• Korean Journal of Remote Sensing
• /
• v.37 no.4
• /
• pp.747-761
• /
• 2021

With the increasing application of UAVs in intelligent transportation systems, vehicle detection for aerial images has become an essential engineering technology and has academic research significance. In this paper, a vehicle detection method for aerial images based on the YOLOv4 deep learning algorithm is presented. At present, the most known datasets are VOC (The PASCAL Visual Object Classes Challenge), ImageNet, and COCO (Microsoft Common Objects in Context), which comply with the vehicle detection from UAV. An integrated dataset not only reflects its quantity and photo quality but also its diversity which affects the detection accuracy. The method integrates three public aerial image datasets VAID, UAVD, DOTA suitable for YOLOv4. The training model presents good test results especially for small objects, rotating objects, as well as compact and dense objects, and meets the real-time detection requirements. For future work, we will integrate one more aerial image dataset acquired by our lab to increase the number and diversity of training samples, at the same time, while meeting the real-time requirements.

• Nuclear Engineering and Technology
• /
• v.55 no.12
• /
• pp.4350-4356
• /
• 2023

Radiation detection systems working at high count rates suffer from the overlapping of their output electric pulses, known as pulse pile-up phenomenon, resulting in spectrum distortion and degradation of the energy resolution. Pulse tail extrapolation is a pile-up correction method which tries to restore the shifted baseline of a piled-up pulse by extrapolating the overlapped part of its preceding pulse. This needs a mathematical model which is almost always nonlinear, fitted usually by a nonlinear least squares (NLS) technique. NLS is an iterative, potentially time-consuming method. The main idea of the present study is to replace the NLS technique by an integration-based non-iterative method (NIM) for pulse tail extrapolation by an exponential model. The idea of linear extrapolation, as another non-iterative method, is also investigated. Analysis of experimental data of a NaI(Tl) radiation detector shows that the proposed non-iterative method is able to provide a corrected spectrum quite similar with the NLS method, with a dramatically reduced computation time and complexity of the algorithm. The linear extrapolation approach suffers from a poor energy resolution and throughput rate in comparison with NIM and NLS techniques, but provides the shortest computation time.

• Structural Engineering and Mechanics
• /
• v.14 no.4
• /
• pp.455-472
• /
• 2002

The paper involves the study on the elastic and elasto-plastic stress wave propagation in the 1-D and 2-D solid media. The Smooth Particle Hydrodynamics equations governing the elastic and elasto-plastic large deformation dynamic response of solid structures are presented. The proposed additional stress points are introduced in the formulation to mitigate the tensile instability inherent in the SPH approach. Both incremental rate approach and leap-frog algorithm for time integration are introduced and the new solution algorithm is developed and implemented. Two examples on stress wave propagation in aluminium bar and 2-D elasto-plastic steel plate are included. Results from the proposed SPH approach are compared with available analytical values and finite element solutions. The comparison illustrates that the stress wave propagation problems can be effectively solved by the proposed SPH method. The study shows that the SPH simulation is a reliable and robust tool and can be used with confidence to treat transient dynamics such as linear and non-linear transient stress wave propagation problems.

• Journal of KIISE:Databases
• /
• v.33 no.7
• /
• pp.715-726
• /
• 2006

Web service technologies provide a standard means for interoperation and integration of heterogeneous applications distributed over the Internet. For efficient execution of hierarchically interacting composite web services, this paper proposes an approach to distribute web service invocations over peer systems effectively, exploiting intensional XML data embedding external service calls as a result of well services. A cost-based optimization problem on the execution of web services using intensional results was formalized, and a heuristic search method to find an optimal solution and a greedy algorithm to generate an efficient invocation plan quickly were suggested in this paper. Experimental evaluation shows that the proposed greedy algorithm provides near-optimal solutions in an acceptable time even for a large number of Web services.

• Journal of Ocean Engineering and Technology
• /
• v.34 no.4
• /
• pp.272-276
• /
• 2020

This paper presents an optimization process for a catamaran hull form. The entire optimization process was managed using the CAD-CFD integration platform CAESES. The resistance of the demi-hull was simulated in calm water using the CFD solver STAR-CCM+, and an inviscid fluid model was used to reduce the computing time. The Free-Form Deformation (FFD) method was used to make local changes in the bulbous bow. For the optimization of the bulbous bow, the Non-dominated Sorting Genetic Algorithm (NSGA)-II was applied, and the optimization variables were the length, breadth, and angle between the bulbous bow and the base line. The Lackenby method was used for global variation of the bow of the hull. Nine hull forms were generated by moving the center of buoyancy while keeping the displacement constant. The optimum bow part was selected by comparing the resistance of the forms. After obtaining the optimum demi-hull, the distance between two demi-hulls was optimized. The results show that the proposed optimization sequence can be used to reduce the resistance of a catamaran in calm water.

• Journal of Platform Technology
• /
• v.6 no.1
• /
• pp.9-16
• /
• 2018

In recent years, many analysts have been using Python and R to analyze data using efficient analysis and statistics. However, it is difficult to solve the requirement to express various analyzes in real time through the web system using only two languages. In addition, it is possible to confirm that the extracted P-Value is different from the value in R because there is no corresponding algorithm in Java or the corresponding source is not verified. Therefore, in this paper, we propose a method to apply analyst 's analysis algorithm contents directly to his system using R in other languages.

• Journal of the Korean Society for Aeronautical & Space Sciences
• /
• v.49 no.10
• /
• pp.849-858
• /
• 2021

It is difficult to establish a quantitative standard because there are many factors to consider, such as environmental conditions, airworthiness, and maintainability, in determining the installation location of equipment in an aircraft. In addition, as the number of equipment increases, the design proposal increases exponentially, so the design is proceeding depending on the experience of the designer much in order to review it within a limited time schedule. In this paper, a method of calculating the length and weight of the wiring harness according to the location of the equipment and a method of optimizing the weight of the wiring harness and the CG of the equipment using genetic algorithms are described in order to create a quantitative standard useful by comparing the optimal design and the actual design.

• Journal of KIBIM
• /
• v.14 no.2
• /
• pp.1-12
• /
• 2024

The paper proposes a methodology that combines the Random Sample Consensus (RANSAC) algorithm and the Point Cloud Encoder-Decoder Network (PCEDNet) algorithm to automatically extract the length of infrastructure elements from point cloud data acquired through 3D LiDAR scans of retaining walls. This methodology is expected to significantly improve time and cost efficiency compared to traditional manual measurement techniques, which are crucial for the data-driven analysis required in the precision-demanding construction sector. Additionally, the extracted positional and dimensional data can contribute to enhanced accuracy and reliability in Scan-to-BIM processes. The results of this study are anticipated to provide important insights that could accelerate the digital transformation of the construction industry. This paper provides empirical data on how the integration of digital technologies can enhance efficiency and accuracy in the construction industry, and offers directions for future research and application.

• Journal of the Institute of Electronics Engineers of Korea TE
• /
• v.37 no.5
• /
• pp.48-55
• /
• 2000

Recently due to rapid improvement of integrated circuit design environment, size of IC design is to become large to possible design System on Chip(SoC) that one chip with multi function enclosed. Also cause to this rapid change, consumption market is require to spend smallest time for new product development. In this paper to propose a methodology can design a large size IC for save time and applied to design of MPEG Layer 2 decoder to can use audio receiver in digital broadcast system. The digital broadcast audio decoder in this paper is pointed to save hardware size as optimizing algorithm. MPEG Layer 2 decoder algorithm is include MAC to can have an effect on hardware size. So coefficients are using sign digit expression. It is for hardware optimization. If using this method can design MAC without multiplier. The designed audio decoder is using 14,000 gates hardware size and save 22% (4000 gates) hardware usage than using multiplier. Also can design MPEG Layer 2 decoder usable digital broadcast receiver for short time.

• International Journal of Aeronautical and Space Sciences
• /
• v.10 no.1
• /
• pp.75-82
• /
• 2009

In this parer, we propose a new way of improving DGNSS service using combination of multiple SBAS information. Because SBAS uses Geostationary Earth Orbit (GEO) satellites, it has very large coverage but it can be unavailable in urban canyon because of visibility problem. R. Chen solved this problem by creating Virtual Reference Stations (VRS) using the SBAS signal [1]. VRS converts SBAS signal to RTCM signals corresponding its location, and broadcast the converted RTCM signals over the wireless internet. This method can solve the visibility problem cost effectively. Furthermore it can solve DGNSS coverage problem by creating just a transmitter instead of a reference station. Developing above method, this paper proposes the methods that integrate two or more SEAS signals into one RTCM signal and broadcast it. In Korea, MSAS signal is available even though it is not officially certified for Korean users. As a Korean own SBAS-like system, there is the internet-based KWTB (Korean WADGPS Test Bed) which we developed and released at ION GNSS 2006. As a result, virtually two different SBAS corrections are available in Korea. In this paper, we propose the integration methods for these two independent SBAS corrections and present the test results using the actual measurements from the two systems. We present the detailed algorithm for these two methods and analyze the features and performances of them. To verify the proposed methods, we conduct the experiment using the logged SBAS corrections from the two systems and the RINEX data logged at Dokdo monitoring station in Korea. The preliminary test results showed the improved performance compared to the results from two independent systems, which shows the potential of our proposed methods. In the future, the newly developed SBASs will be available and the places which can access the multiple SBAS signals will increase. At that time, the integration or combination methods of two or more SBASs will become more important. Our proposed methods can be one of the useful solutions for that. As an additional research, we need to extend this research to the system level integration such as the concept of the decentralized W ADGPS.