• Journal of The Korean Society of Agricultural Engineers
• /
• v.64 no.1
• /
• pp.15-26
• /
• 2022

3D open-field farm model developed from UAV (Unmanned Aerial Vehicle) data could make crop monitoring easier, also could be an important dataset for various fields like remote sensing or precision agriculture. It is essential to separate crops from the non-crop area because labeling in a manual way is extremely laborious and not appropriate for continuous monitoring. We, therefore, made a 3D open-field farm model based on UAV images and developed a crop segmentation model using a supervised machine learning algorithm. We compared performances from various models using different data features like color or geographic coordinates, and two supervised learning algorithms which are SVM (Support Vector Machine) and KNN (K-Nearest Neighbors). The best approach was trained with 2-dimensional data, ExGR (Excess of Green minus Excess of Red) and z coordinate value, using KNN algorithm, whose accuracy, precision, recall, F1 score was 97.85, 96.51, 88.54, 92.35% respectively. Also, we compared our model performance with similar previous work. Our approach showed slightly better accuracy, and it detected the actual crop better than the previous approach, while it also classified actual non-crop points (e.g. weeds) as crops.

• Journal of the Institute of Electronics Engineers of Korea SP
• /
• v.41 no.6
• /
• pp.175-186
• /
• 2004

We present a hybrid approach of texture-based method and connected component (CC)-based method for text extraction in complex images. Two primary methods, which are mainly utilized in this area, are sequentially merged for compensating for their weak points. An automatically constructed MLP-based texture classifier can increase recall rates for complex images with small amount of user intervention and without explicit feature extraction. CC-based filtering based on the shape information using NMF enhances the precision rate without affecting overall performance. As a result, a combination of texture and CC-based methods leads to not only robust but also efficient text extraction. We also enhance the processing speed by adopting appropriate region marking methods for each input image category.

• Journal of the Korean Society for Precision Engineering
• /
• v.31 no.9
• /
• pp.777-782
• /
• 2014

Usually, railway axles are designed for infinite life based on endurance limit of the material and the axle is not fractured immediately when a surface crack initiated. The railway axles have been inspected regularly by NDT such as ultrasonic testing, magnetic testing and eddy current testing and so on. Because the axle failure is profoundly influenced by the probability of missing a fatigue crack during an NDT inspection, it is necessary to evaluate the Non Destructive Interval of railway axle. In the present paper, the Non Destructive Interval of railway axle based on fracture mechanics and finite element analysis was investigated. It was shown that the Non Destructive Interval of railway axle can be evaluated using fracture mechanics approach and extended using NDT which a crack can detect clearly.

• Journal of Internet Computing and Services
• /
• v.24 no.6
• /
• pp.1-11
• /
• 2023

Federated learning (FL) is a ground breaking machine learning paradigm that allow smultiple participants to collaboratively train models in a cloud environment, all while maintaining the privacy of their raw data. This approach is in valuable in applications involving sensitive or geographically distributed data. However, one of the challenges in FL is dealing with heterogeneous and non-independent and identically distributed (non-IID) data across participants, which can result in suboptimal model performance compared to traditionalmachine learning methods. To tackle this, we introduce FedGCD, a novel FL algorithm that employs Graph Neural Network (GNN)-based community detection to enhance model convergence in federated settings. In our experiments, FedGCD consistently outperformed existing FL algorithms in various scenarios: for instance, in a non-IID environment, it achieved an accuracy of 0.9113, a precision of 0.8798,and an F1-Score of 0.8972. In a semi-IID setting, it demonstrated the highest accuracy at 0.9315 and an impressive F1-Score of 0.9312. We also introduce a new metric, nonIIDness, to quantitatively measure the degree of data heterogeneity. Our results indicate that FedGCD not only addresses the challenges of data heterogeneity and non-IIDness but also sets new benchmarks for FL algorithms. The community detection approach adopted in FedGCD has broader implications, suggesting that it could be adapted for other distributed machine learning scenarios, thereby improving model performance and convergence across a range of applications.

• International Journal of Precision Engineering and Manufacturing
• /
• v.3 no.4
• /
• pp.54-63
• /
• 2002

In recent years, mechanical systems such as high speed vehicles and railway trains moving on elastic beam structures have become a very important issue to consider. In this paper, a general approach, which can predict the dynamic behavior of a constrained mechanical system moving on a flexible beam structure, is proposed. Various supporting conditions for the foundation support are considered for the elastic beam structure. The elastic structure is assumed to be a non-uniform and linear Bernoulli-Euler beam with a proportional damping effect. Combined differential-algebraic equation of motion is derived using the multi-body dynamics theory and the finite element method. The proposed equations of motion can be solved numerically using the generalized coordinate partitioning method and predictor-corrector algorithm, which is an implicit multi-step integration method.

• Proceedings of the Korean Society of Precision Engineering Conference
• /
• 2006.05a
• /
• pp.411-412
• /
• 2006

It is very difficult to analyze the dynamic characteristic because railway vehicle is a very complex system which are connected various mass element with railway vehicle system. To realize and analyze actual phenomenon has restriction that usual commercial software calculates creep force under creep theory about wheel-rail contact mechanism as basic analyzing, and approach about contact point are based on two dimensional non-linear contact theory and simplified Hertzian contact which considers just displacement change on the YZ plain. Therefore, to solve these problems there should be a new approach difference with existing one. In this research, a new algorithm for finding wheel-rail contact position, calculation method of contact force and applied force will be presented.

• Journal of the Korean Society for Precision Engineering
• /
• v.26 no.12
• /
• pp.62-68
• /
• 2009

In this paper, we investigate a new method to measure the thickness of thin transparent objects utilizing a step index multi-mode optical fiber sensor. The method mainly depends on the refraction rate of transparent target, the diameter of optical fibers and the distance to reflector. We confirmed the effects of these parameters through the experimental verification tests. The comparison between the theoretical vs. analytical results shows good agreements with each other. The proposed model also enables users to measure the thickness of thin transparent objects without considering the reflection from the target. This approach provides simple, cost-effective and non-contact solutions to measure the thickness.

• Proceedings of the Korean Society of Precision Engineering Conference
• /
• 2005.10a
• /
• pp.24-27
• /
• 2005

The NDT(Non-Destructive Testing) is valid fur the defect detection of rolling stocks because it can be used to detect defects in invisible places. For example, in case of wheelsets fatigue cracks are initiated in the wheel seat that suffers from fretting fatigue damage. But the conventional ICFPD method can not be applied to detect such cracks in press-fit area of the axle by some technical problems. In this study, we introduced a new ICFPD (Induced Current Focusing Potential Drop) method that can be applied in press-fit area of the axle. And we performed the finite element analysis of the new ICFPD method using measured electromagnetic properties of the wheel and axle. It seems that our approach is very useful f3r the detection of defects in invisible places.

• Proceedings of the Korean Society of Precision Engineering Conference
• /
• 2003.06a
• /
• pp.1283-1287
• /
• 2003

Reverse engineering technology refers to the process that creates a CAD model of an existing part using measuring devices. Recently, non-contact scanning devices have become more accurate and the speed of data acquisition has increased drastically. However, they generate thousands of points per second and various types of point data. Therefore. it becomes a important to handle the huge amount and various types of point data to generate a surface model efficiently. This paper proposes a new triangular mesh generation method using 3D grids. The geometric information of a part can be obtained from point cloud data by estimating normal values of the points. In our research, the non-uniform 3D grids are generated first for feature based data reduction based on the geometric information. Then, triangulation is performed with the reduced point data. The grid structure is efficiently used not only for neighbor point search that can speed up the mesh generation process but also for getting surface connectivity information to result in same topology surface with the point data. Through this integrated approach, it is possible to create surface models from scanned point data efficiently.

• Korean Journal of Head & Neck Oncology
• /
• v.39 no.1
• /
• pp.1-9
• /
• 2023

Technological advancement in human genome analysis and ICT (information & communication technologies) brought 'precision medicine' into our clinical practice. Precision medicine is a novel medical approach that provides personalized treatments tailored to each individual by precisely segmenting patient populations, based on robust data including a person's genetic information, disease information, lifestyle information, etc. Precision medicine has a potential to be applied to treating a range of tumors, in addition to non-small cell lung cancer, in which precision oncology has been actively practiced. In this article, we are reviewing precision medicine in head and neck cancer (HNC) with focus on tumor agnostic biomarkers and treatments such as NTRK, MSI-H/dMMR, TMB-H and BRAF V600E, all of which were recently approved by U.S. Food and Drug Administration (FDA).