• Proceedings of the Korean Society of Precision Engineering Conference
• /
• 2002.10a
• /
• pp.1037-1040
• /
• 2002

In micro-stereolithography technology, fabrication conditions that include laser power, laser scan speed, laser scan pitch, and material property of photopolymer such as penetration depth and critical exposure are considered as major process variables. But the existing scan path generation methods based only on CAD model have not taken them into account, which has resulted in cross-section dimension of low accuracy. Thus, to enhance cross-section dimensional accuracy, the physical resin solidification n phenomena should be reflected in laser scan path generation and stage operating code. In this paper, multi-line experiments based on single line solidification model are performed. And the method for improving cross-section dimensional accuracy is presented, which is to apply the database based on experimental results to laser scan path generation.

• Journal of the Korean Society for Library and Information Science
• /
• v.29
• /
• pp.305-343
• /
• 1995

The purposes of this study are to confirm the applicability of the unobtrusive evaluation to Korean libraries, to develop survey forms, check lists and evaluation criteria appropriate for Korean libraries, to analyze the relationship between the test variables and the performance level, to find out the causes of poor service if any, and to propose measures for improving the quality of reference services. The main results of the analyses are as follows : $\cdot$ The accuracy of answers was correlated to appropriateness of search process, way of providing answer, usage of language, approachability and use of second materials in the order named. That is, there was a strong correlation between accuracy and the variables related to librarian's behavior. $\cdot$ There was no statistically significant relationship between accuracy and the environmental aspects of reference department such as number of professional librarians, self-evaluation of reference collection, etc. $\cdot$ The main reasons for failing to give accurate answers were found to be weaknesses in understanding questions. developing search strategy logically and relating the querries to available information sources.

• Proceedings of the KIEE Conference
• /
• 1998.07b
• /
• pp.555-559
• /
• 1998

The human chromosome analysis is widely used to diagnose genetic disease and various congenital anomalies. Many researches on automated chromosome karyotype analysis have been carried out, some of which produced commercial systems. However, there still remains much room for improving the accuracy of chromosome classification. In this paper, We proposed an optimal pattern classifier by neural network to improve the accuracy of chromosome classification. The proposed pattern classifier was built up of two-step multi-layer neural network(TMANN). We reconstructed chromosome image to improve the chromosome classification accuracy and extracted four morphological features parameters such as centromeric index (C.I.), relative length ratio(R.L.), relative area ratio(R.A.) and chromosome length(C.L.). These Parameters employed as input in neural network by preprocessing twenty human chromosome images. The experiment results shown that the chromosome classification error was reduced much more than that of the other classification methods.

• International Journal of Ocean Engineering and Technology Speciallssue:Selected Papers
• /
• v.5 no.1
• /
• pp.40-45
• /
• 2002

The calculation of earthwork plays a major role in the planning and design phases of many civil engineering projects, such as seashore reclamation; thus, improving the accuracy of earthwork calculation has become very important. In this paper, we propose an algorithm for finding a cubic spline surface with the free boundary conditions, which interpolates the given three-dimensional data, by using B-spline and an accurate method to estimate pit-excavation volume. The proposed method should be of interest to surveyors, especially those concerned with accuracy of volume computations. The mathematical models of the conventional methods have a common drawback: the modeling curves form peak points at the joints. To avoid this drawback, the cubic spline polynomial is chosen as the mathematical model of the new method. In this paper, we propose an algorithm of finding a spline surface, which interpolates the given data, and an appropriate method to calculate the earthwork. We present some computational results that show the proposed method, of the Maple program, provides better accuracy than the method presented by Chen and Lin.

• Journal of international Conference on Electrical Machines and Systems
• /
• v.2 no.2
• /
• pp.184-189
• /
• 2013

This paper proposes a method for improving the performance of a position sensorless control system for a slotless surface permanent magnet synchronous motor (SPMSM) in a very low-speed region. In position sensorless control based on a motor model, accurate motor parameters are required because parameter errors would affect position estimation accuracy. Therefore, online parameter identification is applied in the proposed system. The error between the reference voltage and the voltage applied to the motor is also affect position estimation accuracy and stability, thus it is compensated to ensure accuracy and stability of the sensorless control system. In this study, two voltage error compensation methods are used, and the effects of the compensation methods are discussed. The performance of the proposed sensorless control method is evaluated by experimental results.

• Journal of the Korean Data and Information Science Society
• /
• v.20 no.3
• /
• pp.505-514
• /
• 2009

The researcher analyzes the relationship between the number of neighbors and the prediction accuracy in the preference prediction process using collaborative filtering system. The number of neighbors who are involved in the preference prediction process are divided into four groups. Each group shows a little difference in the preference prediction. By using prediction error averages in each group, linear functions are suggested. Through the result of this study, the accuracy of preference prediction can be raised when using linear functions by using the number of neighbors in the suggested system.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2018.05a
• /
• pp.232-235
• /
• 2018

Convolutional Neural Network is arguably the most popular deep learning architecture that is one of the most attractive area of research since it has various applications including face detection and recognition. The cascaded CNN operates at multiple resolution and rejects the background regions in the fast low resolution stages. By considering that advantage, we carry out the study on accuracy of cascaded CNN for face detection applications. The key point for our study is to analysing and improving the accuracy of cascaded CNN by applying simulations of algorithm where by we used Google's Tensorflow GPU as deep learning framework.

• Journal of Korea Multimedia Society
• /
• v.22 no.6
• /
• pp.639-645
• /
• 2019

This paper proposes a salient chromagram components extraction method based on the temporal discrete cosine transform of a chromagram block to improve cover song retrieval accuracy. The proposed salient chromagram emphasizes tonal contents of music, which are well-preserved between an original song and its cover version, while reducing the effects of timbre difference. We apply the proposed salient chromagram extraction method as a preprocessing step for the Fourier-transform based cover song matching. Experiments on two cover song datasets confirm that the proposed salient chromagram improves the cover song search accuracy.

• Journal of IKEEE
• /
• v.22 no.4
• /
• pp.1188-1194
• /
• 2018

This paper proposes an image enhancement-based license plate detection algorithm to improve the overall performance of system. Non-uniform illumination conditions have huge impact on overall plate detection system accuracy. In this paper, we propose an algorithm for color image enhancement-based license plate detection for improving accuracy of images degraded by excessively strong and low sunlight. Firstly, the image is enhanced by Multi-Scale Retinex algorithm. Secondly, a plate detection method is employed to take advantage of geometric properties of connected components, which can significantly reduce the undesired plate regions. Finally, intersection over union method is applied for detecting the accurate location of number plate. Experimental results show that the proposed method significantly improves the accuracy of plate detection system.

• Journal of information and communication convergence engineering
• /
• v.17 no.1
• /
• pp.41-48
• /
• 2019

In recent years, sentiment analysis research has become popular. The research results of sentiment analysis have achieved remarkable results in practical applications, such as in Amazon's book recommendation system and the North American movie box office evaluation system. Analyzing big data based on user preferences and evaluations and recommending hot-selling books and hot-rated movies to users in a targeted manner greatly improve book sales and attendance rate in movies [1, 2]. However, traditional machine learning-based sentiment analysis methods such as the Classification and Regression Tree (CART), Support Vector Machine (SVM), and k-nearest neighbor classification (kNN) had performed poorly in accuracy. In this paper, an improved kNN classification method is proposed. Through the improved method and normalizing of data, the purpose of improving accuracy is achieved. Subsequently, the three classification algorithms and the improved algorithm were compared based on experimental data. Experiments show that the improved method performs best in the kNN classification method, with an accuracy rate of 11.5% and a precision rate of 20.3%.