• KSCE Journal of Civil and Environmental Engineering Research
• /
• v.29 no.1D
• /
• pp.1-9
• /
• 2009

Bus information system(BIS), as a part of the intelligent transportation system(ITS), is one of the most advanced public transportation systems which provide the real-time bus traffic information for the users waiting the buses at the bus stop. However, correct bus information data, such as the present bus location, the user waiting time, the bus arrival time, etc. are not provided for the bus users because the proper bus arrival time predictive models are not used yet in most of the cities operating the bus information system, including the metropolitan City of Ulsan. Thus, the purpose in this study is to investigate real-time bus traffic characteristic data for identifying the bus operation characteristics on the arterial under the study in the metropolitan City of Ulsan, analyze real-time bus traffic characteristic data on the ID locations of the arterial under the study, construct the optimal unit segment models for the unit segments which are the bus stop, node and travel section using the exponential smoothing, weighted smoothing and Kalman Filter methods, respectively, and finally suggest the optimal integrated model for predicting the real-time bus arrival time at the bus stop of the arterial under the study.

• Journal of Distribution Science
• /
• v.17 no.3
• /
• pp.69-83
• /
• 2019

Purpose - This paper investigates the long term impact of RFID investment on firm value in Korea. We wand to find out why the long term performance of some firm's RFID investment is better than others. To understand the dynamics of the long term returns from RFID investment announcements, we divide our events into groups for each of the independent firm characteristic variable such as investment time period, kind of markets, industries, solvency and growth potential. We composed portfolios based on the RFID investment announcement date for each group and evaluate the monthly abnormal excess returns. Research design, data, and methodology - Based on these calendar-time portfolios, we measure the long term returns from 86 RFID investment announcements of 46 firms from 2003 to 2017. We construct the calendar-time portfolio for 3, 6, 9, 12 months of holding periods. Using the weighted least squares method, we regress the raw monthly returns of the portfolios on the Fama-French model and Carhart(1997) model. As a result, we can get the estimated risk adjusted mean monthly abnormal excess return α_P for each of the calendar-time portfolio. Results - We found that early adopters, large firms, non-manufacturing firms have very significant excess returns. We also found modestly significant excess returns for financially stable firms and slow growing firms. Put together, top managers of the firms which plan to invest RFID should understand the strategic role of RFID adoption and the generalized business process of distribution information technology investment in Korea. Moreover, the findings of this paper provide useful trading strategies to the managers of large funds who are considering on investing in RFID adopting firms. Conclusions - Put together, the results of this paper give us a new insight into how the RFID and IT technology in general and other characteristic factors' interactions affect the long term performance of firms. Using the unbiased estimates of long term returns of the calendar-time portfolios, this paper extends the understandings on short term impact of RFID adoption of existing studies. This paper also extends the current understandings of firm characteristics that affect the long term performance of RFID adopting firms.

• Journal of Korean Society of Transportation
• /
• v.28 no.1
• /
• pp.135-142
• /
• 2010

The convenient techniques for predicting the bus arrival time have used the data obtained from the buses belong to the same company only. Consequently, the conventional techniques have often failed to predict the bus arrival time at the downstream bus stops due to the lack of the data during congestion time period. The primary objective of this study is to overcome the weakness of the conventional techniques. The estimation model developed based on the data obtained from Bus Information System(BIS) and Bus management System(BMS). The proposed model predicts the bus arrival time at bus stops by using the data of all buses travelling same roadway section during the same time period. In the tests, the proposed model had a good accuracy of predicting the bus arrival time at the bus stops in terms of statistical measurements (e.g., root mean square error). Overall, the empirical results were very encouraging: the model maintains a prediction job during the morning and evening peak periods and delivers excellent results for the severely congested roadways that are of the most practical interest.

• The Journal of the Korea institute of electronic communication sciences
• /
• v.11 no.8
• /
• pp.759-766
• /
• 2016

This paper is on a system design of recognizing sound sources and tracing locations from detecting a section of sound sources which is strong in surrounding environmental sounds about sound sources occurring in an abnormal situation by using signals within the section. In detection of the section with strong sound sources, weighted average delta energy of a short section is calculated from audio signals received. After inputting it into a low-pass filter, through comparison of values of the output result, a section strong in background sound is defined. In recognition of sound sources, from data of the detected section, using an HMM(: Hidden Markov Model) as a traditional recognition method, learning and recognition are realized from creating information to recognize sound sources. About signals of sound sources that surrounding background sounds are included, by using energy of existing signals, after detecting the section, compared with the recognition through the HMM, a recognition rate of 3.94% increase is shown. Also, based on the recognition result, location grasping by using TDOA(: Time Delay of Arrival) between signals in the section accords with 97.44% of angles of a real occurrence location.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.10 no.1
• /
• pp.364-380
• /
• 2016

The most popular approach in object classification is based on the bag of visual-words model, which has several fundamental problems that restricting the performance of this method, such as low time efficiency, the synonym and polysemy of visual words, and the lack of spatial information between visual words. In view of this, an object classification based on weakly supervised E2LSH and saliency map weighting is proposed. Firstly, E2LSH (Exact Euclidean Locality Sensitive Hashing) is employed to generate a group of weakly randomized visual dictionary by clustering SIFT features of the training dataset, and the selecting process of hash functions is effectively supervised inspired by the random forest ideas to reduce the randomcity of E2LSH. Secondly, graph-based visual saliency (GBVS) algorithm is applied to detect the saliency map of different images and weight the visual words according to the saliency prior. Finally, saliency map weighted visual language model is carried out to accomplish object classification. Experimental results datasets of Pascal 2007 and Caltech-256 indicate that the distinguishability of objects is effectively improved and our method is superior to the state-of-the-art object classification methods.

• Journal of The Korean Astronomical Society
• /
• v.47 no.6
• /
• pp.209-214
• /
• 2014

In this study we develop a set of solar proton event (SPE) forecast models with NOAA scales by Multi Layer Perceptron (MLP), one of neural network methods, using GOES solar X-ray flare data from 1976 to 2011. Our MLP models are the first attempt to forecast the SPE scales by the neural network method. The combinations of X-ray flare class, impulsive time, and location are used for input data. For this study we make a number of trials by changing the number of layers and nodes as well as combinations of the input data. To find the best model, we use the summation of F-scores weighted by SPE scales, where F-score is the harmonic mean of PODy (recall) and precision (positive predictive value), in order to minimize both misses and false alarms. We find that the MLP models are much better than the multiple linear regression model and one layer MLP model gives the best result.

• Journal of Korean Port Research
• /
• v.14 no.4
• /
• pp.451-461
• /
• 2000

The design and maintenance of navigation channel and water facilities of an harbor which is located at the mouth of river or at the estuary area are difficult due to the complexity of estuarial water and sediment circulation. Effects of deepening navigable waterways, of changing coastline configurations, or of discharging dredged material to the open sea are necessary to be investigated and predicted in terms of water quality and possible physical changes to the coastal environment. A borad analysis of the transport mechanism in the estuary area was made in terms of sediment property, falling velocity, concentration and flow characteristics. In order to simulate the transport processes, a two-dimensional finite element model is developed, which includes erosion, transport and deposition mechanism of suspended sediments. Galerkin’s weighted residual method is used to solve the transient convection-diffusion equation. The fluid domain is subdivided into a series of triangular elements in which a quadratic approximation is made for suspended sediment concentration. Model could deal with a continuous aggregation by stipulating the settling velocity of the flocs in each element. The model provides suspended sediment concentration, bed shear stress, erosion versus deposition rate and bed profile at the given time step.

• Proceedings of the Acoustical Society of Korea Conference
• /
• 1994.06a
• /
• pp.893-898
• /
• 1994

In this paper, the influence of voice source estimation and modeling on speech synthesis and coding is examined and then their new estimation and modeling techniques are proposed and verified by computer simulation. It is known that the existing speech synthesizer produced the speech which is dull and inanimated. These problems are arised from the fact that existing estimation and modeling techniques can not give more accurate voice parameters. Therefore, in this paper we propose a new voice source estimation algorithm and modeling techniques which can not give more accurate voice parameters. Therefore, in this paper we propose a new voice source estimation algorithm and modeling techniques which can represent a variety of source characteristics. First, we divide speech samples in one pitch region into four parts having different characteristics. Second, the vocal-tract parameters and voice source waveforms are estimated in each regions differently using sequential SVD. Third, we propose composite source model as a new voice source model which is represented by weighted sum of pre-defined basis functions. And finally, the weights and time-shift parameters of the proposed composite source model are estimeted uning EM(estimate maximize) algorithm. Experimental results indicate that the proposed estimation and modeling methods can estimate more accurate voice source waveforms and represent various source characteristics.

• Journal of Korea Multimedia Society
• /
• v.25 no.9
• /
• pp.1257-1265
• /
• 2022

Recognition of sign language is challenging due to the occlusion of hands, accuracy of hand gestures, and high computational costs. In recent years, deep learning techniques have made significant advances in this field. Although these methods are larger and more complex, they cannot manage long-term sequential data and lack the ability to capture useful information through efficient information processing with faster convergence. In order to overcome these challenges, we propose a word-level sign language recognition (SLR) system that combines a real-time human pose detection library with the minimized version of the gated recurrent unit (GRU) model. Each gate unit is optimized by discarding the depth-weighted reset gate in GRU cells and considering only current input. Furthermore, we use sigmoid rather than hyperbolic tangent activation in standard GRUs due to performance loss associated with the former in deeper networks. Experimental results demonstrate that our pose-based optimized GRU (Pose-OGRU) outperforms the standard GRU model in terms of prediction accuracy, convergency, and information processing capability.

• Journal of environmental and Sanitary engineering
• /
• v.16 no.1
• /
• pp.1-8
• /
• 2001

Since most people spend over 80% of their time indoor, indoor air quality tends to be the dominant contributor to personal exposure. In this study, indoor and outdoor $NO_2$concentrations were measured and compared with simultaneously personal exposures of 27 house-wives and female workers of kindergarten. Time activity pattern and house characteristics were used to determine the effects of these factors on personal exposure. Since house-wives student spent most their times in indoor with mean of 89.8%, their $NO_2$ exposure was associated with indoor $NO_2$ level(r= 0.92) rather than outdoor $NO_2$ level(r= 0.87). female workers were also associated with indoor $NO_2$ level(r= 0.70) though sample number were small. Using time-weighted average model, $NO_2$ exposures of house-wives were estimated by $NO_2$ measurements in indoor home and outdoor home levels. Estimated $NO_2$ personal exposures were significantly correlated with measured $NO_2$ personal exposures (r= 0.90). These results might mean that air pollutants exposure of old and feeble persons, and infants could be estimated by measuring concentrations of indoor home.