• Journal of the Institute of Electronics Engineers of Korea SP
• /
• v.41 no.4
• /
• pp.93-101
• /
• 2004

A face boundary can be approximated by an ellipse with five-dimensional parameters. This property allows an ellipse detection algorithm to be adapted to detecting faces. However, the construction of a huge five-dimensional parameter space for a Hough transform is quite unpractical. Accordingly, we Propose a selectively attentional Hough transform method for detecting faces from a symmetric contour in an image. The idea is based on the use of a constant aspect ratio for a face, gradient information, and scan-line-based orientation decomposition, thereby allowing a 5-dimensional problem to be decomposed into a two-dimensional one to compute a center with a specific orientation and an one-dimensional one to estimate a short axis. In addition, a two-point selection constraint using geometric and gradient information is also employed to increase the speed and cope with a cluttered background. After detecting candidate face regions using the proposed Hough transform, a multi-layer perceptron verifier is adopted to reject false positives. The proposed method was found to be relatively fast and promising.

• Journal of the Institute of Convergence Signal Processing
• /
• v.10 no.1
• /
• pp.7-12
• /
• 2009

In this paper, we propose a directional feature extraction method for off-line handwritten numerals by using the morphological operations. Direction features are obtained from four directional line images, each of which contains horizontal, vertical, right-diagonal and left-diagonal lines in entire numeral lines. Conventional method for extracting directional features uses Kirsch masks which generate edge-shaped double line images for each direction, whereas our method uses directional erosion operations and generate single line images for each direction. To apply these directional erosion operations to the numeral image, preprocessing steps such as thinning and dilation are required, but resultant directional lines are more similar to numeral lines themselves. Our four [$4{\times}4$] directional features of a numeral are obtained from four directional line images through a zoning method. For obtaining the higher recognition rates of the handwrittern numerals, we use the multiple feature which is comprised of our proposed feature and the conventional features of a kirsch directional feature and a concavity feature. For recognition test with given features, we use a multi-layer perceptron neural network classifier which is trained with the back propagation algorithm. Through the experiments with the CENPARMI numeral database of Concordia University, we have achieved a recognition rate of 98.35%.

• Geophysics and Geophysical Exploration
• /
• v.12 no.1
• /
• pp.105-113
• /
• 2009

The interpretation of observed waveform characteristics identified in refraction and wide-angle reflection data increases confidence in the crustal structure model obtained. When calculating traveltimes and raypaths, wavefront methods on a regular grid based on graph theory are robust even with complicated structures, but basically compute only first arrivals. In this paper, we develop new algorithms to compute traveltimes and raypaths not only for first arrivals, but also for fast and later reflection arrivals, later refraction arrivals, and converted waves between P and S, using the modified wavefront method based on slowness network nodes mapped on a multi-layer model. Using the new algorithm, we can interpret reflected arrivals, Pg-later arrivals, strong arrivals appearing behind Pn, triplicated Moho reflected arrivals (PmP) to obtain the shape of the Moho, and phases involving conversion between P and S. Using two models of an ocean-continent transition zone and an oceanic ridge or seamount, we show the usefulness of this algorithm, which is confirmed by synthetic seismograms using the 2D Finite Difference Method (2D-FDM). Characteristics of arrivals and raypaths of the two models differ from each other in that using only first-arrival traveltime data for crustal structure analysis involves risk of erroneous interpretation in the ocean-continent transition zone, or the region around a ridge or seamount.

• Journal of the Society of Cosmetic Scientists of Korea
• /
• v.30 no.3 s.47
• /
• pp.369-375
• /
• 2004

In this study, liquid crystalline (LC) is formed using Gemini surfactant (GS) type and moisturizing effect in vivo is measured. $3.0\;wt\%$ of sodium dicocoyl ethylene diamine (PEG)-15 sulfate (SCD-PEG-15S) is used as GS and $4.0\;wt\%$ of hydrogenated dimer acid esters (HDAE) as booster. For stabilizers, $2.0\;wt\%$ of behenyl alcohol (BA) and $1.0\;wt\%$ of Iyso-lecithin (LyL) are utilized. It is stabilized in pH from 4.0 to 10.5 and the best condition is in pH 6.5. The value of viscosity is $8,000\pm500$ cP. The most excellent particles are formed within the range of 4.0 to 15.5 um. Formed LC is observed around LC particles using polarization microscope. It is also observed that lamellar gel network structure is formed around LC particles. Moisturizing effect is improved by $13.6\%$ (P<0.05) compared to control when measured 30 min later after coating samples. After 1 h, moisturizing effect is improved by 1$12.6\%$ (P<0.05) than control while showing $28.3\%$ (P<0.05) of improvement after 4 h. These results may be caused from that manufactured LC forms lamellar structure so that it has better water-holding ability and absorbance of oil increases. This formula could be utilized by delivery system (DS) on skin so that this technology can be applied for manufactuing pharmaceuticals and cosmetics.

• KIPS Transactions on Software and Data Engineering
• /
• v.12 no.1
• /
• pp.51-58
• /
• 2023

In this paper, we propose an organ segmentation technique for the automatic extraction of medical diagnostic indicators from X-ray images. In order to calculate diagnostic indicators of heart disease and spinal disease such as VHS(vertebral heart scale) and Cobb angle, it is necessary to accurately segment the thoracic spine, carina, and heart in a chest X-ray image. A deep neural network model in which the high-resolution representation of the image for each layer and the structure converted into a low-resolution feature map are connected in parallel was adopted. This structure enables the relative position information in the image to be effectively reflected in the segmentation process. It is shown that learning performance can be improved by combining the OCR module, in which pixel information and object information are mutually interacted in a multi-step process, and the channel attention module, which allows each channel of the network to be reflected as different weight values. In addition, a method of augmenting learning data is presented in order to provide robust performance against changes in the position, shape, and size of the subject in the X-ray image. The effectiveness of the proposed theory was evaluated through an experiment using 145 human chest X-ray images and 118 animal X-ray images.

• The Journal of the Korea institute of electronic communication sciences
• /
• v.17 no.2
• /
• pp.343-350
• /
• 2022

It is worth verifying the effectiveness of data integration between data with different features. This study investigated whether the data integration affects the accuracy of deep neural network (DNN), and which integration method shows the best improvement. This study used two different public datasets. One public dataset was taken in an actual farm in India. And another was taken in a laboratory environment in Korea. Leaf images were selected from two different public datasets to have five classes which includes normal and four different types of plant diseases. DNN used pre-trained VGG16 as a feature extractor and multi-layer perceptron as a classifier. Data were integrated into three different ways to be used for the training process. DNN was trained in a supervised manner via the integrated data. The trained DNN was evaluated by using a test dataset taken in an actual farm. DNN shows the best accuracy for the test dataset when DNN was first trained by images taken in the laboratory environment and then trained by images taken in the actual farm. The results show that data integration between plant images taken in a different environment helps improve the performance of deep neural networks. And the results also confirmed that independent use of plant images taken in different environments during the training process is more effective in improving the performance of DNN.

• Journal of Korean Tunnelling and Underground Space Association
• /
• v.20 no.6
• /
• pp.899-915
• /
• 2018

In this study, in order to construct an eco-friendly advanced road transportation network, the multi-layer tunnel, which is a small-sized car road, is designed to have a height of less than 60 cm. However, the shape of the tunnel is low and the height of the traffic sign is small. In order to solve these problems, traffic sign characters were designed in three dimensions, and the possibility of applying the design of the three - dimensional sign that can obtain greater visibility than the existing signs at the same distance and the possibility verification through virtual simulation were performed. The three-dimensional sign is horizontally installed on the ceiling of the multi-layer tunnel. To be seen vertically, it is enlarged by a certain ratio by the perspective, and the width and height are enlarged. Respectively. In addition, 3D simulation was performed to verify the visibility of the stereoscopic signs when the driver ran through the stereoscopic sign design specifications. As a result of the design and experimental study, it was confirmed that the stereoscopic sign could be designed through the theoretical formula and that it could provide the driver with a larger traffic sign character because there is no limitation of the facility limit compared to the existing vertical traffic sign. Also, we confirmed that it can be implemented in the side wall by using the stereoscopic sign design principle installed on the ceiling part. It was confirmed that the design of the stereoscopic sign can be designed to be smaller as the distance that the driver visually recognizes the sperm is shorter, the height of the protrusion vertically at the lower part of the stereoscopic sign becomes higher. As a result of 3D simulation running experiment based on the design information of the stereoscopic sign, it was confirmed that the stereoscopic sign is visually the same as the vertical sign at the planned distance. Although the detailed research and institutional improvement of stereoscopic signs have not been made in Korea and abroad, it is evolved into a core technology of new road traffic facilities through various studies through the possibility of designing and applying stereoscopic signs developed through this study Expect.

• Polymer(Korea)
• /
• v.38 no.6
• /
• pp.752-759
• /
• 2014

Crosslinked polyimides (PIs) were synthesized by reacting 4,4'-(hexafluoroisopropylidene)-diphthalic anhydride (6FDA) and 2,2'-bis(trifluoromethyl)benzidine (TFDB) with various ratios of the cross-linkable, end-capping agent cis-1,2,3,6-tetrahydrophthalic anhydride (CDBA) via ring-opening metathesis polymerization. Residual stress behaviors were investigated in-situ during thermal imidization of the crosslinked PI precursors using a thin film stress analyzer (TFSA) by wafer bending method. The thermal properties were investigated via differential scanning calorimetry (DSC), thermomechanical analysis (TMA), and thermogravimetric analysis (TGA). The optical properties were measured by ultraviolet-visible spectrophotometer (UV-vis) and spectrophotometry. All properties were interpreted with respect to their morphology of crosslinked networks. With increasing the amounts of the end-capping agent, the residual stress decreased from 27.9 to -1.3 MPa, exhibited ultra-low stress and high thermal properties. The minimized residual stress and enhanced thermal properties of the crosslinked PI makes them potential candidates for versatile high-density multi-layer structure applications.

• The Korean Journal of Applied Statistics
• /
• v.35 no.4
• /
• pp.457-468
• /
• 2022

In aged society, it is important to prevent older people from being disability needing long-term care. The purpose of this study is to develop a prediction model to discover high-risk groups who are likely to be beneficiaries of Long-Term Care Insurance. This study is a retrospective study using database of National Health Insurance Service (NHIS) collected in the past of the study subjects. The study subjects are 7,724,101, the population over 65 years of age registered for medical insurance. To develop the prediction model, we used logistic regression, decision tree, random forest, and multi-layer perceptron neural network. Finally, random forest was selected as the prediction model based on the performances of models obtained through internal and external validation. Random forest could predict about 90% of the older people in need of long-term care using DB without any information from the assessment of eligibility for long-term care. The findings might be useful in evidencebased health management for prevention services and can contribute to preemptively discovering those who need preventive services in older people.

• Journal of the Korea Society of Computer and Information
• /
• v.14 no.12
• /
• pp.157-167
• /
• 2009

Recently, the leakage of confidential information and personal information is taking place on the Internet more frequently than ever before. Most of such online security incidents are caused by attacks on vulnerabilities in web applications developed carelessly. It is impossible to detect an attack on a web application with existing firewalls and intrusion detection systems. Besides, the signature-based detection has a limited capability in detecting new threats. Therefore, many researches concerning the method to detect attacks on web applications are employing anomaly-based detection methods that use the web traffic analysis. Much research about anomaly-based detection through the normal web traffic analysis focus on three problems - the method to accurately analyze given web traffic, system performance needed for inspecting application payload of the packet required to detect attack on application layer and the maintenance and costs of lots of network security devices newly installed. The UTM(Unified Threat Management) system, a suggested solution for the problem, had a goal of resolving all of security problems at a time, but is not being widely used due to its low efficiency and high costs. Besides, the web filter that performs one of the functions of the UTM system, can not adequately detect a variety of recent sophisticated attacks on web applications. In order to resolve such problems, studies are being carried out on the web application firewall to introduce a new network security system. As such studies focus on speeding up packet processing by depending on high-priced hardware, the costs to deploy a web application firewall are rising. In addition, the current anomaly-based detection technologies that do not take into account the characteristics of the web application is causing lots of false positives and false negatives. In order to reduce false positives and false negatives, this study suggested a realtime anomaly detection method based on the analysis of the length of parameter value contained in the web client's request. In addition, it designed and suggested a WAF(Web Application Firewall) that can be applied to a low-priced system or legacy system to process application data without the help of an exclusive hardware. Furthermore, it suggested a method to resolve sluggish performance attributed to copying packets into application area for application data processing, Consequently, this study provide to deploy an effective web application firewall at a low cost at the moment when the deployment of an additional security system was considered burdened due to lots of network security systems currently used.