• Journal of the Korean Institute of Intelligent Systems
• /
• v.11 no.4
• /
• pp.354-358
• /
• 2001

This paper presents a fuzzy neural network model which solves the underutilization problem. This fuzzy neural network has both stability and flexibility because it uses the control structure similar to AHT(Adaptive Resonance Theory)-l neural network. And this fuzzy nenral network does not need to initialize weights and is less sensitive to noise than ART-l neural network is. The learning rule of this fuzzy neural network is the modified and fuzzified version of Kohonen learning rule and is based on the fuzzification of leaky competitive leaming and the fuzzification of conditional probability. The similarity measure of vigilance test, which is performed after selecting a winner among output neurons, is the relative distance. This relative distance considers Euclidean distance and the relative location between a datum and the prototypes of clusters. To compare the performance of the proposed fuzzy neural network with that of Kohonen Self-Organizing Feature Map the IRIS data and Gaussian-distributed data are used.

• KSCE Journal of Civil and Environmental Engineering Research
• /
• v.33 no.5
• /
• pp.1871-1885
• /
• 2013

In this study, spatio-temporal distribution of future drought in South Korea was predicted by using the meteorological data generated from GCMs on which a variety of climate changing scenarios are applied. Drought phenomena was quantitatively analyzed using SPI(Standardized Precipitation Index). In addition, potential drought hazard maps for different drought duration and return period were made for the South Koreaby drought frequency analysis after deriving SDF(Severity-Duration-Frequency) curves using the 54 weather stations throughout the country. From the potential drought hazard maps, drought is expected to be severer in Nakdong River basin and Seomjin River basin under A2 scenario. It was also predicted that drought would be severe in the Han River basin by RCP8.5 scenario. In the future, potential drought hazard area would be expanded until the Eastern part of Nakdong River basin as compared with that of past under A2 scenario condition. Research results indicated that future drought would be extensively occurred all areas of South Korea not limiting in the southern part of country.

• The Journal of the Korea Contents Association
• /
• v.9 no.7
• /
• pp.159-170
• /
• 2009

It is very important to extract the expression data and capture a face image from a video for online-based 3D face animation. In recently, there are many researches on vision-based approach that captures the expression of an actor in a video and applies them to 3D face model. In this paper, we propose an automatic data extraction system, which extracts and traces a face and expression data from realtime video inputs. The procedures of our system consist of three steps: face detection, face feature extraction, and face tracing. In face detection, we detect skin pixels using YCbCr skin color model and verifies the face area using Haar-based classifier. We use the brightness and color information for extracting the eyes and lips data related facial expression. We extract 10 feature points from eyes and lips area considering FAP defined in MPEG-4. Then, we trace the displacement of the extracted features from continuous frames using color probabilistic distribution model. The experiments showed that our system could trace the expression data to about 8fps.

• KIPS Transactions on Software and Data Engineering
• /
• v.6 no.12
• /
• pp.573-580
• /
• 2017

Deep Neural Network (DNN) is a large layered neural network which is consisted of a number of layers of non-linear units. Deep Learning which represented as DNN has been applied very successfully in various applications. However, many issues in DNN have been identified through past researches. Among these issues, generalization is the most well-known problem. A Recent study, Dropout, successfully addressed this problem. Also, Dropout plays a role as noise, and so it helps to learn robust feature during learning in DNN such as Denoising AutoEncoder. However, because of a large computations required in Dropout, training takes a lot of time. Since Dropout keeps changing an inter-layer representation during the training session, the learning rates should be small, which makes training time longer. In this paper, using mutation operation, we reduce computation and improve generalization performance compared with Dropout. Also, we experimented proposed method to compare with Dropout method and showed that our method is superior to the Dropout one.

• Journal of the Earthquake Engineering Society of Korea
• /
• v.12 no.5
• /
• pp.23-29
• /
• 2008

This study investigates the dynamic instability of strength-limited bilinear single degree of freedom (SDF) systems under seismic excitation. The strength-limited bilinear hysteretic model best replicates the hysteretic behavior of the steel moment resisting frames. To estimate the dynamic instability of SDF systems, the collapse strength ratio is used, which is the yield-strength reduction factor when collapse occurs. Statistical studies are carried out to estimate median collapse strength ratios and those dispersions of strength-limited bilinear SDF systems with given natural periods, hardening stiffness ratios, post-capping stiffness ratios, ductility and damping ratios ranging from 2 to 20% subjected to 240 earthquake ground motions recorded on stiff soil sites. Equations to calculate median and standard deviation of collapse strength ratios in strength-limited bilinear SDF systems are obtained through nonlinear regression analysis. By using the proposed equations, this study estimated the probabilistic distribution of collapse strength ratios, and compared this with the exact values from which the accuracy of the proposed equations was verified.

• Journal of KIISE
• /
• v.43 no.7
• /
• pp.795-800
• /
• 2016

In image classification, recent CNNs compete with human performance. However, there are limitations in more general recognition. Herein we deal with indoor images that contain too much information to be directly processed and require information reduction before recognition. To reduce the amount of data processing, typically variational inference or variational Bayesian methods are suggested for object detection. However, these methods suffer from the difficulty of marginalizing over the given space. In this study, we propose an image-text integrated recognition system using active vision based on Spatial Transformer Networks. The system attempts to efficiently sample a partial region of a given image for a given language information. Our experimental results demonstrate a significant improvement over traditional approaches. We also discuss the results of qualitative analysis of sampled images, model characteristics, and its limitations.

• KSCE Journal of Civil and Environmental Engineering Research
• /
• v.26 no.4D
• /
• pp.593-600
• /
• 2006

Joint opening of jointed concrete pavement is caused by change in temperature and humidity of adjoined slab. The magnitude of joint opening influences on the load-transfer-efficiency and the behavior of sealant. If temperature or humidity decreases, joint opening increases. Generally maximum joint opening of a given joint is predicted by using AASHTO equation. While different magnitudes of joint opening at the individual joints have been observed in a given pavement section, AASHTO equation is limited to predict average joint opening in a given pavement section. Therefore the AASHTO equation may underestimate maximum joint for the half of joint in a given pavement section. Joints showing larger opening than the designed may experience early joint sealant failure, early faulting. Also unexpected spalling may be followed due to invasion of fine aggregate into the joints after sealant pop-off. In this study, the variation of the joint opening in a given pavement section was investigated based on the LTPP SMP data. Factors affecting on the variation are explored. Finally a probabilistic joint opening model is developed. This model can account for the reliability of the magnitude of joint opening so that the designer can select the ratio of underestimated joint opening.

• Journal of Korean Tunnelling and Underground Space Association
• /
• v.6 no.1
• /
• pp.25-40
• /
• 2004

The supersonic shock wave generated by fully coupled explosion will change into subsonic shock wave, plastic wave, and elastic wave consecutively as the wave propagates through rock mass. While the estimation of the blast-induced peak pressure was the main aim of the companion paper, this paper will concentrate on the estimation of the rise time of blast-induced pressure. The rise time can be expressed as a function of explosive density, isentropic exponent, detonation velocity, exponential coefficient of the peak pressure attenuation, dynamic yield stress, plastic wave velocity, elastic wave velocity, rock density, Hugoniot parameters, etc. Parametric analysis was performed to pinpoint the most influential parameter that affects the rise time and it was found that rock properties are more sensitive than explosive properties. The probabilistic distribution of the rise time is evaluated by the Rosenblueth'S point estimate method from the probabilistic distributions of explosive properties and rock properties. Numerical analysis was performed to figure out the effect of rock properties and explosive properties on the uncertainty of blast-induced vibration. Uncertainty analysis showed that uncertainty of rock properties constitutes the main portion of blast-induced vibration uncertainty rather than that of explosive properties. Numerical analysis also showed that the loading rate, which is the ratio of the peak blasting pressure to the rise time, is the main influential factor on blast-induced vibration. The loading rate is again more influenced by rock properties than by explosive properties.

• Journal of the Institute of Electronics Engineers of Korea CI
• /
• v.48 no.1
• /
• pp.127-132
• /
• 2011

In this paper, we propose a novel method of adult image classification by combining skin color regions and edges in an input image. The proposed method consists of four steps. In the first step, initial skin color regions are detected by logical AND operation of all skin color regions detected by the existing methods of skin color detection. In the second step, a skin color probability map is created by modeling the distribution of skin color in the initial regions. Then, a binary image is generated by using threshold value from the skin color probability map. In the third step, after using the binary image and edge information, we detect final skin color regions using a region growing method. In the final step, adult image classification is performed by support vector machine(SVM). To this end, a feature vector is extracted by combining the final skin color regions and neighboring edges of them. As experimental results, the proposed method improves performance of the adult image classification by 9.6%, compared to the existing method.

• KSCE Journal of Civil and Environmental Engineering Research
• /
• v.43 no.1
• /
• pp.109-117
• /
• 2023

Schedule delays present a major risk factor, as they can adversely affect construction projects, such as through increasing construction costs, claims from a client, and/or a decrease in construction quality due to trims to stages to catch up on lost time. Risk management has been conducted according to the importance and priority of schedule delay risk, but quantification of risk on the depth of schedule delay tends to be inadequate due to limitations in data collection. Therefore, this research used the BERT (Bidirectional Encoder Representations from Transformers) language model to convert the contents of aconstruction log, which comprised unstructured data, into WBS (Work Breakdown Structure)-based structured data, and to form a model of classification and quantification of risk. A process was applied to eight highway construction sites, and 75 cases of rain schedule delay risk were obtained from 8 out of 39 detailed work kinds. Through a K-S test, a significant probability distribution was derived for fourkinds of work, and the risk impact was compared. The process presented in this study can be used to derive various schedule delay risks in construction projects and to quantify their depth.