• Journal of Advanced Marine Engineering and Technology
• /
• v.24 no.4
• /
• pp.405-413
• /
• 2000

This paper present study on the application of sound pressure and vibration signals to detect the presence of defects in a rolling element bearing using a statistical analysis method. The well established statistical parameters such as the crest factor and the distribution of moments including kurtosis and skew are utilized in this study. In addition, other statistical parameters derived from the beta distribution function are also used. A comparison study on the performance of the different types of parameter used is also performed. The statistical analysis is used because of its simplicity and quick computation. Under ideal conditions, the statistical method can be used to identify the different types of defect present in the bearing. In addition, the results also reveal that there is no significant advantages in using the beta function parameters when compared to using kurtosis and the crest factor for detecting and identifying defects in rolling element bearings from both sound and vibration signals.

• Proceedings of the KSRS Conference
• /
• 2003.11a
• /
• pp.885-887
• /
• 2003

The purpose of this investigation is to generate true orthophotos from high resolution satellite images. The major works of this research include 4 parts: (1) determination of orientation parameters, (2) generating traditional orthophotos using terrain model, (3) relief correction for buildings, and (4) process for hidden areas. To determine the position of satellites, we correct the onboard orientation parameters to fine tune the orbit. In the generation of traditional orthophotos, we employ orientation parameters and digital terrain model(DTM) to rectify tilt displacements and relief displacements for terrain. We, then, compute relief displacements for buildings with digital building model (DBM). To avoid double mapping, we detect hidden areas. Due to the satellite’s small field of view, an efficient method for the detection of hidden areas and building rectification will be proposed in this paper. Test areas cover the city of Kaohsiung in southern Taiwan. Test images are from the QuickBird satellite.

• Proceedings of the KIEE Conference
• /
• 2004.11b
• /
• pp.51-53
• /
• 2004

In this paper, we are going to propose the new algorithms to detect, classify, discriminate the transient and the reflected signal from noise and thus discriminate the fault section and locale the fault accurately on underground power cable system. Actually, at this system, it's very difficult to discriminate the transient because of the reflected signal including many noises. Therefore, how to solve the noise interference is a big problem. In this paper, authors present a solution based on multiple scales correlation of the transient using stationary wavelet transform. It's simple, quick and straightforward. For applying all algorithms, we just use the signal captured in single end.

• Journal of IKEEE
• /
• v.23 no.3
• /
• pp.1033-1037
• /
• 2019

GaN-based sensors have been widely investigated thanks to its potential in detecting the presence of hydrogen. In this study, we fabricated hydrogen gas sensors with AlGaN/GaN heterojunction and investigated how the sensing performance to be affected by SiN surface passivation. The gas sensor employed a high electron mobility transistors (HEMTs) with 30 nm platinum catalyst as a gate to detect the hydrogen presence. SiN layer was deposited by inductively-coupled chemical vapor deposition as post-passivation. The sensors with SiN passivation exhibited hydrogen sensing characteristics with various gas flow rates and concentrations of hydrogen in inert background gas at $200^{\circ}C$ similar to the ones without passivation. Aside from quick response time for both sensors, there are differences in sensitivity and recovery time because of the existence of the passivation layer. The results also confirmed the dependence of sensing performance on gas flow rate and gas concentration.

• International Journal of Computer Science & Network Security
• /
• v.21 no.6
• /
• pp.200-206
• /
• 2021

Fraud in e-commerce transaction increased in the last decade especially with the increasing number of online stores and the lockdown that forced more people to pay for services and groceries online using their credit card. Several machine learning methods were proposed to detect fraudulent transaction. Neural networks showed promising results, but it has some few drawbacks that can be overcome using optimization methods. There are two categories of learning optimization methods, first-order methods which utilizes gradient information to construct the next training iteration whereas, and second-order methods which derivatives use Hessian to calculate the iteration based on the optimization trajectory. There also some training refinements procedures that aims to potentially enhance the original accuracy while possibly reduce the model size. This paper investigate the performance of several NN models in detecting fraud in e-commerce transaction. The backpropagation model which is classified as first learning algorithm achieved the best accuracy 96% among all the models.

• Journal of the Korea Society of Computer and Information
• /
• v.15 no.2
• /
• pp.55-64
• /
• 2010

This paper proposes a face detection method using Free Rectangle feature which possesses a quick execution time and a high efficiency. The proposed mask of Free Rectangle feature is composed of two separable rectangles with the same area. In order to increase the feature diversity, Haar-like feature generally uses a complex mask composed of two or more rectangles. But the proposed feature mask can get a lot of very efficient features according to any position and scale of two rectangles on the feature window. Moreover, the Free Rectangle feature can largely reduce the execution time since it is defined as the only difference of the sum of pixels of two rectangles irrespective of the mask type. Since it yields a quick detection speed and good detection rates on real world images, the proposed face detection method based on Adaboost algorithm is easily applied to detect another object by changing the training dataset.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.6 no.1
• /
• pp.98-115
• /
• 2012

New trend called ubiquitous leads the recent business by standardization and integration. It should be the main issue how to guarantee the integration and accountability on each business, especially in mission critical system which is mainly supported by M2M (Machine to Machine) control mechanism. This study is from the analysis of digital forensics case study that is from the M2M Sensing Control Mechanism problem of the "Imjin River" case in 2009, where a group of family is swept away to death by water due to M2M control error. The ubiquitous surroundings bring the changes in the field of criminal investigation to real time controls such as M2M systems. The needs of digital forensics on M2M control are increasing on every crime scene but we suffer from the lack of control metrics to get this done efficiently. The court asks for more accurately analyzed results accounting high quality product development design. Investigators in the crime scene need real-time analysis against the crime caused by poor quality of mission critical systems. It seems to be every need of Real-Time-Enterprise, so called ubiquitous society on the case. We try to find the efficiency and productivity in discovering non-functional design defects in M2M convergence products focusing on three metrics in study model with quick implementation. Digital forensics system in present status depends on know-how of each investigator and is hard to expect professional analysis on every field. This study set up a hypothesis "Co-working of professional investigators on each field will qualify Performance and Integrity" especially in mission critical system such as M2M and suggests "Online co-work analysis model" to efficiently detect and prevent mission critical errors in advance. At the conclusion, this study proved the statistical research that was surveyed by digital forensics specialists around M2M crime scene cases with quick implementation of dash board.

• Journal of Power Electronics
• /
• v.17 no.1
• /
• pp.56-66
• /
• 2017

Fast and accurate fault detection is the primary step and one of the most important tasks in fault tolerant converters. In this paper, a fast and simple method is proposed to detect and diagnosis the faulty cell in a cascaded H-bridge multilevel inverter under a short circuit fault. In this method, the reference voltage is calculated using switching control pulses and DC-Link voltages. The comparison result of the output voltage and the reference voltage is used in conjunction with active cell pulses to detect the faulty cell. To achieve this goal, the cell which is active when the Fault signal turns to "0" is detected as the faulty cell. Furthermore, consideration of generating the active cell pulses is completely described. Since the main advantage of this method is its simplicity, it can be easily implemented in a programmable digital device. Experimental results obtained with an 11-level inverter prototype confirm the effectiveness of the proposed fault detection technique. In addition, they show that the diagnosis method is unaffected by variations of the modulation index.

• Journal of Biosystems Engineering
• /
• v.42 no.4
• /
• pp.301-313
• /
• 2017

Purpose: This study was conducted to discriminate between normal hulled barley and Fusarium (Fusarium asiaticum and Fusarium graminearum) infected hulled barley by using the near-infrared spectroscopy (NIRS) technique. Methods: Fusarium asiaticum and Fusarium graminearum were artificially inoculated in hulled barley and the reflectance spectrum of the barley spike was obtained by using a near-infrared spectral sensor with wavelength band in the range 1,175-2,170 nm. After obtaining the spectrum of the specimen, the hulled barley was cultivated in a greenhouse and visually inspected for infections. Results: From a partial least squares discriminant analysis (PLS-DA) prediction model developed from the raw spectrum data of the hulled barley, the discrimination accuracy for the normal and infected hulled barley was 99.82% (563/564) and 100% (672/672), respectively. Conclusions: NIRS is effective as a quick and nondestructive method to detect whether hulled barley has been infected with Fusarium. Further, it expected that NIRS will be able to detect Fusarium infections in other grains as well.

• Journal of the Korean Society for Aeronautical & Space Sciences
• /
• v.39 no.3
• /
• pp.225-231
• /
• 2011

The satellite clock anomalies, one of the abnormal signal factors of the GPS satellites, can have a significant impact on the GPS measurements. However, it can be difficult to detect the anomalies of the satellites clock before the range of the satellites clock error becomes bigger than the range of the other factors, due to the measurement including error of the orbit, ionosphere delay, troposphere delay, multipath and receiver clock. In order to perform quick and accurate detection by minimization of critical range in anomalies of the satellites clock, this paper suggested a solution to detect precise anomalies of the satellites clock after application of carrier smoothing filter from measurement by dual-frequency and adjustment of errors which can be occurred by other factor and the receiver clock errors. The performance of the proposed method was confirmed by comparing to the satellite clock biases which are provided by IGS.