• Journal of the Korea Computer Graphics Society
• /
• v.25 no.3
• /
• pp.123-131
• /
• 2019

In this study, we modified CT images of femoral head in consideration of anatomically meaningful structure, proposing the method to augment the training data of convolution Neural network for segmentation of femur mesh model. First, the femur mesh model is obtained from the CT image. Then divide the mesh model into meaningful parts by using cluster analysis on geometric characteristic of mesh surface. Finally, transform the segments by using an appropriate mesh deformation algorithm, then create new CT images by warping CT images accordingly. Deep learning models using the data enhancement methods of this study show better image division performance compared to data augmentation methods which have been commonly used, such as geometric conversion or color conversion.

• Journal of Auto-vehicle Safety Association
• /
• v.9 no.3
• /
• pp.19-23
• /
• 2017

This paper presents a lateral driver model with neuromuscular system to evaluate the performance of electric power steering (EPS). Output of most previously developed driver models is steering angle. However, in order to evaluate EPS system, driver model which results in steering torque output is needed. The proposed lateral driver model mainly consists of 2 parts: desired steering angle calculation and conversion of steering angle into steering torque. Desired steering angle calculation part results in steering angle to track desired yaw rate for path tracking. Conversion of steering angle into torque is consideration with neuromuscular system. The proposed driver model is investigated via actual driving data. Compared to other algorithms, the proposed algorithm shows similar pattern of steering angle with human driver. The proposed driver can be utilized to efficiently evaluate EPS system in simulation level.

• Spatial Information Research
• /
• v.21 no.5
• /
• pp.73-82
• /
• 2013

In recent years, according to the increase of interests in indoor space, various researches are being carried out for the construction and services of indoor spatial information. BIM data is very useful to build indoor spatial information. Accordingly, many studies for the use of BIM data on GIS part are in progress. In order to take advantage of BIM data on GIS part, the conversion technology for building indoor data and visualization techniques are required. However, most of the previous researches are focused on the conversion technology to construct indoor spatial information by importing BIM data into GIS applications while there is few research on visualization. In this study, an indoor LOD(Level of Detail) model is proposed to apply to on indoor facility management system when indoor data was constructed based on BIM data for the linkage between BIM and GIS.

• Journal of Korean Society of Forest Science
• /
• v.98 no.4
• /
• pp.409-416
• /
• 2009

Forest biomass estimation is essential for greenhouse gas inventories and terrestrial carbon accounting. Remote sensing allows for estimating forest biomass over a large area. This study was conducted to estimate forest biomass and to produce a forest biomass map for Muju county using forest biomass conversion table developed by field plot data from the 5th National Forest Inventory and Landsat TM-5. Correlation analysis was carried out to select suitable independent variables for developing regression models. It was resulted that the height class, crown closure density, and age class were highly correlated with forest biomass. Six regression models were used with the combination of these three stand variables and verified by validation statistics such as root mean square error (RMSE) and mean bias. It was found that a regression model with crown closure density and height class (Model V) was better than others for estimating forest biomass. A biomass conversion table by model V was produced and then used for estimating forest biomass in the study site. The total forest biomass of the Muju county was estimated about 8.8 million ton, or 128.3 ton/ha by the conversion table.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.24 no.6B
• /
• pp.1113-1122
• /
• 1999

In this paper, we propose MSS(Multicast Synchronization Server) and QCS(QoS Conversion Server) models that can support IP multicast and QoS(Quality of Service) over the ATM network more efficiently. In the MSS model, it is possible to establish shortcut VCs(Virtual Circuits) among all hosts in the ATM network and to transfer multicast data at high speed. Also the MSS model is more scalable, because the number of inter-cluster VCs needed in the MSS model is less than that of EARTH. In the QCS model, ATM switch is modified to support one QoS service and best effort service through the one point-to-multipoint VC at the same time so required network resources are reduced, and dedicated server is used for QoS conversion to accept heterogeneous receivers more efficiently. In this Paper, the proposed MSS model and QCS model have solved both efficiency problem and scalability problem. It is proved through the comparison of the VCs required in each model.

• Proceedings of the Korean Operations and Management Science Society Conference
• /
• 2008.10a
• /
• pp.444-447
• /
• 2008

This study is intended to develop a improved version of Cross Impact Analysis Model based on Scaling Theory. In developing the model, we applied the scale transformation technique and regression technique to existing CIA model. Improved CIA model is composed of two sub-models: 'model for impact value measurement,' and 'model for impact value conversion'. We applied a technique which measures data by ordinal scale and then transforms them into interval scale and ratio scale data to CIA model. The accuracy of forecasting and the usability of CIA application have been improved.

• Industrial Engineering and Management Systems
• /
• v.7 no.2
• /
• pp.160-170
• /
• 2008

This paper proposes a mathematical model for converting conveyor assembly line to cellular manufacturing in complex production environments. Complex production environments refer to the situations with multi-products, variant demand, different batch sizes and the worker abilities varying with work stations and products respectively. The model proposed in this paper aims to determine (1) how many cells should be formatted; (2) how many workers should be assigned in each cell; (3) and how many workers should be rested in shortened conveyor line when a conveyor assembly line should be converted, in order to optimize system performances which are defined as the total throughput time and total labor power. We refer the model to a new production system. Such model can be used as an evaluation tool in the cases of (i) when a company wants to change its production system (usually a belt conveyor line) to a new one (including cell manufacturing); (ii) when a company wants to evaluate the performance of its converted system. Simulation experiments based on the data collected from the previous documents are used to estimate the marginal impact that each factor change has had on the estimated performance improvement resulting from the conversion.

• Structural Monitoring and Maintenance
• /
• v.2 no.3
• /
• pp.213-236
• /
• 2015

The paper presents an investigation into the mode conversion and scattering characteristics of guided waves at delaminations in laminated composite beams. A three-dimensional (3D) finite element (FE) model, which is experimentally verified using data measured by 3D scanning laser vibrometer, is used in the investigation. The study consists of two parts. The first part investigates the excitability of the fundamental anti-symmetric mode ($A_0$) of guided wave in laminated composite beams. It is found that there are some unique phenomena, which do not exist for guided waves in plate structures, make the analysis become more complicated. The phenomena are observed in numerical study using 3D FE simulations. In the second part, several delaminated composite beams are studied numerically to investigate the mode conversion and scattering characteristics of the $A_0$ guided wave at delaminations. Different sizes, locations and through-thickness locations of the delaminations are investigated in detail. The mode conversion and scattering phenomena of guided waves at the delaminations are studied by calculating reflection and transmission coefficients. The results show that the sizes, locations and through-thickness locations of the delaminations have significant effects on the scattering characteristics of guided waves at the delaminations. The results of this research would provide better understanding of guided waves propagation and scattering at the delaminations in the laminated composite beams, and improve the performance of guided wave damage detection methods.

• Journal of the Korean Society of Industry Convergence
• /
• v.23 no.6_2
• /
• pp.949-956
• /
• 2020

It's important to measure fuel consumption of vehicles. It's possible to monitor green house gas from vehicles for various traffic conditions with the measured data. It's effective to eco-drive for drivers with fuel consumption data also. There's a display of fuel consumption in the modern vehicles, but it's not useful to get the data from the display. An estimating method for fuel consumption of a vehicle is suggested in the study. It's a simple but an effective method using GPS data. The GPS data(speed, acceleration, road slope) and vehicle data(weight, frontal area, model year, certified fuel economy) is necessary to estimate the fuel consumption for the method. It calculates driving resistance force to estimate engine power. Then it estimates the necessary fuel consumption to maintain the engine power with fuel-power conversion factor. The conversion factor is corrected with certified fuel economy, model year and rated power. The precision of the methods is checked with road test data. The test driving data was measured with GPS and OBD. The error of the estimated fuel consumption for the measured one is about 1.8%. But the error is large for the 1000 and 100 data number from the total data number of about 10,000. The error is from the larger change range of the GPS data than the one of the measured fuel consumption data. But the proposed estimating method is useful to percept the fuel consumption change for better fuel economy with simple gadget like smart phone or other GPS instruments.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.18 no.2
• /
• pp.311-326
• /
• 2024

The rapid development of neural network technology promotes the neural network model driven by big data to overcome the texture effect of complex objects. Due to the limitations in complex scenes, it is necessary to establish custom template matching and apply it to the research of many fields of computational vision technology. The dependence on high-quality small label sample database data is not very strong, and the machine learning system of deep feature connection to complete the task of texture effect inference and speculation is relatively poor. The style transfer algorithm based on neural network collects and preserves the data of patterns, extracts and modernizes their features. Through the algorithm model, it is easier to present the texture color of patterns and display them digitally. In this paper, according to the texture effect reasoning of custom template matching, the 3D visualization of the target is transformed into a 3D model. The high similarity between the scene to be inferred and the user-defined template is calculated by the user-defined template of the multi-dimensional external feature label. The convolutional neural network is adopted to optimize the external area of the object to improve the sampling quality and computational performance of the sample pyramid structure. The results indicate that the proposed algorithm can accurately capture the significant target, achieve more ablation noise, and improve the visualization results. The proposed deep convolutional neural network optimization algorithm has good rapidity, data accuracy and robustness. The proposed algorithm can adapt to the calculation of more task scenes, display the redundant vision-related information of image conversion, enhance the powerful computing power, and further improve the computational efficiency and accuracy of convolutional networks, which has a high research significance for the study of image information conversion.