiDAR data which combines the laser measurement skill and GPS has been introduced to obtain high resolution accuracy in the altitude measurement. In this paper, we first investigate the LiDAR based researches in advanced foreign countries, then we propose data a generation scheme and an algorithm for the optimal manage and synthesis of railway facility system in our 3-D spatial terrain information. For this object, LiDAR based height data transformed to DEM, and the realtime unification of the vector via digital image mapping and raster via exactness evaluation is transformed to make it possible to trace the model of generated 3-dimensional railway model with long distance for 3D tract model generation.

• Phonetics and Speech Sciences
• /
• v.12 no.1
• /
• pp.75-83
• /
• 2020

The purpose of this study was to examine how a singing program using self-voice monitoring for children with cochlear implants (CI) influences on the intonation and the accuracy of pitch production. To verify and estimate the effectiveness, a program was conducted with participants of 7 prelingual CI users, whose aged between 4 years and 7 years. The program adopted three stages from the self-voice monitoring: Listen, Explore, and Reproduce (LER stage). All participants received 8 singing sessions over 8 weeks, including pre-test, intervention, and post-test. For the pre and post-test, participants' singing of an excerpt of a song "happy birthday" and speaking three assertive sentences and three interrogative sentences were recorded and analyzed in terms of the intonation slopes at the end of the sentences and the melodic contour. From the sentence speeches, we found that the intonation slopes of the interrogative sentences significantly improved as they showed similar patterns with that of the average normal hearing group. Also, in regard to singing, we observed that the melody contour had progressed, as well as the range of pitch production had extended. The positive result from the intervention indicates that the singing program was effective for children with CI to develop the intonation skill and accuracy of pitch production.

• Proceedings of the KSR Conference
• /
• 2008.11b
• /
• pp.30-36
• /
• 2008

The Spatial Image contents of Geomorphology 3-D environment is focused by the requirement and importance in the fields such as, national land development plan, telecommunication facility management, railway construction, general construction engineering, Ubiquitous city development, safety and disaster prevention engineering. The currently used DEM system using contour lines, which embodies geographic information based on the 2-D digital maps and facility information has limitation in implementation in reproducing the 3-D spatial city. Moreover, this method often neglects the altitude of the rail way infrastructure which has narrow width and long length. There it is needed to apply laser measurement technique in the spatial target object to obtain accuracy. Currently, the LiDAR data which combines the laser measurement skill and GPS has been introduced to obtain high resolution accuracy in the altitude measurement. In this paper, we tested of the railway facilities using laser surveying system, then we propose data a generation of spatial images for the optimal manage and synthesis of railway facility system in our 3-D spatial terrain information. For this object, LiDAR based height data transformed to DEM, and the realtime unification of the vector via digital image mapping and raster via exactness evaluation is transformed to make it possible to trace the model of generated 3-dimensional railway model with long distance for 3D tract model generation. As the results, We confirmed the solutions of varieties application for railway facilities management using 3-D spatial image contents.