• Journal of the Semiconductor & Display Technology
• /
• v.22 no.4
• /
• pp.166-171
• /
• 2023

This paper presents the synaptic characteristics of IGZO memristors in neuromorphic computing, using MATLAB/Simulink and NeuroSim. In order to investigate the variations in the conductivity of IGZO memristor and the corresponding changes in the hidden layer, simulations are conducted by using the MNIST dataset. It was observed from simulation results that the recognition accuracy could be dependent on various parameters of IGZO memristor, along with the experimental exploration. Moreover, we identified optimal parameters to achieve high accuracy, showing an outstanding accuracy of 96.83% in image classification.

• Proceedings of the Korean Society of Precision Engineering Conference
• /
• 2000.11a
• /
• pp.330-333
• /
• 2000

Dimensional accuracy and surface quality are very important in rapid prototyping especially when the models are used for the production of tools. This paper presents the development of benchmarking part to investigate dimensional accuracy and surface finish. A new test part is designed to perform benchmarking of major rapid prototyping processes such as selective laser sintering, laminated object manufacturing, stereolithography apparatus, and fused deposition modeling. The test part design includes basic manufacturing features such as holes, walls, squares, cylinder and etc. In addition, the small features are included in order to evaluate the fine details that can be manufactured by a specific RP process. The CMM program that automatically measures different features in the test part is also developed. The evaluation of accuracy as well as surface roughness are discussed for major rapid prototyping processes.

• Annual Conference of KIPS
• /
• 2021.05a
• /
• pp.358-361
• /
• 2021

In order to improve the accuracy of small target detection more effectively, this paper proposes an improved single shot detector (SSD) target detection and recognition method based on cspdarknet53, which introduces lightweight ECA attention mechanism and Feature Pyramid Network (FPN). First, the original SSD backbone network is replaced with cspdarknet53 to enhance the learning ability of the network. Then, a lightweight ECA attention mechanism is added to the basic convolution block to optimize the network. Finally, FPN is used to gradually fuse the multi-scale feature maps used for detection in the SSD from the deep to the shallow layers of the network to improve the positioning accuracy and classification accuracy of the network. Experiments show that the proposed target detection algorithm has better detection accuracy, and it improves the detection accuracy especially for small targets.

• Journal of Advanced Marine Engineering and Technology
• /
• v.20 no.4
• /
• pp.90-100
• /
• 1996

To maintain high spacial accuracy and rapid CPU time in interpolating data from grid to random position or inversely in PIV, proposed many technuques are compared and discussed mainly in terms of interpolating error and computing time. And artificial PIV atmosphere data is furnished by CFD result. First, for interpolation from grid to random position, multiquadric method gives the highest accuracy with the longest CPU time and Taylor series expansion methods give reasonable accuracy with less calculating load. Secondly, the sub-pixel resolution analysis in estimating the coordinates of the maximum correlation coefficients essential in the grey level correlation PIV reveal that 8-neighbours 2nd-order least square interpolation gives utmost accuracy in terms of the real flow conditions.

• Journal of Advanced Marine Engineering and Technology
• /
• v.20 no.4
• /
• pp.412-412
• /
• 1996

To maintain high spacial accuracy and rapid CPU time in interpolating data from grid to random position or inversely in PIV, proposed many technuques are compared and discussed mainly in terms of interpolating error and computing time. And artificial PIV atmosphere data is furnished by CFD result. First, for interpolation from grid to random position, multiquadric method gives the highest accuracy with the longest CPU time and Taylor series expansion methods give reasonable accuracy with less calculating load. Secondly, the sub-pixel resolution analysis in estimating the coordinates of the maximum correlation coefficients essential in the grey level correlation PIV reveal that 8-neighbours 2nd-order least square interpolation gives utmost accuracy in terms of the real flow conditions.

• Journal of the Korea Safety Management & Science
• /
• v.12 no.1
• /
• pp.127-136
• /
• 2010

In this paper, the author proposed following two methods to improve the accuracy of the recommender system. First, in order to classify the users more accurately, the author used a EMC(Expanded Moving Center) heuristic algorithm which improved clustering accuracy. Second, the author proposed the Neighborhood-oriented preference prediction method that improved the conventional preference prediction methods, so the accuracy of the recommender system is improved. The test result of the recommender system which adapted the above two methods suggested in this paper was improved the accuracy than the conventional recommendation methods.

• Journal of radiological science and technology
• /
• v.44 no.4
• /
• pp.367-373
• /
• 2021

A quality assurance of computed tomography(CT) have done seven items that were water attenuation coefficient, noise, homogeneity, spatial resolution, contrast resolution, slice thickness, artifact using by standard phantom. But there is no quality assurance items and methods for CT simulator at domestic institutions yet. Therefore the study aimed to access the CT dose index(CTDI), table tilting, image distortion, laser accuracy, table movement accuracy and CT seven items for CT simulator quality assurance. The CTDI at the center of the head phantom was 0.81 for 80 kVp, 1.55 for 100 kVp, 2.50 for 120 mm, 0.22 for 80 kVp at the center of the body phantom, 0.469 for 100 kVp, and 0.81 for 120 kVp. The table tilting was within the tolerance range of ±1.0° or less. Image distortion had 1 mm distortion in the left and right images based on the center, and the laser accuracy was measured within ±2 mm tolerance. The purpose of this study is to improve the quality assurance items suitable for the current situation in Korea in order to protect the normal tissues during the radiation treatment process and manage the CT simulator that is implemented to find the location of the tumor more clearly. In order to improve the accuracy of the CT simulator when looking at the results, the error range of each item should be small. It is hoped that the quality assurance items of the CT simulator will be improved by suggesting the quality assurance direction of the CT simulator in this study, and the results of radiation therapy will also improve.

• Magazine of the Korean Society of Agricultural Engineers
• /
• v.19 no.4
• /
• pp.4533-4543
• /
• 1977

The purpose of this study was to investigate and compare flow discharges of rectangular, V-notch and trapezoidal type of cutthrooat flumes, and the published data for trapezoidal parshall flumes. And the trapezoidal cutthroat flumes were also compared in their accuracy of discharge measurements for various convergence ratios in the inlet section and divergence ratios in the outlet section. Five flumes were studied, and all the flumes were 45cm long with flat-bottom and were made of well-finished transparent acryl plate of 3mm thickness. One rectangular, one V-notch and three trapezoidal types were numbered 1 to 5 as shown in Fig. III-1. The measured depth of water was ranged from 5 to 20cm. The results obtained in this study are summarized as follows: 1. The general discharge equations for tested prototypes are listed for free flow in Table IV-1 and for submergence flow in Table IV-4. 2. In both free and submerged flow, the accuracy of the discharge formula obtained by this test is highly significant at 1% level as shown in Table IV-2 and Table IV-6. The accuracy of disharges measured depends upon the convergence and divergence ratios in the trapezoidal types: the less the ratios of convergence as well as divergence, the lower the accuracy. 3. Submergence ratios tend to increase in the order of flume number except flume No. 4. This implies that trapezoidal cutthroat flumes are more acceptable than rectangular or V-notch ones for free flow. 4. The transition submergence for the trapezoidal Parshall flumes ranges from 80-85 percent, which is slightly higher than the tested flume. However, the trapezoidal cutthroat flume No. 5 has higher transition submergence ratio, ranging from 73-78 percent, than other trapezoidal ones. The difference between the trapezoidal Parshall flumes and the trapezoidal cutthroat flumes in transition submergence seems small enough to be ignored in their field use. 5. Trapezoidal cutthroat flume is simple and economical to construct in existing openchannels whose shapes are generally trapezoidal. In order to obtain the best rating accuracy, flume No. 3 among the tested trapezoidal types is recommended, because it shows the highest accuracy for both free and submerged flow.

• KSCE Journal of Civil and Environmental Engineering Research
• /
• v.39 no.1
• /
• pp.239-246
• /
• 2019

The drone photogrammetry is actively utilized for obtaining highly accurate spatial information and other various monitoring purposes. It is general to plan the drone photogrammetry by referring to previous experiences or cases in order to obtain the required accuracy, but the drone photogrammetry is often carried out again due to poor accuracy. Since the required spatial accuracy of the drone photogrammetry process result becomes the means of objective evaluation regardless of the type of result, it should be determined carefully. Therefore, it is necessary to determine flight height, overlap, number and arrangement of ground control point, and exterior orientation factor acquisition method in order to meet the required 3D positional accuracy for the design of drone photogrammetry project. In this study, previous study cases for the analysis of drone photogrammetry accuracy were carefully analyzed and verified by applying such cases to testing area, and design guideline of drone photogrammetry project for a small area was prepared based on the analysis result. The presented project design guideline is expected to be a great help to business practice although it is not perfect, and if the design guideline is prepared through comprehensive analysis in future, it would be possible to provide a perfect manual.

• Journal of the Korean Society of Surveying, Geodesy, Photogrammetry and Cartography
• /
• v.28 no.6
• /
• pp.573-578
• /
• 2010

Precise determination of satellite positions is necessary to improve positioning accuracy in GNSS. In this study, GLONASS orbits were predicted from broadcast ephemeris using the 4th-order Runge-Kutta numerical integration method and their accuracy dependence on the integration step and the integration time was analyzed. The 3D RMS (Root Mean Square) differences between the results from I-second integration step and 300-second integration step was about 3 cm, but the processing time was one hundred times less for the I-second integration time case. For trials of different integration times, the 3D RMS errors were 8.3 m, 187.3 m, and 661.5 m for 30-, 150-, and 300-minutes of integration time, respectively. Though this integration-time analysis, we concluded that the accuracy gets higher with a shorter integration time. Thus we suggest forward and backward integration methods to improve GLONASS positioning accuracy, and with this method we can achieve a 5-meter level of 3-D orbit accuracy.