• KSCE Journal of Civil and Environmental Engineering Research
• /
• v.28 no.6D
• /
• pp.759-766
• /
• 2008

Road classification system is the first step for determining the road function and design standards. Currently, roads are classified by various indices such as road location and function. In this study, we classify road using various traffic indices as well as to identify traffic characteristics for each type of road. To accomplish the objectives, mixture model was applied for classifying road and analyzing traffic characteristics using traffic data that observed at permanent traffic count stations. A total of 8 variables were applied: annual average daily traffic(AADT), $K_{30}$ coefficient, heavy vehicle proportion, day volume proportion, peak hour volume proportion, sunday coefficient, vacation coefficient, and coefficient of variation(COV). A total of 350 permanent traffic count points were categorized into three groups : Group I (Urban road), Group II (Rural road), and Group III (Recreational road). AADT were 30,000 for urban, 16,000 for rural, and 5,000 for recreational road. Group III was typical recreational road showing higher average daily traffic volume during Sunday and vacational periods. Group I showed AM peak and PM peak, while group II and group III did not show AM peak and PM peak.

• Information Systems Review
• /
• v.12 no.2
• /
• pp.189-203
• /
• 2010

Is presenting guidance of information performance measurement as government PRM version 2.0 these common reference models in public sector. Government PRM is consisted of assessment classification system and standard line of sight and performance management standard form. Through this, is sorting performance element and define cause-and effect. Government PRM is supplying measurement categories at assessment classification system, but relative importance for application standard by measurement categories is not presenting. In this study, importance for government PRM's measurement categories been applying by commonness Test of information performance measurement of public sector wishes to deduce estimation and priority. Research model used Fuzzy QFD, and designed so that can reflect well PRM's development purpose. I applied Fuzzy AHP and FPP method that graft together fuzzy theory to minimize uncertainty and ambiguity in that expert opinion. Is drawn to element that "Standard model offer for information department and management" is the most important in government PRM's development purpose. "Quality of service" is showing the highest priority in customer results in measurement category. Importance for government PRM's measurement categories can offer common valuation basis in government and public institution. Hereafter if examine closely quantitative cause-and effect for structure model of measurement classification system when study government PRM more objective and efficient reference model become.

• Journal of the Institute of Convergence Signal Processing
• /
• v.24 no.3
• /
• pp.134-139
• /
• 2023

Rehabilitators perform outpatient treatment and daily rehabilitation exercises to recover physical function with the aim of quickly returning to society after surgical treatment. Unlike performing exercises in a hospital with the help of a professional therapist, there are many difficulties in performing rehabilitation exercises by the patient on a daily basis. In this paper, we propose a CNN-LSTM-based upper limb rehabilitation real-time monitoring system so that patients can perform rehabilitation efficiently and with correct posture on a daily basis. The proposed system measures biological signals through shoulder-mounted hardware equipped with EMG and IMU, performs preprocessing and normalization for learning, and uses them as a learning dataset. The implemented model consists of three polling layers of three synthetic stacks for feature detection and two LSTM layers for classification, and we were able to confirm a learning result of 97.44% on the validation data. After that, we conducted a comparative evaluation with the Teachable machine, and as a result of the comparative evaluation, we confirmed that the model was implemented at 93.6% and the Teachable machine at 94.4%, and both models showed similar classification performance.

• The Korean Journal of Archival Studies
• /
• no.21
• /
• pp.169-246
• /
• 2009

The national records management system of the Korean Government has been developed in a close relationship with changes in the administrative system. The national records management system established immediately after the establishment of the Korean Government, began to be reformed as a system with a new feature during the quick transition of the administrative system during the early 1960s. Particularly this new system holds an important meaning in that it began to cope with the mass production system of records and was established on the government level for the first time since the establishment of the government. Also this was a basic framework that defined the records management pattern of the Korean Government for the later 40 years. Therefore, this study aims to identify the origin and the meaning of the national records management system established during the early 1960s. At the time of establishing the government, the administrative system of the Korean Government was not completely free from the framework of the administrative system of the Chosen General Government. This was mainly because the Korean Government had no capability to renovate the administrative system. This was not an exception also for the national records management system. In other words, the forms and preparation methods of official document, an official document management process, and the classification and appraisal system used the records management system of the Chosen General Government without any alteration. Main factors that brought about the reform of the national records management system as well as the change in the Korean administrative system during the early 1960s, were being created in Korean society, starting from the mid 1950s. This resulted from the growth of Korean Army, public officers, and students of administrative science as being the intrinsic elites of Korean society through their respective experience of the US administration. In particular, the reform of the creation, classification, filing, transfer, and preservation system shown during the introduction of a scientific management system of the US Army in the Korean Army was a meaningful change given the historic developing process of Korean records management system history. This change had a decisive effect on the reform of the national records management system during the early 1960s. As the Korean Army, public officers, and students of administrative science, who had posted growth beginning in the mid-1950s, emerged as administrative elites during the early 1960s, the administrative system of the Korean Government brought about a change, which was different from the past in terms of its quality, and the modernization work of documentary administration pursued during the period, became extended to the reform of the national records management system. Then, the direction of reform was 'the efficient and effective control' over records based on scientific management, which was advanced through the medium of the work that accommodate the US office management system and a decimal filing system to Korean administrative circumstances. Consequently, Various official document forms, standards, and the gist of process were improved and standardized, and the appraisal system based on the function-based classification were unified on the government level by introducing a decimal filing system.

• Journal of Environmental Impact Assessment
• /
• v.14 no.3
• /
• pp.97-107
• /
• 2005

The purpose of this study is to develop a system, which estimates watershed pollutant loading rate through the combination of GIS and computational mode. Also, the applicability of this study was estimated by the application of the above system for Chuncheon City. The detailed results of these studies are as follows; The pollutant loading estimation system was developed for more convenient estimation of pollutant loading rate in watershed, and the system load was minimized by the separation of estimation module for point and non-point source. This system on the basis of GIS is very economical and efficient because it can be applied to other watershed with the watershed map. System modification is not needed. The pollutant loading estimation system for point source was developed to estimate the pollutant loading rate in watershed through the extraction of the proper data from all districts and yearly data and the execution of spatial analysis which is main function of GIS. From the verification result of spatial analysis, real watershed area and the administrative districtarea extracted by spatial analysis were $1,114,893,340.15m^2$ and $1,114,878,683.68m^2$, respectively. It shows that the spatial analysis results were very exact with only 0.001% error. The pollutant loading estimation system for non-point source was developed to calculate the pollutant loading rate through the overlaying of land-use and watershed map after the construction of new land-use map using the land register database with most exact land use classification. Application result for Chuncheon City shows that the proposed system results in one percent land use error while the statistical method results in five percent. More exact nonpoint source pollutant loading was estimated from this system.

• Journal of the Korean Association of Geographic Information Studies
• /
• v.11 no.4
• /
• pp.101-111
• /
• 2008

To provide more natural elements in the harsh urban environment, 'planting trees as urban forests' has been emphasized as having an important role, and trees are expected to be as functional as the trees under more natural conditions in rural areas, and provide people with benefits. To do this, urban forest policies needed a better idea of planting methods and management of trees through the theory of landscape ecology, and also the feedback system according to the evaluation and assessment of urban forests. In this case study, a new principle and assessment indices for the evaluation are applied for the 4 urban forests in two Korean metropolitan cities, Daejeon and Ulsan. The evaluation of Korean urban forest-function as biotope and the assessment for the classification of biotope diversity types are carried out. The AUEM(Adding Up Estimation Matrix) is applied for the analysis of urban forests. Unlikely previous researches on the other Korean metropolitan cities, the size of urban forest has less influence on the vegetation diversity. The most frequent biotope grade is the 3rd grade at Namseon park and Mt. Hamwol, while Mt. Bomun and Mt. Yeompo show the 4th grade. The grades of forest-function as biotope are from 3rd to 5th in which lower than average in forest-function grades. This means that the 4 sites are still not-matured forests and less-functional forests as the urban biotope.

• Journal of the Korea Institute of Building Construction
• /
• v.23 no.4
• /
• pp.417-428
• /
• 2023

The demolition of domestic infrastructures mirrors other significant construction initiatives in presenting a markedly high accident rate. A comprehensive investigation into the origins of such accidents is crucial for the prevention of future incidents. Upon detailed inspection, the causes of demolition construction accidents are multifarious, encompassing unsafe worker behavior, hazardous conditions, psychological and physical states, and site management deficiencies. While statistics relating to demolition construction accidents are consistently collated and reported, there exists an exigent need for a more foundational cause categorization system based on accident type. Drawing from Heinrich's Domino Theory, this study classifies the origins of accidents(unsafe behavior, unsafe conditions) and human errors(human factors) as per the type of accidents experienced during demolition construction. In this study, a three-step model of QFD-FMEA(Quality Function Deployment - Failure Mode Effect Analysis) is employed to systematically categorize accident causes according to the types of accidents that occur during demolition construction. The QFD-FMEA method offers a technique for cause classification at each stage of the demolition process, including direct causes(unsafe behavior, unsafe environment), and human errors(human factors) through a tri-stage process. The results of this accident cause classification can serve as safety knowledge and reference checklists for accident prevention efforts.

• Journal of Korean Society of Archives and Records Management
• /
• v.16 no.2
• /
• pp.81-112
• /
• 2016

This study aims to propose the adoption of a business management system, as well as suggest the functions and development directions for public organizations, which are required to establish the record management and information disclosure system under Government 3.0, rapidly respond to the needs for strengthening the responsibilities for explanation, and improve work efficiency. Recently, some of the public organizations that introduced the record management and information disclosure system adopted the Electronic Document System, which focuses on the function of electronic approval, and developed a records classification scheme for the system. This study aims to review the case of A organization, which recently developed an in-house records management system and established information strategy planning to adopt a customized business management system after establishing a business reference model throughout the organization, and suggests the directions of the electronic record production system for public organizations.

• Journal of Intelligence and Information Systems
• /
• v.25 no.4
• /
• pp.53-65
• /
• 2019

Object tracking is one of important steps to achieve video-based surveillance systems. Object tracking is considered as an essential task similar to object detection and recognition. In order to perform object tracking, various machine learning methods (e.g., least-squares, perceptron and support vector machine) can be applied for different designs of tracking systems. In general, generative methods (e.g., principal component analysis) were utilized due to its simplicity and effectiveness. However, the generative methods were only focused on modeling the target object. Due to this limitation, discriminative methods (e.g., binary classification) were adopted to distinguish the target object and the background. Among the machine learning methods for binary classification, total error rate minimization can be used as one of successful machine learning methods for binary classification. The total error rate minimization can achieve a global minimum due to a quadratic approximation to a step function while other methods (e.g., support vector machine) seek local minima using nonlinear functions (e.g., hinge loss function). Due to this quadratic approximation, the total error rate minimization could obtain appropriate properties in solving optimization problems for binary classification. However, this total error rate minimization was based on a batch mode setting. The batch mode setting can be limited to several applications under offline learning. Due to limited computing resources, offline learning could not handle large scale data sets. Compared to offline learning, online learning can update its solution without storing all training samples in learning process. Due to increment of large scale data sets, online learning becomes one of essential properties for various applications. Since object tracking needs to handle data samples in real time, online learning based total error rate minimization methods are necessary to efficiently address object tracking problems. Due to the need of the online learning, an online learning based total error rate minimization method was developed. However, an approximately reweighted technique was developed. Although the approximation technique is utilized, this online version of the total error rate minimization could achieve good performances in biometric applications. However, this method is assumed that the total error rate minimization can be asymptotically achieved when only the number of training samples is infinite. Although there is the assumption to achieve the total error rate minimization, the approximation issue can continuously accumulate learning errors according to increment of training samples. Due to this reason, the approximated online learning solution can then lead a wrong solution. The wrong solution can make significant errors when it is applied to surveillance systems. In this paper, we propose an exactly reweighted technique to recursively update the solution of the total error rate minimization in online learning manner. Compared to the approximately reweighted online total error rate minimization, an exactly reweighted online total error rate minimization is achieved. The proposed exact online learning method based on the total error rate minimization is then applied to object tracking problems. In our object tracking system, particle filtering is adopted. In particle filtering, our observation model is consisted of both generative and discriminative methods to leverage the advantages between generative and discriminative properties. In our experiments, our proposed object tracking system achieves promising performances on 8 public video sequences over competing object tracking systems. The paired t-test is also reported to evaluate its quality of the results. Our proposed online learning method can be extended under the deep learning architecture which can cover the shallow and deep networks. Moreover, online learning methods, that need the exact reweighting process, can use our proposed reweighting technique. In addition to object tracking, the proposed online learning method can be easily applied to object detection and recognition. Therefore, our proposed methods can contribute to online learning community and object tracking, detection and recognition communities.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.22 no.2
• /
• pp.135-140
• /
• 2012

The mirror neuron system in the cerebrum, which are handled by visual information-based imitative learning. When we observe the observer's range of mirror neuron system, we can assume intention of performance through progress of neural activation as specific range, in include of partially hidden range. It is goal of our paper that imitative learning is applied to 3D vision-based intelligent system. We have experiment as stereo camera-based restoration about acquired 3D image our previous research Using Optical flow, unscented Kalman filter. At this point, 3D input image is sequential continuous image as including of partially hidden range. We used Hidden Markov Model to perform the intention recognition about performance as result of restoration-based hidden range. The dynamic inference function about sequential input data have compatible properties such as hand gesture recognition include of hidden range. In this paper, for proposed intention recognition, we already had a simulation about object outline and feature extraction in the previous research, we generated temporal continuous feature vector about feature extraction and when we apply to Hidden Markov Model, make a result of simulation about hand gesture classification according to intention pattern. We got the result of hand gesture classification as value of posterior probability, and proved the accuracy outstandingness through the result.