• Journal of Korean Institute of Industrial Engineers
• /
• v.36 no.4
• /
• pp.212-218
• /
• 2010

This paper shows accuracy comparison results of reliability estimation methods for one-shot systems with respect to sample sizes. To compare accuracy in reliability estimation methods, quantal-response data, characterizing one-shot systems, were simulated using failure times of LED obtained through the accelerated life test, and then the true reliability over time was evaluated using the failure times. The simulated quantal-response data were used to estimate the true reliability through applying reliability estimation methods in open literature. Accuracy of each reliability estimation method was compared in terms of both SSE (Sum of Squared Error) and MSE (Mean Squared Error), and then estimation trend for each method is found. Feasible bounds which true reliability would exist within were estimated through applying the found trends to quantal-response data set of a real weapon system.

• Proceedings of the SAREK Conference
• /
• 2008.06a
• /
• pp.383-388
• /
• 2008

We are now developing a CFD program, AIRVIEW, with several numerical models and the SIMPLER solving method for constructing flow field and thermal comfort. This study is carried out for evaluating an accuracy of AIRVIEW. Comparisons of accuracy are carried out using Phoenics Version 3.4. In this study, we compare the turbulent kinetic energy distribution and local turbulent Re number obtained with Phoenics with those results simulated by AIRVIEW for three kinds of duct. It is observed from comparison of results that the turbulent kinetic energy values are significant due to the large velocity gradients in the region of flow. Numerical results for turbulent kinetic energy distribution and local turbulent Re number are that a good degree of agreement is found.

• Journal of IKEEE
• /
• v.19 no.4
• /
• pp.500-505
• /
• 2015

Through Montecarlo simulation, this study compares how accurately the existing three phase deviation models estimate the time-varying variance of a planar oscillator perturbed by Gaussian noises. The comparison reveals that Kaertner model estimates the time-varying variance with about 1000 times higher accuracy than ISF or PP model exhibits. Additionally, it finds that the estimation accuracy of PP model is somewhat higher than that of ISF model.

• Korean Journal of Remote Sensing
• /
• v.18 no.3
• /
• pp.163-169
• /
• 2002

The land uses of Korean peninsula are very complicated and high-density. Therefore, the image classification using coarse resolution satellite images may not provide good results for the land cover classification. The purpose of this paper is to compare the classification accuracy of visual interpretation with that of digital image classification of satellite remote sensing data such as 20m SPOT and 30m TM. In this study, hybrid classification was used. Classification accuracy was assessed by comparing each classification result with reference data obtained from KOMPSAT-1 EOC imagery, air photos, and field surveys.

• Journal of the Semiconductor & Display Technology
• /
• v.21 no.3
• /
• pp.63-68
• /
• 2022

In this study, we produced defect detection models using parallel convolution neural networks. If convolution neural networks are constructed parallel type, the model's detection accuracy will increase and detection time will decrease. We produced parallel-type defect detection models using 4 types of convolutional algorithms. The performance of models was evaluated using evaluation indicators. The model's performance is detection accuracy and detection time. We compared the performance of each parallel model. The detection accuracy of the model using AlexNet is 97 % and the detection time is 0.3 seconds. We confirmed that when AlexNet algorithm is constructed parallel type, the model has the highest performance.

• Journal of the Semiconductor & Display Technology
• /
• v.22 no.2
• /
• pp.40-44
• /
• 2023

The defect detection based on deep learning requires minimal loss and high accuracy to pinpoint product defects. In this paper, we confirm the loss rate of deep learning training based on disc-shaped artificial scaffold images. It is intended to compare the performance of Cross-Entropy functions used in object detection algorithms. The model was constructed using normal, defective artificial scaffold images and category cross entropy and sparse category cross entropy. The data was repeatedly learned five times using each loss function. The average loss rate, average accuracy, final loss rate, and final accuracy according to the loss function were confirmed.

• Proceedings of the KSRS Conference
• /
• 2002.10a
• /
• pp.388-393
• /
• 2002

Remote sensing data is collected and analyzed to enhance understanding of the terrestrial surface. Since Landsat satellite was launched in 1972, many researches using multispectral data has been achieved. Recently, with the availability of airborne and satellite hyperspectral data, the study on hyperspectral data are being increased. It is known that as the number of spectral bands of high-spectral resolution data increases, the ability to detect more detailed cases should also increase, and the classification accuracy should increase as well. In this paper, we classified the hyperspectral and multispectral data and tested the classification accuracy. The MASTER(MODIS/ASTER Airborne Simulator, 50channels, 0.4~13$\mu$m) and Landsat TM(7channels) imagery including Yeong-Gwang area were used and we adjusted the classification items in several cases and tested their classification accuracy through statistical comparison. As a result of this study, it is shown that hyperspectral data offer more information than multispectral data.

• Proceedings of the Korean Society of Precision Engineering Conference
• /
• 2005.10a
• /
• pp.436-439
• /
• 2005

In general a square master and a dial gauge are used to measure the axis squareness on the spot. This method is a comparison measurement and its accuracy depends on the square accuracy wholly. Therefore the accuracy of a square master is very important and it is impossible that the accuracy of a square measurement is superior to the accuracy of a square master. In this paper, the new method of square measurement is proposed for measuring square without a square master and easily. This method is an absolute measurement by using a reversal method and can be used to measurement the accuracy of a square master.

• Korean Journal of Digital Imaging in Medicine
• /
• v.13 no.1
• /
• pp.39-48
• /
• 2011

Uses a Tomographic scan image and Table Object Distance(TOD) price after measuring, uses accuracy and usability of blood vessel diameter(Vessel Diameter) measurement under comparison evaluating boil TOD Calibration. The patient who enforces Prosecuting Attorney abdomen Tomographic scan in the object the superior mesentery artery uses PACS View from abdomen fault image and from blood vessel diameter and the table measures the height until of the blood vessel. Uses Angio Catheter from Angiography(5 Fr.) and enforces is measured from PACS View the height until of the table which and the blood vessel at TOD Calibration price and the size of the superior mesentery artery inputs measures an superior mesentery artery building skill. Catheter Calibration input Agnio Catheter where uses in Angiography the size of the superior mesentery artery at Catheter Calibration price and they measure. Produced an accuracy from monitoring data and comparison evaluated. The statistical program used SPSS. TOD Calibration accuracy was 96.53%, standard deviation is 0.03829. Catheter Calibration accuracy of 92.91%, standard deviation is 0.05085. Represents a statistically significant difference(p = 0). According to age and gender was not statistically significant(p > 0.05). TOD Calibration correlation coefficient R-squared of 88.8%, Catheter Calibration of the R-squared is 75.5%. High accuracy of both methods. Through this study, CT images using the measured distance between the table and the Object, TOD Calibration accuracy higher than two Catheter Calibration was measured. TOD and Catheter Calibration represents a statistically significant difference(p = 0).

• Nuclear Engineering and Technology
• /
• v.54 no.8
• /
• pp.3027-3033
• /
• 2022

Background: In this study, various types of deep-learning models for predicting in vitro radiosensitivity from gene-expression profiling were compared. Methods: The clonogenic surviving fractions at 2 Gy from previous publications and microarray gene-expression data from the National Cancer Institute-60 cell lines were used to measure the radiosensitivity. Seven different prediction models including three distinct multi-layered perceptrons (MLP), four different convolutional neural networks (CNN) were compared. Folded cross-validation was applied to train and evaluate model performance. The criteria for correct prediction were absolute error < 0.02 or relative error < 10%. The models were compared in terms of prediction accuracy, training time per epoch, training fluctuations, and required calculation resources. Results: The strength of MLP-based models was their fast initial convergence and short training time per epoch. They represented significantly different prediction accuracy depending on the model configuration. The CNN-based models showed relatively high prediction accuracy, low training fluctuations, and a relatively small increase in the memory requirement as the model deepens. Conclusion: Our findings suggest that a CNN-based model with moderate depth would be appropriate when the prediction accuracy is important, and a shallow MLP-based model can be recommended when either the training resources or time are limited.