• KIPS Transactions on Software and Data Engineering
• /
• v.5 no.5
• /
• pp.251-260
• /
• 2016

Face recognition is a technology to extract feature from a facial image, learn the features through various algorithms, and recognize a person by comparing the learned data with feature of a new facial image. Especially, in order to improve the rate of face recognition, face recognition requires various processing methods. In the training stage of face recognition, feature should be extracted from a facial image. As for the existing method of extracting facial feature, linear discriminant analysis (LDA) is being mainly used. The LDA method is to express a facial image with dots on the high-dimensional space, and extract facial feature to distinguish a person by analyzing the class information and the distribution of dots. As the position of a dot is determined by pixel values of a facial image on the high-dimensional space, if unnecessary areas or frequently changing areas are included on a facial image, incorrect facial feature could be extracted by LDA. Especially, if a camera image is used for face recognition, the size of a face could vary with the distance between the face and the camera, deteriorating the rate of face recognition. Thus, in order to solve this problem, this paper detected a facial area by using a camera, removed unnecessary areas using the facial feature area calculated via a Gabor filter, and normalized the size of the facial area. Facial feature were extracted through LDA using the normalized facial image and were learned through the artificial neural network for face recognition. As a result, it was possible to improve the rate of face recognition by approx. 13% compared to the existing face recognition method including unnecessary areas.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.21 no.6
• /
• pp.218-226
• /
• 2020

An experimental study was conducted on baseline model updating and damage estimation techniques for the health monitoring of offshore wind turbine tripod substructures. First, a procedure for substructure health monitoring was proposed. An initial baseline model for a scaled model of a tripod substructure was established. A baseline model was updated based on the natural frequencies and the mode shapes measured in the healthy state. A training pattern was then generated using the updated baseline model, and the damage was estimated by inputting the modal parameters measured in the damaged state into the trained neural network. The baseline model could be updated reasonably using the effective fixity model. The damage tests were performed, and the damage locations could be estimated reasonably. In addition, the estimated damage severity also increased as the actual damage severity increased. On the other hand, when the damage severity was relatively small, the corresponding damage location was detected, but it was more difficult to identify than the other cases. Further studies on small damage estimation and stiffness reduction quantification will be needed before the presented method can be used effectively for the health monitoring of tripod substructures.

• Journal of the Korea Society of Computer and Information
• /
• v.25 no.9
• /
• pp.37-44
• /
• 2020

AI technology has developed in the form of decision support technology in law, patent, finance and national defense and is applied to disease diagnosis and legal judgment. To search real-time information with Deep Learning, Big data Analysis and Deep Learning Algorithm are required. In this paper, we try to predict the entrance rate to high-ranking universities using a Deep Learning model, RNN(Recurrent Neural Network). First, we analyzed the current status of private academies in administrative districts and the number of students by age in administrative districts, and established a socially accepted hypothesis that students residing in areas with a high educational fever have a high rate of enrollment in high-ranking universities. This is to verify based on the data analyzed using the predicted hypothesis and the government's public data. The predictive model uses data from 2015 to 2017 to learn to predict the top enrollment rate, and the trained model predicts the top enrollment rate in 2018. A prediction experiment was performed using RNN, a Deep Learning model, for the high-ranking enrollment rate in the special education zone. In this paper, we define the correlation between the high-ranking enrollment rate by analyzing the household income and the participation rate of private education about the current status of private institutes in regions with high education fever and the effect on the number of students by age.

• Journal of the Korean Association of Geographic Information Studies
• /
• v.23 no.3
• /
• pp.26-67
• /
• 2020

Synthetic Aperture Radar(SAR) is able to photograph the earth's surface regardless of weather conditions, day and night. Because of its possibility to search for hydrological factors such as soil moisture and groundwater, and its importance is gradually increasing in the field of water resources. SAR began to be mounted on satellites in the 1970s, and about 15 or more satellites were launched as of 2020, which around 10 satellites will be launched within the next 5 years. Recently, various types of SAR technologies such as enhancement of observation width and resolution, multiple polarization and multiple frequencies, and diversification of observation angles were being developed and utilized. In this paper, a brief history of the SAR system, as well as studies for estimating soil moisture and hydrological components were investigated. Up to now hydrological components that can be estimated using SAR satellites include soil moisture, subsurface groundwater discharge, precipitation, snow cover area, leaf area index(LAI), and normalized difference vegetation index(NDVI) and among them, soil moisture is being studied in 17 countries in South Korea, North America, Europe, and India by using the physical model, the IEM(Integral Equation Model) and the artificial intelligence-based ANN(Artificial Neural Network). RADARSAT-1, ENVISAT, ASAR, and ERS-1/2 were the most widely used satellite, but the operation has ended, and utilization of RADARSAT-2, Sentinel-1, and SMAP, which are currently in operation, is gradually increasing. Since Korea is developing a medium-sized satellite for water resources and water disasters equipped with C-band SAR with the goal of launching in 2025, various hydrological components estimation researches using SAR are expected to be active.

• Journal of the Computational Structural Engineering Institute of Korea
• /
• v.24 no.1
• /
• pp.79-85
• /
• 2011

Bridge Weigh-in-Motion(BWIM) system calculates a travelling vehicle's weight without interruption of traffic flow by analyzing the signals that are acquired from various sensors installed in the bridge. BWIM system or data accumulated from the BWIM system can be utilized to development of updated live load model for highway bridge design, fatigue load model for estimation of remaining life of bridges, etc. Field test with moving trucks including various load cases should be performed to guarantee successful development of precise BWIM system. In this paper, a numerical simulation technique is adopted as an alternative or supplement to the vehicle traveling test that is indispensible but expensive in time and budget. The constructed numerical model is validated by comparison experimentally measured signal with numerically generated signal. Also vehicles with various dynamic characteristics and travelling conditions are considered in numerical simulation to investigate the variation of bridge responses. Considered parameters in the numerical study are vehicle velocity, natural frequency of the vehicle, height of entry bump, and lateral position of the vehicle. By analyzing the results, it is revealed that the lateral position and natural frequency of the vehicle should be considered to increase precision of developing BWIM system. Since generation of vehicle travelling signal by the numerical simulation technique costs much less than field test, a large number of test parameters can effectively be considered to validate the developed BWIM algorithm. Also, when artificial neural network technique is applied, voluminous data set required for training and testing of the neural network can be prepared by numerical generation. Consequently, proposed numerical simulation technique may contribute to improve precision and performance of BWIM systems.

• The Journal of the Convergence on Culture Technology
• /
• v.4 no.2
• /
• pp.197-204
• /
• 2018

One of the hottest issues in our 21st century is AI. Just as the automation of manual labor has been achieved through the Industrial Revolution in the agricultural society, the intelligence information society has come through the SW Revolution in the information society. With the advent of Google 'Alpha Go', the computer has learned and predicted its own machine learning, and now the time has come for the computer to surpass the human, even to the world of Baduk, in other words, the computer. Machine learning ML (machine learning) is a field of artificial intelligence. Machine learning ML (machine learning) is a field of artificial intelligence, which means that AI technology is developed to allow the computer to learn by itself. The time has come when computers are beyond human beings. Many companies use machine learning, for example, to keep learning images on Facebook, and then telling them who they are. We also used a neural network to build an efficient energy usage model for Google's data center optimization. As another example, Microsoft's real-time interpretation model is a more sophisticated translation model as the language-related input data increases through translation learning. As machine learning has been increasingly used in many fields, we have to jump into the AI industry to move forward in our 21st century society.

• KSCE Journal of Civil and Environmental Engineering Research
• /
• v.38 no.4
• /
• pp.579-586
• /
• 2018

Short-term prediction of travel speed has been widely studied using data-driven non-parametric techniques. There is, however, a lack of research on the prediction aimed at urban areas due to their complex dynamics stemming from traffic signals and intersections. The purpose of this study is to develop a hybrid approach combining ensemble empirical mode decomposition (EEMD) and artificial neural network (ANN) for predicting urban travel speed. The EEMD decomposes the time-series data of travel speed into intrinsic mode functions (IMFs) and residue. The decomposed IMFs represent local characteristics of time-scale components and they are predicted using an ANN, respectively. The IMFs can be predicted more accurately than their original travel speed since they mitigate the complexity of the original data such as non-linearity, non-stationarity, and oscillation. The predicted IMFs are summed up to represent the predicted travel speed. To evaluate the proposed method, the travel speed data from the dedicated short range communication (DSRC) in Daegu City are used. Performance evaluations are conducted targeting on the links that are particularly hard to predict. The results show the developed model has the mean absolute error rate of 10.41% in the normal condition and 25.35% in the break down for the 15-min-ahead prediction, respectively, and it outperforms the simple ANN model. The developed model contributes to the provision of the reliable traffic information in urban transportation management systems.

• Journal of the Korean Society of Surveying, Geodesy, Photogrammetry and Cartography
• /
• v.37 no.6
• /
• pp.435-444
• /
• 2019

Recently, DL (Deep Learning) has been rapidly applied in various fields. In particular, classification and object recognition from images are major tasks in computer vision. Most of the DL utilizing imagery is primarily based on the CNN (Convolutional Neural Network) and improving performance of the DL model is main issue. While most CNNs are involve with images for training data, this paper aims to classify and recognize objects using DSM (Digital Surface Model), and slope and aspect information derived from the DSM instead of images. The DSM data sets used in the experiment were established by DGPF (German Society for Photogrammetry, Remote Sensing and Geoinformatics) and provided by ISPRS (International Society for Photogrammetry and Remote Sensing). The CNN-based SegNet model, that is evaluated as having excellent efficiency and performance, was used to train the data sets. In addition, this paper proposed a scheme for training data generation efficiently from the limited number of data. The results demonstrated DSM and derived data could be feasible for semantic classification with desirable accuracy using DL.

• Investigative Magnetic Resonance Imaging
• /
• v.13 no.2
• /
• pp.127-136
• /
• 2009

Purpose : Functional magnetic resonance imaging (fMRI) was used to assess the temporal response of neural activation in healthy subjects while they performed the Iowa Gambling Test (IGT), which utilizes decisions involving ambiguity and risk. The IGT was divided into five blocks of 20 trials; analysis showed that activity in the medial prefrontal cortex (mPFC) moves gradually from the dorsal to the ventral mPFC over the course of the IGT. These findings suggest that cognitive division of the mPFC, including the dorsal portion of the anterior cingulated cortex (ACC), plays a major role in ambiguous decision making and that the aspect of the IGT corresponding to risky decision making is associated with significant activity within the corticolimbic network strongly implicated in emotion and reinforcement. Our results also suggest that decisions made under ambiguity and decisions made under risk situations can be further divided into sub-phases based on the neural network involved.

• Journal of the Korea Concrete Institute
• /
• v.23 no.4
• /
• pp.487-494
• /
• 2011

In this research, the possibility of using bottom ash as a binder for the alkali-activated cement mortar is studied. Several experiments were performed to investigate the variation of the material properties according to the mix proportion. In the experimental program, the flowability and compressive strength were evaluated for various values of water/ash ratio, activator/ash ratio, sodium silicate to sodium hydroxide ratio, curing temperature, and the fineness of bottom ash as the main variables. The experimental results showed that high strength of 40 MPa or greater could be achieved in $60^{\circ}C$ high temperature curing condition with proper flowability. For $20^{\circ}C$ ambient temperature curing, the 28 days compressive strength of approximately 30MPa could be obtained although the early-age strength development was very slow. Based on the results, the range of optimized mix design of bottom-ash based alkali-activated cement mortar was suggested. In addition, using the artificial neural network analysis, the flowability and compressive strength were predicted with the difference in the mix proportion of the bottom-ash based alkali-activated cement mortar.