• Journal of the Korean Society of Radiology
• /
• v.18 no.5
• /
• pp.531-540
• /
• 2024

Recently, research and development on automatic diagnosis solutions in the medical imaging field using deep learning are actively underway. In this study, we sought to find a fast and accurate classification deep learning modeling for classification of pneumonia in chest images using Inception V3, a deep learning model based on a convolutional artificial neural network. For this reason, after applying the optimization algorithms AdaGrad, RMS Prop, and Adam to deep learning modeling, deep learning modeling was implemented by selectively applying learning rates of 0.01 and 0.001, and then the performance of chest X-ray image pneumonia classification was compared and evaluated. As a result of the study, in verification modeling that can evaluate the performance of the classification model and the learning state of the artificial neural network, it was found that the performance of deep learning modeling for classification of the presence or absence of pneumonia in chest X-ray images was the best when applying Adam as the optimization algorithm with a learning rate of 0.001. I was able to. And in the case of Adam, which is mainly applied as an optimization algorithm when designing deep learning modeling, it showed excellent performance and excellent metric results when selectively applying learning rates of 0.01 and 0.001. In the metric evaluation of test modeling, AdaGrad, which applied a learning rate of 0.1, showed the best results. Based on these results, when designing deep learning modeling for binary-based medical image classification, in order to expect quick and accurate performance, a learning rate of 0.01 is preferentially applied when applying Adam as an optimization algorithm, and a learning rate of 0.01 is preferentially applied when applying AdaGrad. I recommend doing this. In addition, it is expected that the results of this study will be presented as basic data during similar research in the future, and it is expected to be used as useful data in the health and bio industries for the purpose of automatic diagnosis of medical images using deep learning.

• Journal of Biomedical Engineering Research
• /
• v.41 no.4
• /
• pp.147-153
• /
• 2020

This study was to evaluate the usability of self-developed phantom for evaluating automatic exposure control (AEC) using three-dimensions (3D) printer. 3D printer of fused deposition modeling (FDM) type was utilized to make the self-developed AEC phantom and image acquisitions were conducted by two different type of scanners. The self-developed AEC phantom consisted of four different size of portions. As a result, two types of phantom (pyramid and pentagon shape) were created according to the combination of the layers. For evaluating the radiation dose with the two types of phantom, the values of tube current, computed tomography dose index volume (CTDIvol), and dose length product (DLP) were compared. As a result, it was confirmed that the values of tube current were properly reflected according to the thickness, and the CTDIvol and DLP were not significantly changed regardless of AEC functions of different scanners. In conclusion, the self-developed phantom by using 3D printer could assess whether the AEC function works well. So, we confirmed the possibility that a self-made phantom could replace the commercially expensive AEC performance evaluation phantom.

• Journal of Convergence Society for SMB
• /
• v.2 no.1
• /
• pp.33-40
• /
• 2012

Recently, embedded environment suffers a huge change, by growth of hardware and turning to be software-controlled. This has improved embedded software complexity. It also brought us the limit of the old development way to resolve the problem. Model-driven development is one solution to solve the limit common software development by previous way, and it became a one uses for embedded environment also. In this paper, we propose model based development approach for embedded software, witch consists of diagram editor and automatic code generator. The diagram editors are implemented by GMF, which include additional functions to solve memory restrictions and concurrent execution problems without OS environment to a automatic code generator. In order to verify the generated code, it will be tested in main control model of UAV by replacing existing module with generated one.

• Proceedings of the Korean Society of Computer Information Conference
• /
• 2011.01a
• /
• pp.107-108
• /
• 2011

Since the development of railway technology, the current urban Railway the first train line in the country for safe operation control automatic/unattended operation, automatic train operation equipment available (ATO) on time and reliable operation has introduced. ATO Automatic operation controlled by the value (Target velocity) and the feedback value (Actual velocity) by the error between the backing and braking of the train by repeated low energy efficiency. In this paper, given a fixed distance stations between time operation with minimal energy in the driving characteristics and driving trains are modeled. Therefore, in line 5 real route time sectional drive straight sections for experimental data analysis / draft Section / curved and section of the train on that line is selected according to the changing driving patterns to minimize the energy optimal driving patterns were presented.

• The Journal of the Convergence on Culture Technology
• /
• v.9 no.1
• /
• pp.347-353
• /
• 2023

Currently, soldiers enlisted in the military (Army) are receiving measurements (automatic, manual) of body parts and trying on sample clothing at boot training centers, and then receiving clothing in the desired size. Due to the low accuracy of the measured size during the measurement process, in the military, which uses a relatively more detailed sizing system than civilian casual clothes, the supplied clothes do not fit properly, so the frequency of changing the clothes is very frequent. In addition, there is a problem in that inventory is managed inefficiently by applying the measurement system based on the old generation body shape data collected more than a decade ago without reflecting the western-changed body type change of the MZ generation. That is, military uniforms of the necessary size are insufficient, and many unnecessary-sized military uniforms are in stock. Therefore, in order to reduce the frequency of clothing replacement and improve the efficiency of stock management, deep learning-based automatic measurement of body size, big data analysis, and machine learning-based "Personalized Combat Uniform Automatic Recommendation System for Enlisted Soldiers" is proposed.

• KSCE Journal of Civil and Environmental Engineering Research
• /
• v.44 no.4
• /
• pp.425-432
• /
• 2024

In this study, we propose a self-restoring friction slit damper by combining the concepts of self-restoring dampers, friction dampers, and steel dampers that are currently used and researched. For this purpose, an innovative damper structure was designed using superelastic shape memory alloy for automatic recovery and combining the concepts of friction damper and slit damper. Afterwards, detailed design was carried out and variables such as material, with of strut, and bolt fastening force were set. Modeling was performed using the ABAQUS program for a total of 12 dampers, and finite element analysis was performed by substituting the designed loading protocol. As a result, the self-recovering friction slit damper using superelastic shape memory alloy was excellent in terms of load, but the energy dissipation ability was not significantly secured due to the excellent recovery performance. However, friction slit dampers made of Gr.50 steel have dramatically improved performance in terms of load and energy dissipation through innovative structural improvements. Through this, the innovative structure of the damper, which combines the mechanisms of a friction damper and a steel damper, was demonstrated.

• Journal of the Korean Association of Geographic Information Studies
• /
• v.22 no.4
• /
• pp.229-240
• /
• 2019

3D geo-spatial model have been widely used in the field of Civil Engineering, Medical, Computer Graphics, Urban Management and many other. Especially, the demand for high quality 3D spatial information such as precise road map construction has explosively increased, MMS and UAV techniques have been actively used to acquire them more easily and conveniently in surveying and geo-spatial field. However, in order to perform 3D modeling by integrating the two data set from MMS and UAV, its so needed an proper registration method is required to efficiently correct the difference between the raw data acquisition sensor, the point cloud data generation method, and the observation accuracy occurred when the two techniques are applied. In this study, we obtained UAV point colud data in Yeouido area as the study area in order to determine the automatic registration performance between MMS and UAV point cloud data using ICP(Iterative Closet Point) method. MMS observations was then performed in the study area by dividing 4 zones according to the level of overlap ratio and observation noise with based on UAV data. After we manually registered the MMS data to the UAV data, then compared the results which automatic registered using ICP method. In conclusion, the higher the overlap ratio and the lower the noise level, can bring the more accurate results in the automatic registration using ICP method.

• The Journal of The Korea Institute of Intelligent Transport Systems
• /
• v.14 no.3
• /
• pp.9-23
• /
• 2015

A lane change maneuver which has a high driver cognitive workload and skills sometimes leads to severe traffic accidents. In this study, the Advanced Lane Change Assist System (ALCAS) was developed to assist with the automatic lane changes in merging sections which is mainly based on an automatic control algorithm for detecting an available gap, determining the Optimal Lane Change Start Point (OLCSP) in various traffic conditions, and positioning the merging vehicle at the OLCSP safely by longitudinal automatic controlling. The analysis of lane change behavior and modeling of fundamental lane change feature were performed for determining the default parameters and the boundary conditions of the algorithm. The algorithm was composed of six steps with closed-loop. In order to confirm the algorithm performance, numerical scenario tests were performed in various surrounding vehicles conditions. Moreover, feasibility of the developed system was verified in microscopic traffic simulation(VISSIM 5.3 version). The results showed that merging vehicles using the system had a tendency to find the OLCSP readily and precisely, so improved merging performance was observed when the system was applied. The system is also effective even during increases in vehicle volume of the mainline.

• Journal of Korea Water Resources Association
• /
• v.51 no.9
• /
• pp.803-812
• /
• 2018

The purpose of this study is to develop Particle Swarm Optimization (PSO) automatic calibration algorithm with multi-objective functions by Python, and to evaluate the applicability by applying the algorithm to the Soil and Water Assessment Tool (SWAT) watershed modeling. The study area is the upstream watershed of Gongdo observation station of Anseongcheon watershed ($364.8km^2$) and the daily observed streamflow data from 2000 to 2015 were used. The PSO automatic algorithm calibrated SWAT streamflow by coefficient of determination ($R^2$), root mean square error (RMSE), Nash-Sutcliffe efficiency ($NSE_Q$), and especially including $NSE_{INQ}$ (Inverse Q) for lateral, base flow calibration. The results between automatic and manual calibration showed $R^2$ of 0.64 and 0.55, RMSE of 0.59 and 0.58, $NSE_Q$ of 0.78 and 0.75, and $NSE_{INQ}$ of 0.45 and 0.09, respectively. The PSO automatic calibration algorithm showed an improvement especially the streamflow recession phase and remedied the limitation of manual calibration by including new parameter (RCHRG_DP) and considering parameters range.

• Journal of Institute of Control, Robotics and Systems
• /
• v.18 no.7
• /
• pp.688-698
• /
• 2012

This paper deals with ROLSO (Reduced Order Luenberger State Observer) design to prevent jackknifing phenomenon of articulated vehicles consisting of the tractor and semi-trailer by using GPS. In addition, by applying the regulator system using ROLSO feedback system, simulation's result presents that articulated vehicle's states are stabilized than the human's PR time (Preception Response time) rapidly. This simulation verifies that the automatic control of articulated vehicle's can be applied for the accident prevention for the time that the driver is unable to manage with the sudden accident. For this simulation, by using the equation of planar motion, the modeling of the articulated vehicle was performed. This modeling was expressed in the state space model. And FOLSO (Full Order Luenberger State Observer), ROLSO were designed by using the state space model of an articulated vehicle's dynamics.