• The Journal of the Korea institute of electronic communication sciences
• /
• v.15 no.1
• /
• pp.133-138
• /
• 2020

This paper introduces a method for improving accuracy using the first convolution layer, which is not used in most modified CNN(: Convolution Neural Networks). In CNN, such as GoogLeNet and DenseNet, the first convolution layer uses only the traditional methods(3×3 convolutional computation, batch normalization, and activation functions), replacing this with RGB-csb. In addition to the results of preceding studies that can improve accuracy by applying RGB values to feature maps, the accuracy is compared with existing CNN using a limited number of images. The method proposed in this paper shows that the smaller the number of images, the greater the learning accuracy deviation, the more unstable, but the higher the accuracy on average compared to the existing CNN. As the number of images increases, the difference in accuracy between the existing CNN and the proposed method decreases, and the proposed method does not seem to have a significant effect.

• Journal of Digital Convergence
• /
• v.18 no.2
• /
• pp.285-290
• /
• 2020

This study focus on improving the text language proficiency regarding users' written text. In order to tone up accuracy improvement in writing, Computer Assisted Language Learning(CALL) can be primarily used as one of the most efficient tools. This study proposes a English Grammar Checking Application that can improve the accuracy over the current applications. The proposed system is capable of defining the difference between a Noun and a Noun Phrase which is critical in improving grammar accuracy for those who use Englilsh as a foreign language in English writing.

• International Journal of Computer Science & Network Security
• /
• v.24 no.7
• /
• pp.170-176
• /
• 2024

Since the World Health Organization (WHO) has declared COVID-19 as pandemic, many researchers have started working on developing vaccine and developing AI systems to detect COVID-19 patient using Chest X-ray images. The purpose of this work is to improve the performance of pre-trained Deep convolution neural nets (DCNNs) on Chest X-ray images dataset specially COVID-19 which is developed by collecting from different sources such as GitHub, Kaggle. To improve the performance of Deep CNNs, data augmentation is used in this study. The COVID-19 dataset collected from GitHub was containing 257 images while the other two classes normal and pneumonia were having more than 500 images each class. There were two issues whike training DCNN model on this dataset, one is unbalanced and second is the data is very less. In order to handle these both issues, we performed data augmentation such as rotation, flipping to increase and balance the dataset. After data augmentation each class contains 510 images. Results show that augmentation on Chest X-ray images helps in improving accuracy. The accuracy before and after augmentation produced by our proposed architecture is 96.8% and 98.4% respectively.

• Journal of Korean Society of Industrial and Systems Engineering
• /
• v.47 no.2
• /
• pp.1-9
• /
• 2024

New motor development requires high-speed load testing using dynamo equipment to calculate the efficiency of the motor. Abnormal noise and vibration may occur in the test equipment rotating at high speed due to misalignment of the connecting shaft or looseness of the fixation, which may lead to safety accidents. In this study, three single-axis vibration sensors for X, Y, and Z axes were attached on the surface of the test motor to measure the vibration value of vibration. Analog data collected from these sensors was used in classification models for anomaly detection. Since the classification accuracy was around only 93%, commonly used hyperparameter optimization techniques such as Grid search, Random search, and Bayesian Optimization were applied to increase accuracy. In addition, Response Surface Method based on Design of Experiment was also used for hyperparameter optimization. However, it was found that there were limits to improving accuracy with these methods. The reason is that the sampling data from an analog signal does not reflect the patterns hidden in the signal. Therefore, in order to find pattern information of the sampling data, we obtained descriptive statistics such as mean, variance, skewness, kurtosis, and percentiles of the analog data, and applied them to the classification models. Classification models using descriptive statistics showed excellent performance improvement. The developed model can be used as a monitoring system that detects abnormal conditions of the motor test.

• Proceedings of the Korean Society of Precision Engineering Conference
• /
• 2005.06a
• /
• pp.10-14
• /
• 2005

This report deals with a feeding system of the Coaxal grinding machine, processing optical ferrule. This report also examines the applicability of using the feeding system for the Coaxial grinding machine, by mean of conducting performance estimation. The results are as follow; Repeatability of regulating wheel is $17{\mu}m$, R/W rotation accuracy is between $30\;\~\;40{\mu}m$. This means 'Rotation accuracy' is lower than the concentricity level. Backlash generation level at the feeding system of the grinding wheel is under $1{\mu}m$, thereby positioning accuracy is controlled within $2{\mu}m$ In terms of repeatability, you can find occasional error at the returning process from the starting point. This error is resulted from the measurement tolerance of the starting point sensor. We will get the repeatability level under control by $1{\mu}m$, through improving the soft-ware used and up-grading the sensor at the starting point.

• Journal of Communications and Networks
• /
• v.18 no.5
• /
• pp.796-805
• /
• 2016

In many wireless sensor network (WSN) applications, the information transmitted by an individual entity or node is of limited use without the knowledge of its location. Research in node localization is mostly geared towards multi-hop range-free localization algorithms to achieve accuracy by minimizing localization errors between the node's actual and estimated position. The existing localization algorithms are focused on improving localization accuracy without considering efficiency in terms of energy costs and algorithm convergence time. In this work, we show that our proposed localization scheme, called DV-maxHop, can achieve good accuracy and efficiency. We formulate the multi-objective optimization functions to minimize localization errors as well as the number of transmission during localization phase. We evaluate the performance of our scheme using extensive simulation on several anisotropic and isotropic topologies. Our scheme can achieve dual objective of accuracy and efficiency for various scenarios. Furthermore, the recently proposed algorithms require random uniform distribution of anchors. We also utilized our proposed scheme to compare and study some practical anchor distribution schemes.

• Journal of the Korean Society for Precision Engineering
• /
• v.16 no.2 s.95
• /
• pp.201-210
• /
• 1999

One of the major limitations of productivity and quality in metal cutting is the machining accuracy of machine tools. The machining accuracy is affected by geometric and thermal errors of the machine tools. Therefore, a key requirement for improving te machining accuracy and product quality is to reduce the geometric and thermal errors of machine tools. This study models geometric error for error analysis and develops on-machine measurement system by which the volumetric erors are measured. The geometric error is modeled using form shaping function(FSF) which is defined as the mathematical relationship between form shaping motion of machine tool and machined surface. The constant terms included in the error model are found from the measurement results of on-machine measurement system. The developed on-machine measurement system consists of the spherical ball artifact (SBA), the touch probe unit with a star type stylus, the thermal data logger and the personal computer. Experiments, performed with the developed measurement system, show that the system provides a high measuring accuracy, with repeatability of ${\pm}2{\mu}m$ in X, Y and Z directions.

• Journal of Sensor Science and Technology
• /
• v.28 no.6
• /
• pp.414-419
• /
• 2019

An ultrasonic based magnetostrictive position sensor (MPS) provides an indication of real target position. It determines the real target position by multiplying the propagation speed of ultrasonic wave and the time-of-flight between the receiving signals; one is the initial signal by an excitation current and the other is the reflection signal by the ultrasonic wave. The propagation speed of the ultrasonic wave depends on the temperature of the waveguide. Hence, the change of the propagation speed in various environments is a critical factor in terms of the positioning accuracy in the MPS. This means that the influence of the changes in the waveguide temperature needs to be compensated. In this paper, we presents a novel way to improve the positioning accuracy of MPSs using temperature compensation for waveguide. The proposed method used the inherent measurement blind area for the structure of the MPS, which can simultaneously measure the position of the moving target and the temperature of the waveguide without any additional devices. The average positional error was approximately -23.9 mm and -1.9 mm before and after compensation, respectively. It was confirmed that the positioning accuracy was improved by approximately 93%.

• Journal of the Korea Institute of Military Science and Technology
• /
• v.21 no.4
• /
• pp.471-480
• /
• 2018

Accurate target coordinate is very important in military operations especially field artillery's ground-to-ground attack and air-force's air-to-ground attack. DOS(or TAS) is used to acquire target coordinates from long distance. DOS is comprised of LRF and goniometer. LRF measures distance between DOS and target. Goniometer is comprised of azimuth and vertical angular sensors, DMC and internal GPS receiver. DOS must set the position and orientation(finding grid north) before measurement step(target coordinate acquisition). To improve accuracy of target coordinate, VRS RTK and reference point method are proposed in DOS setup step. VRS RTK provides accurate location coordinate with small deviations, providing high accuracy and precision in positioning and orientation. As a result, horizontal coordinate(easting and northing) accuracy is improved from 2.68 mil(C.L. = 0.95) mil to 0.58 mil(C.L. = 0.95).

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.11 no.12
• /
• pp.6152-6168
• /
• 2017

Recently, the biometric-based authentication systems such as FIDO (Fast Identity Online) are increased in mobile computing environments. The biometric-based authentication systems are performed on the mobile devices with the battery, the improving energy efficiency is important issue. In the case, the size of images (i.e., face, fingerprint, iris, and etc.) affects both recognition accuracy and energy consumption, and hence the tradeoff analysis between the both recognition accuracy and energy consumption is necessary. In this paper, we propose an energy-efficient way to authenticate based on biometric information with tradeoff analysis between the both recognition accuracy and energy consumption in multimedia IoT (Internet of Things) transmission environments. We select the facial information among biometric information, and especially consider the multicore-based mobile devices. Based on our experimental results, we prove that the proposed approach can enhance the energy efficiency of GABOR+LBP+GRAY VALUE, GABOR+LBP, GABOR, and LBP by factors of 6.8, 3.6, 3.6, and 2.4 over the baseline, respectively, while satisfying user's face recognition accuracy.