• KIPS Transactions on Software and Data Engineering
• /
• v.8 no.3
• /
• pp.115-122
• /
• 2019

For Korean phoneme recognition, Hidden Markov-Gaussian Mixture model(HMM-GMM) or hybrid models which combine artificial neural network with HMM have been mainly used. However, current approach has limitations in that such models require force-aligned corpus training data that is manually annotated by experts. Recently, researchers used neural network based phoneme recognition model which combines recurrent neural network(RNN)-based structure with connectionist temporal classification(CTC) algorithm to overcome the problem of obtaining manually annotated training data. Yet, in terms of implementation, these RNN-based models have another difficulty in that the amount of data gets larger as the structure gets more sophisticated. This problem of large data size is particularly problematic in the Korean language, which lacks refined corpora. In this study, we introduce CTC algorithm that does not require force-alignment to create a Korean phoneme recognition model. Specifically, the phoneme recognition model is based on convolutional neural network(CNN) which requires relatively small amount of data and can be trained faster when compared to RNN based models. We present the results from two different experiments and a resulting best performing phoneme recognition model which distinguishes 49 Korean phonemes. The best performing phoneme recognition model combines CNN with 3hop Bidirectional LSTM with the final Phoneme Error Rate(PER) at 3.26. The PER is a considerable improvement compared to existing Korean phoneme recognition models that report PER ranging from 10 to 12.

• KSCE Journal of Civil and Environmental Engineering Research
• /
• v.29 no.1D
• /
• pp.135-143
• /
• 2009

Integrating the Global Positioning System (GPS) and Inertial Navigation System (INS) sensor technologies using the precise GPS Carrier phase measurements is a methodology that has been widely applied in those application fields requiring accurate and reliable positioning and attitude determination; ranging from 'kinematic geodesy', to mobile mapping and imaging, to precise navigation. However, such integrated system may not fulfil the demanding performance requirements when the baseline length between reference and mobil user GPS receiver is grater than a few tens of kilometers. This is because their positioning/attitude determination is still very dependent on the errors of the GPS observations, so-called "baseline dependent errors". This limitation can be remedied by the integration of GPS and INS sensors, using multiple reference stations. Hence, in order to derive the GPS distance dependent errors, this research proposes measurement processing algorithms for multiple reference stations, such as a reference station ambiguity resolution procedure using linear combination techniques, a error estimation based on Kalman filter and a error interpolation. In addition, all the algorithms are evaluated by processing real observations and results are summarized in this paper.

• KSCE Journal of Civil and Environmental Engineering Research
• /
• v.28 no.6D
• /
• pp.931-936
• /
• 2008

Surveying by using terrestrial LiDAR(Light Detection And Ranging) is more rapid than by using total station which enables tunnel section profile surveying to be done in suitable time and minimize centerline error, occurrence of overcut and undercut. Therefore, utilization of terrestrial LiDAR has increased more and more in section profile survey and measurement field Moreover, studies of terrestrial LiDAR for accurate and efficient utilization is now ongoing vigorously. Average end area formula, which was generally used to calculate overcut and undercut, was compared with existing methods such as total station survey and photogrammetry. However, there are no criteria of spacing distance for calculating overcut and undercut through terrestrial LiDAR surveying which can acquire 3D information of whole tunnel. This research performed reverse engineering to decide optimal spacing distance when surveying tunnel section profile by comparing whole tunnel volume and tunnel volume in difference spacing distance. This result was utilized to produce CAD drawing for the test tunnel site where there is no design drawings. In addition to this, efficiency of LiDAR and accuracy of CAD drawing was compared with targetless total station surveying of tunnel section profile. Finally, error analysis of target coordinate's accuracy and incidence angle was done in order to verify the accuracy of terrestrial LiDAR technology.

• KSCE Journal of Civil and Environmental Engineering Research
• /
• v.26 no.3B
• /
• pp.279-289
• /
• 2006

The reconstruction of low dimension nonlinear behavior from the hydrologic time series has been an active area of research in the last decade. In this study, we present the applications of a powerful state space reconstruction methodology using the method of Support Vector Machines (SVM) to the Great Salt Lake (GSL) volume. SVMs are machine learning systems that use a hypothesis space of linear functions in a Kernel induced higher dimensional feature space. SVMs are optimized by minimizing a bound on a generalized error (risk) measure, rather than just the mean square error over a training set. The utility of this SVM regression approach is demonstrated through applications to the short term forecasts of the biweekly GSL volume. The SVM based reconstruction is used to develop time series forecasts for multiple lead times ranging from the period of two weeks to several months. The reliability of the algorithm in learning and forecasting the dynamics is tested using split sample sensitivity analyses, with a particular interest in forecasting extreme states. Unlike previously reported methodologies, SVMs are able to extract the dynamics using only a few past observed data points (Support Vectors, SV) out of the training examples. Considering statistical measures, the prediction model based on SVM demonstrated encouraging and promising results in a short-term prediction. Thus, the SVM method presented in this study suggests a competitive methodology for the forecast of hydrologic time series.

• Tuberculosis and Respiratory Diseases
• /
• v.48 no.2
• /
• pp.180-190
• /
• 2000

Background: Genomic instability, which is manifested by the replication error(RER) phenotype, has been proposed for the promotion of genetic alterations necessary for carcinogenesis. Merlo et al. reported frequent microsatellite instability in primary small cell lung cancers. However, Kim et al. found that instability occurred in only 1% of the loci tested and did not resemble the replication error-positive phenotype. The significance of microsatellite instability in the tumorigenesis of small cell lung cancer as well as the relationship between microsatellite instability and its clinical prognosis was investigated in our study. Methods: Fifteen primary small cell lung cancers were chosen for this study. The DNAs extracted from paraffin-embedded tissue blocks with primary tumor and corresponding control tissue were investigated. Forty microsatellite markers on chromosome 1p, 2p, 3p, 5q, 6p, 6q, 9p, 9q, 13q, and 17p were used in the microsatellite analysis. Results: Thirteen(86.7%) of 15 tumors exhibited LOH in at least one of the tested microsatellite markers. Three of 13 tumors exhibiting LOH lost a larger area in chromosome 9p. LOH was shown in 72.7% on chromosome 2p, 40% on 3p, 50% on 5q, 46.7% on 9p, 69.2% on 13q, and 66.7% on 17p(Table 1). Nine(60%) of 15 tumors exhibited shifted bands in at least one of the tested microsatellite markers. Nine cases exhibiting shifted bands showed altered loci ranging 2.5~52.5%(mean $9.4%\pm16.19$)(Table 2). Shifted bands occurred in 5.7% (34 of 600) of the loci tested(Table 2). Nine cases with shifted bands exhibited LOH ranging between 0~83.3%, and the median survival duration of those cases was 35 weeks. Six cases without shifted bands exhibited LOH ranging between 0~83.3%, and the median survival duration of those cases was 73 weeks. There was no significant difference between median survival durations of the two groups(p=0.4712). Conclusion: Microsatellite instability as well as the inactivation of several tumor suppressor genes may play important roles in the development and progression process of tumors. However, the relationship between microsatellite instability and its clinical prognosis in primary small cell lung cancer could not be established.

• Korean Journal of Remote Sensing
• /
• v.39 no.5_3
• /
• pp.1009-1029
• /
• 2023

Urban trees play a vital role in urban ecosystems,significantly reducing impervious surfaces and impacting carbon cycling within the city. Although previous research has demonstrated the efficacy of employing artificial intelligence in conjunction with airborne light detection and ranging (LiDAR) data to generate urban tree information, the availability and cost constraints associated with LiDAR data pose limitations. Consequently, this study employed freely accessible, high-resolution multispectral satellite imagery (i.e., Sentinel-2 data) to estimate fractional tree canopy cover (FTC) within the urban confines of Suwon, South Korea, employing machine learning techniques. This study leveraged a median composite image derived from a time series of Sentinel-2 images. In order to account for the diverse land cover found in urban areas, the model incorporated three types of input variables: average (mean) and standard deviation (std) values within a 30-meter grid from 10 m resolution of optical indices from Sentinel-2, and fractional coverage for distinct land cover classes within 30 m grids from the existing level 3 land cover map. Four schemes with different combinations of input variables were compared. Notably, when all three factors (i.e., mean, std, and fractional cover) were used to consider the variation of landcover in urban areas(Scheme 4, S4), the machine learning model exhibited improved performance compared to using only the mean of optical indices (Scheme 1). Of the various models proposed, the random forest (RF) model with S4 demonstrated the most remarkable performance, achieving R² of 0.8196, and mean absolute error (MAE) of 0.0749, and a root mean squared error (RMSE) of 0.1022. The std variable exhibited the highest impact on model outputs within the heterogeneous land covers based on the variable importance analysis. This trained RF model with S4 was then applied to the entire Suwon region, consistently delivering robust results with an R² of 0.8702, MAE of 0.0873, and RMSE of 0.1335. The FTC estimation method developed in this study is expected to offer advantages for application in various regions, providing fundamental data for a better understanding of carbon dynamics in urban ecosystems in the future.

• KSCE Journal of Civil and Environmental Engineering Research
• /
• v.44 no.5
• /
• pp.637-648
• /
• 2024

The frequency of extreme weather events such as heavy and extreme rainfall has been increasing due to global climate change. Accordingly, it is essential to quantify hydrometeorological variables for efficient water resource management. Among the various hydro-meteorological variables, Land Surface Temperature (LST) and Evapotranspiration (ET) play key roles in understanding the interaction between the surface and the atmosphere. In Korea, LST and ET are mainly observed through ground-based stations, which also have limitation in obtaining data from ungauged watersheds, and thus, it hinders to estimate spatial behavior of LST and ET. Alternatively, remote sensing-based methods have been used to overcome the limitation of ground-based stations. In this study, we evaluated the applicability of the National Aeronautics and Space Administration's (NASA) ECOsystem Spaceborne Thermal Radiometer Experiment on Space Station (ECOSTRESS) LST and ET data estimated across Korea (from July 1, 2018 to December 31, 2022). For validation, we utilized NASA's MODerate Resolution Imaging Spectroradiometer (MODIS) data and eddy covariance flux tower observations managed by agencies under the Ministry of Environment of South Korea. Overall, results indicated that ECOSTRESS-based LSTs showed similar temporal trends (R: 0.47~0.73) to MODIS and ground-based observations. The index of agreement also showed a good agreement of ECOSTRESS-based LST with reference datasets (ranging from 0.82 to 0.91), although it also revealed distinctive uncertainties depending on the season. The ECOSTRESS-based ET demonstrated the capability to capture the temporal trends observed in MODIS and ground-based ET data, but higher Mean Absolute Error and Root Mean Square Error were also exhibited. This is likely due to the low acquisition rate of the ECOSTRESS data and environmental factors such as cooling effect of evapotranspiration, overestimation during the morning. This study suggests conducting additional validation of ECOSTRESS-based LST and ET, particularly in topographical and hydrological aspects. Such validation efforts could enhance the practical application of ECOSTRESS for estimating basin-scale LST and ET in Korea.

• Korean Journal of Remote Sensing
• /
• v.38 no.6_3
• /
• pp.1827-1836
• /
• 2022

Disasters that occur unexpectedly are difficult to predict. In addition, the scale and damage are increasing compared to the past. Sometimes one disaster can develop into another disaster. Among the four stages of disaster management, search and rescue are carried out in the response stage when an emergency occurs. Therefore, personnel such as firefighters who are put into the scene are put in at a lot of risk. In this respect, in the initial response process at the disaster site, robots are a technology with high potential to reduce damage to human life and property. In addition, Light Detection And Ranging (LiDAR) can acquire a relatively wide range of 3D information using a laser. Due to its high accuracy and precision, it is a very useful sensor when considering the characteristics of a disaster site. Therefore, in this study, development and experiments were conducted so that the robot could perform real-time monitoring at the disaster site. Multi-sensor module was developed by combining LiDAR, Inertial Measurement Unit (IMU) sensor, and computing board. Then, this module was mounted on the robot, and a customized Simultaneous Localization and Mapping (SLAM) algorithm was developed. A method for stably mounting a multi-sensor module to a robot to maintain optimal accuracy at disaster sites was studied. And to check the performance of the module, SLAM was tested inside the disaster building, and various SLAM algorithms and distance comparisons were performed. As a result, PackSLAM developed in this study showed lower error compared to other algorithms, showing the possibility of application in disaster sites. In the future, in order to further enhance usability at disaster sites, various experiments will be conducted by establishing a rough terrain environment with many obstacles.

• Journal of the Korean Association of Geographic Information Studies
• /
• v.14 no.4
• /
• pp.137-149
• /
• 2011

In this study, we analyzed the effects of seismic displacements due to the mega thrust earthquake occurred near Tohoku-Oki area on Mar. 11, 2011 with Mw 9.0 magnitude in the context of evaluation of position change by the earthquake on the Korean and Asian GPS permanent stations. For this, two weeks GPS data observed around the event of Tohoku-Oki earthquake (Mar. 4 ~ Mar. 18, 2011) were obtained from 22 GPS permanent stations in the vicinity of epicenter (Korea, Japan, Russia, China and Taiwan) and 284 IGS global stations. All available GPS data were processed and adjusted by GAMIT/GLOBK software to estimate the co-seismic horizontal displacements at each station. As the results of GPS analysis, the co-seismic displacements due to Tohoku-Oki earthquake were clearly revealed in the GPS stations of Asian region, Japan and its neighboring countries, and even to affect the horizontal position of GPS station (WUHN in China) which are located about 2,702km away from the epicenter. In conclusion, it was found that the Tohoku-Oki earthquake had resulted in the horizontal displacements ranging from 14.9 mm to 58.3 mm in Korea. So, these displacements can cause the position error of GPS geodetic survey up to 20 mm without updating the coordinates of Korean geodetic network.

• Journal of IKEEE
• /
• v.15 no.1
• /
• pp.76-85
• /
• 2011

In this paper, we analysed RSS characteristics by external factors and presented an efficient algorithm for real-time location tracking and its hardware system. The proposed algorithm enhanced the ranging accuracy using Kalman Filter based on the RSS DB. The location tracking system that consists of the tag, AP(Access Point), a data collector(Data Receiver) with IEEE 802.15.4(ZigBee) network environment, and location tracking application that reveal locations of each tag is implemented for the test environment. The location tracking system presented in this paper is implemented with MSP430 microprocessor manufactured by TI(Texas Instrument), CC2420 RF chipset and the location tracking application. With the results of the experiment, the proposed algorithm and the system can achieve the efficiency and the accuracy of location tracking with the average error of 19.12cm, and its standard deviation of 5.31cm in outdoor circumstance. Also, the experimental result shows that exact tracking of position in indoor circumstance cannot achieve because of vulnerable RSS with external circumstance.