• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.39A no.2
• /
• pp.92-99
• /
• 2014

This paper proposes the method of facial expression recognition based on decision tree structure. In the image of facial expression, ASM(Active Shape Model) and LBP(Local Binary Pattern) make the local features of a facial expressions extracted. The discriminant features gotten from local features make the two facial expressions of all combination classified. Through the sum of true related to classification, the combination of facial expression and local region are decided. The integration of branch classifications generates decision tree. The facial expression recognition based on decision tree shows better recognition performance than the method which doesn't use that.

• Transactions of the Korean Society of Mechanical Engineers B
• /
• v.31 no.5
• /
• pp.457-466
• /
• 2007

A numerical and experimental study on the evaluation and the prediction of cleanliness level in the mini-environment system was carried out. Using the concept of local mean air-age (LMA) and effective flow rate, the new direct method for estimating the mini-environment was developed and compared with the previous performance index of airflow pattern characteristics. It was found out that the airflow pattern analysis is a restricted method to estimate the real performance of the mini-environment. The reason is that the airflow pattern cannot predict the effect of the increment of the ventilation rate on the cleanliness level in the mini-environment. While LMA is capable of showing the effects of the contaminant accumulation caused by turbulent intensity, eddy, and the increment of the effective flow rate. This result showed that LMA is more exact and effective performance index than the previous one like the airflow pattern characteristics.

• Journal of Korea Society of Digital Industry and Information Management
• /
• v.9 no.2
• /
• pp.111-119
• /
• 2013

Face recognition has recently become one of the most popular research areas in the fields of computer vision, machine learning, and pattern recognition because it spans numerous applications, such as access control, surveillance, security, credit-card verification, and criminal identification. In this paper, we propose a simple descriptor called an ECSP(Extended Center-Symmetric Pattern) for illumination-robust face recognition. The ECSP operator encodes the texture information of a local face region by emphasizing diagonal components of a previous CS-LBP(Center-Symmetric Local Binary Pattern). Here, the diagonal components are emphasized because facial textures along the diagonal direction contain much more information than those of other directions. The facial texture information of the ECSP operator is then used as the input image of an image covariance-based feature extraction algorithm such as 2D-PCA(Two-Dimensional Principal Component Analysis). Performance evaluation of the proposed approach was carried out using various binary pattern operators and recognition algorithms on the Yale B database. The experimental results demonstrated that the proposed approach achieved better recognition accuracy than other approaches, and we confirmed that the proposed approach is effective against illumination variation.

• Transactions of the Korean Society of Mechanical Engineers
• /
• v.10 no.6
• /
• pp.814-822
• /
• 1986

Flow patterns and its transitions in vertical two phase flow of air-water isothermal flow are identified objectively by void output signals and moments computed from the Probability Density Function which is associated with the statistical measurement for time average local void fractions using conductance probe. It has been shown that the probe output signals, PDF distributions and its moments are deterministic criteria of flow pattern and transition classification.

• Proceedings of the IEEK Conference
• /
• 2003.07e
• /
• pp.2108-2111
• /
• 2003

This paper introduces a new local feature extraction method and image matching method for the localization and classification of targets. Proposed method is based on the block-by-block projection associated with directional pattern of blocks. Each pattern has its own eigen-vertors called as CEBs(Classified Eigen-Blocks). Also proposed block-based image matching method is robust to translation and occlusion. Performance of proposed feature extraction and matching method is verified by the face localization and FLIR-vehicle-image classification test.

• Proceedings of the KIEE Conference
• /
• 2006.04a
• /
• pp.117-119
• /
• 2006

In this work, we suggest another method to localize DFT in spatial domain. This enables DFT algorithm to be used for local pattern matching. Once calculated, it costs same load to calculate localized DFT regardless of the size or the position of local region In spatial domain. We applied this method to face detection problem and got the results which prove the utility of our method.

• Proceedings of the IEEK Conference
• /
• 2007.07a
• /
• pp.387-388
• /
• 2007

In this paper, a content-based image retrieval method considering both local information and spatial correlation of image is proposed. In order to efficiently represent the spatial correlation, texture structure is classified into three kinds of pattern. In experiment result, our method improves $3.94%{\sim}11.23%$ precision rate over the existing methods.

• Transactions of the Korean hydrogen and new energy society
• /
• v.25 no.6
• /
• pp.629-635
• /
• 2014

This paper presents annual energy production (AEP) by a 1.5kW wind turbine due to be installed in Deokjeok-Do island. Local wind data is determined by geometric shape of Deokjeok-Do island and annual wind data from Korea Institute of Energy Research at three places considered to be installed the wind turbine. Numerical simulation using WindSim is performed to obtain flow pattern for the whole island. The length of each computation grid is 40 m, and k-e turbulence model is imposed. AEP is determined by the power curve of the wind turbine and the local wind data obtained from numerical simulation. To capture the more detailed flow pattern at the specific local region, Urumsil-maul inside the island, fine mesh having the grid length of 10m is evaluated. It is noted that the input data for numerical simulation to the local region is used the wind data obtained by the numerical results for the whole island. From the numerical analysis, it is found that a local AEP at the Urumsil-maul has almost same value of 1.72 MWh regardless the grid resolutions used in the present calculation. It is noted that relatively fine mesh used for local region is effective to understand the flow pattern clearly.

• Electrical & Electronic Materials
• /
• v.10 no.4
• /
• pp.349-354
• /
• 1997

This paper aims at studying the improvement of test method for t-weight sensitive fault (t-wsf) detect. The development of RAM fabrication technology results in not only the increase at device density on chips but also the decrease in line widths in VLSI. But, the chip size that was large and complex is shortened and simplified while the cost of chips remains at the present level, in many cases, even lowering. First of all, The testing patterns for RAM fault detect, which is apt to be complicated , need to be simplified. This new testing method made use of Local Lower Bound (L.L.B) which has the memory with the beginning pattern of 0(l) and the finishing pattern of 0(1). The proposed testing patterns can detect all of RAM faults which contain stuck-at faults, coupling faults. The number of operation is 6N at 1-weight sensitive fault, 9,5N at 2-weight sensitive fault, 7N at 3-weight sensitive fault, and 3N at 4-weight sensitive fault. This test techniques can reduce the number of test pattern in memory cells, saving much more time in test, This testing patterns can detect all static weight sensitive faults and pattern sensitive faults in RAM.

• Progress in Superconductivity and Cryogenics
• /
• v.17 no.2
• /
• pp.25-30
• /
• 2015

In order to realize economical applications, it is important to reduce the ac loss of 2G high-temperature superconductor coated conductors. It seems to be reasonable that a multi-filamentary wire can decrease the magnetization loss. In this study, we prepared two samples of YBCO coated conductors with striations. We measured local superconducting properties of both samples by using Low Temperature Scanning Laser and Hall Probe Microscopy (LTSLHPM). The distribution of the local critical temperature of samples was analyzed from experimental results of Low Temperature Scanning Laser Microscopy (LTSLM) near the superconducting transition temperature. According to LTSLM results, spatial distributions of the local critical temperature of both samples are homogeneous. The local current density and the local magnetization in samples were explored from measuring stray fields by using Scanning Hall Probe Microscopy (SHPM). From SHPM results, the remanent field pattern of the one bridge sample in an external magnetic field confirms the Bean's critical state model and the three bridge sample has similar remanent field pattern of the one bridge sample. The local magnetization curve in the three bridge sample was measured from external fields from -500 Oe to 500 Oe. We visualized that the distribution of local hysteresis loss are related in the distribution of the remanent field of the three bridge sample. Although the field dependence of the critical current density must be taken into account, the relation of the local hysteresis loss and the remanent field from Bean's model was useful.