• ETRI Journal
• /
• v.32 no.5
• /
• pp.784-794
• /
• 2010

Automatic facial expression recognition has many potential applications in different areas of human computer interaction. However, they are not yet fully realized due to the lack of an effective facial feature descriptor. In this paper, we present a new appearance-based feature descriptor, the local directional pattern (LDP), to represent facial geometry and analyze its performance in expression recognition. An LDP feature is obtained by computing the edge response values in 8 directions at each pixel and encoding them into an 8 bit binary number using the relative strength of these edge responses. The LDP descriptor, a distribution of LDP codes within an image or image patch, is used to describe each expression image. The effectiveness of dimensionality reduction techniques, such as principal component analysis and AdaBoost, is also analyzed in terms of computational cost saving and classification accuracy. Two well-known machine learning methods, template matching and support vector machine, are used for classification using the Cohn-Kanade and Japanese female facial expression databases. Better classification accuracy shows the superiority of LDP descriptor against other appearance-based feature descriptors.

• Journal of Korea Multimedia Society
• /
• v.19 no.4
• /
• pp.747-755
• /
• 2016

Automated guided vehicles have been used for a long time to increase work efficiency in the logistics field, but it is difficult to apply to a variety of logistics sites due to either the restricted movement mechanism or expensive devices. In this paper, we present a flexible software architecture that is hardware-independent for a mobile power cart of the follow mode and implement it using a ROS software platform. Through the SCV analysis for the system functionalities, we design a package to track a user movement and a package to control a new hardware platform. It has an advantage to use a variety of movement algorithms and hardware platforms by applying the strategy pattern and the template method pattern for the design of a software architecture. Through the performance evaluation, we show that the proposed design is maintainable in terms of a software complexity and it detects a user's movement by obtaining a user skeleton information so that it can control a hardware platform to move at a certain distance.

• Korean Journal of Agricultural Science
• /
• v.49 no.4
• /
• pp.821-831
• /
• 2022

American plum line pattern virus (APLPV), a member of the genus Ilarvirus in the family Bromoviridae, is one of the plant quarantine pathogens in Korea. In this study, 15 candidate primer sets were designed and examined to develop a reverse transcription polymerase chain reaction (RT-PCR) assay for plant quarantine inspection of APLPV. Using APLPV-infected and healthy samples, the primer sets were assessed for APLPV detection. To confirm the occurrence of nonspecific reactions, six ilarviruses (Apple mosaic virus, Asparagus virus 2, Blueberry shock virus, Prune dwarf virus, Prunus necrotic ringspot virus, and Tobacco streak virus) and 10 target plants (Prunus mume, P. yedoensis, P. persica, P. armeniaca, P. dulcis, P. tomentosa, P. avium, P. glandulosa, P. salicina, and P. cerasifera) were examined. Finally, two primer sets were selected. These primer sets could generate the expected amplicons even with at least 1 ng of the total RNA template in concentration-dependent amplifications. In addition, a positive clone was developed for use as a positive control in the abovementioned RT-PCR assay.

• Journal of Broadcast Engineering
• /
• v.25 no.6
• /
• pp.861-869
• /
• 2020

We introduce the application that automatically makes several images stored in user's device into one video by using the different climax patterns appearing for each film genre. For the classification of the genre characteristics of movies, a climax pattern model style was created by analyzing the genre of domestic movie drama, action, horror and foreign movie drama, action, and horror. The climax pattern was characterized by the change in shot size, the length of the shot, and the frequency of insert use in a specific scene part of the movie, and the result was visualized. The model visualized by genre developed as a template using Firebase DB. Images stored in the user's device were selected and matched with the climax pattern model developed as a template for each genre. Although it is a short video, it is a feature of the proposed application that it can create an emotional story video that reflects the characteristics of the genre. Recently, platform operators such as YouTube and Naver are upgrading applications that automatically generate video using a picture or video taken by the user directly with a smartphone. However, applications that have genre characteristics like movies or include video-generation technology to show stories are still insufficient. It is predicted that the proposed automatic video editing has the potential to develop into a video editing application capable of transmitting emotions.

• Journal of KIISE:Computing Practices and Letters
• /
• v.12 no.2
• /
• pp.129-140
• /
• 2006

The existing CBD development methods deal with the analysis phase in a superficial manner. Applying such a superficial analysis to business applications with a number of subsystems makes analysis models be inconsistent with levels and styles, only depending on experiences of the analysts. This inconsistent analysis might cause more serious problems during the subsequent development phases, resulting in the failure of the projects. In this paper, we propose a rule-based analysis pattern that provides an analysis template for business applications. This pattern analyzes the concepts of business applications by using external events and internal rules that process the events. Employing this pattern, a huge business application can be developed by a couple of co-analysts who work together in a consistent and systematic manner. This paper also describes an efficient way to develop business components with the suggested analysis pattern using banking deposit case study through UML Components development process.

• Journal of the Korean Crystal Growth and Crystal Technology
• /
• v.20 no.4
• /
• pp.164-167
• /
• 2010

Bulk GaN single crystal with 1.5 mm thickness was successfully grown by hydride vapor phase epitaxy (HVPE) technique. Free-standing GaN substrates of $10{\times}10,\;15{\times}15$ mm size were fabricate after lift-off of sapphire substrate and their optical properties were characterized properties for device applications. X-ray diffraction patterns showed (002) and (004) peak, and the FWHM of the X-ray rocking curve (XRC) measurement in (002) was 98 arcsec. A sharp photoluminescence spectrum at 363 nm was observed and defect spectrum at visible range was not detected. The hexagonal-shaped etch-pits are formed on the GaN surface in $200^{\circ}C\;H_3PO_4$ at 5 minutes. The defect density calculated from observed etch-pits on surface was around $5{\times}10^6/cm^2$. This indicates that the fabricated GaN substrates can be used for applications in the field of optodevice, and high power electronics.

• Proceedings of the Korean Vacuum Society Conference
• /
• 2013.08a
• /
• pp.221-221
• /
• 2013

Vertical LED (VLED) has been recognized as a way to obtain the high-power LED due to their advantages [1]. However, approximately 4% of the light generated from the active region is extracted, if the light extraction from side walls and back side is neglected because of Fresnel reflection (FR) and total internal reflection (TIR) [2,3]. In this study, the optical simulation of the VLED with the various microstructures was performed. Among them, the microlens having the diameter of 3 ${\mu}m$ and the height of 1.5 ${\mu}m$ shown the best result was chosen, and then, optimized microlens was formed on a GaN template using conventional semiconductor process. Various microstructures were proposed to improve the light extraction efficiency (LEE) of the VLED for the simulation. The LEE was simulated using LightTools based on a Monte Carlo ray tracing. The microstructures with hemisphere, cone, truncated and cylinder pattern having diameter of 3 ${\mu}m$ were employed on the top layer of the VLED respectively. The improvement of the LEE by using the microstructure is 87% for the hemisphere, 77% for the cone, 53% for the truncated, 21% for the cylinder, compared with the LEE of the flat surface at the reflectance of 85%. The LEE was increased by 88% at the height of 1.5 ${\mu}m$, compared with the LEE of the flat surface. We found that the microlens on the top layer is the most suitable for increasing the LEE. In order to apply the proposed microlens on n-GaN surface, we fabricated microlens on a GaN template. A photoresist array having hexagonal-closed packed microlens was fabricated on the GaN template. Then, optimization of etching the GaN template was performed using a dry etching process with ICP-RIE. The dry etching carried out using a gas mixture of Cl2 and Ar, each having a flow rate of 16 sccm and 10 sccm, respectively with RF power of 50 W, ICP power of 900 W and chamber pressure of 2 mTorr was the optimum etching condition as shown in Fig. 2(a).

• Journal of Intelligence and Information Systems
• /
• v.16 no.2
• /
• pp.95-108
• /
• 2010

In this paper, we present a hybrid neural network model for dynamic hand gesture recognition. The model consists of two modules, feature extraction module and pattern classification module. We first propose a modified CNN(convolutional Neural Network) a pattern recognition model for the feature extraction module. Then we introduce a weighted fuzzy min-max(WFMM) neural network for the pattern classification module. The data representation proposed in this research is a spatiotemporal template which is based on the motion information of the target object. To minimize the influence caused by the spatial and temporal variation of the feature points, we extend the receptive field of the CNN model to a three-dimensional structure. We discuss the learning capability of the WFMM neural networks in which the weight concept is added to represent the frequency factor in training pattern set. The model can overcome the performance degradation which may be caused by the hyperbox contraction process of conventional FMM neural networks. From the experimental results of human action recognition and dynamic hand gesture recognition for remote-control electric home appliances, the validity of the proposed models is discussed.

• Journal of Korea Multimedia Society
• /
• v.15 no.6
• /
• pp.784-793
• /
• 2012

Defects of TFT-LCD is detected by thresholding the difference image between the input image and template one because LCD panel has its inherent patterns. However, the pitch corresponding to pattern period is gradually changed according to the distance from the center of camera due to geometric distortion of camera characteristics. This paper presents a method to detect defects through comparing the pitch area with neighbor pitch areas where the perspective transform is performed with the extracted features to correct the distortion. The experimental results show that the performance of the proposed method is very effective for real data.

• Journal of Power System Engineering
• /
• v.17 no.5
• /
• pp.121-127
• /
• 2013

In this paper, we propose a distance measurement system for automatic berthing control using a stereo camera mounted on a rotation control device, and a radial pattern target. Automatically controlling the position and attitude of a ship aims to prevent maritime accidents due to human error. Our goal is to measure the relative distance between a ship and an onshore or offshore target for berthing. Therefore, the distance should be continuously measured while tracking a fixed point on a target. To this end, we developed a stereo camerabased distance measurement system that satisfied these requirements. This paper describes the structure and principle of the measurement system. We validate the distance error for target incline due to the relative position and attitude between a camera and a target in miniature scale. In addition, the findings of an experiment in an outdoor environment demonstrate that the proposed measurement system has accuracy within 1 m at a range of 20-100 m which is the acceptable accuracy for automatic berthing.