• 디지털융복합연구
• /
• 제10권3호
• /
• pp.173-178
• /
• 2012

본 논문에서는 영상의 관심 영역을 선택추출하여 효과적으로 객체를 추출 할 수 있는 관심 영역 지도(Saliency Map) 생성 기법을 제안하였다. 제안하는 방법은 객체의 윤곽선에 초점을 맞추어 단일영상의 에지(Edge), HSV 색상 모델의 H(Hue)성분, 포커스(Focus), 엔트로피(Entropy)의 네 가지 특징 정보를 이용한 각각의 특징 지도(Feature Map)를 생성하고, 생성된 특징 지도들을 중심 주변 차이(Center Surround Differences)를 이용하여 중요도 지도(conspicuity map)를 생성하게 된다. 이후 생성된 중요도 지도들을 조합함으로써 관심 영역 지도를 생성하게 된다. 제안한 기법을 이용하여 생성한 관심 영역 지도를 기존 기법의 관심 영역 지도와 비교한 결과 제안한 기법의 우수함을 알 수 있었다.

• Transactions on Electrical and Electronic Materials
• /
• 제17권6호
• /
• pp.363-368
• /
• 2016

Rectangular shape detection plays an important role in many image recognition systems. However, it requires continued research for its improved performance. In this study, we propose a strong rectangular shape detection algorithm, which combines the canny edge and line detection algorithms based on the perpendicularity and parallelism of a rectangle. First, we use the canny edge detection algorithm in order to obtain an image edge map. We then find the edge of the contour by using the connected component and find each edge contour from the edge map by using a DP (douglas-peucker) algorithm, and convert the contour into a polyline segment by using a DP algorithm. Each of the segments is compared with each other to calculate parallelism, whether or not the segment intersects the perpendicularity intersecting corner necessary to detect the rectangular shape. Using the perpendicularity and the parallelism, the four best line segments are selected and whether a determined the rectangular shape about the combination. According to the result of the experiment, the proposed rectangular shape detection algorithm strongly showed the size, location, direction, and color of the various objects. In addition, the proposed algorithm is applied to the license plate detecting and it wants to show the strength of the results.

• 한국진공학회:학술대회논문집
• /
• 한국진공학회 2016년도 제50회 동계 정기학술대회 초록집
• /
• pp.243.1-243.1
• /
• 2016

Recently, many groups have attempted to fabricate 3-dimensional (3D) structures of GaN such as pyramids, rods, stripes and annulars. Since quantum structures on non-polar and semi-polar planes of 3D structures have less influence of internal electric filed, multi quantum wells (MQWs) formed on those planes have high quantum efficiency. Especially, pyramidal and annular structures consist of various crystal planes with different emission wavelength, providing a possibillity of phosphor-free white light emtting diodes (WLEDs).[1] However, it still has problem to obtain high color rendering index (CRI) number because of narrow-band emission and poor indium composition caused by the formation of few number of facets during metal-organic chemical vapor deposition growth.[2] If we can fabricate 3D structure having more various facets, we can make broad-band emittied WLEDs and improve CRI number. In this study, we suggest a simple method to fabricate 3D structures having various facet and containing high indium composition by means of a combination of metal-organic chemical vapor deposition and wet chemical etching techniques.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제9권2호
• /
• pp.775-792
• /
• 2015

Person re-identification is an important and challenging task in computer vision with numerous real world applications. Despite significant progress has been made in the past few years, person re-identification remains an unsolved problem. This paper presents a novel appearance-based approach to person re-identification. The approach exploits region covariance matrix and color histograms to capture the statistical properties and chromatic information of each object. Robustness against low resolution, viewpoint changes and pose variations is achieved by a novel signature, that is, the combination of Log Covariance Matrix feature and HSV histogram (LCMH). In order to further improve re-identification performance, third-party image sets are utilized as a common reference to sufficiently represent any image set with the same type. Distinctive and reliable features for a given image set are extracted through decision boundary between the specific set and a third-party image set supervised by max-margin criteria. This method enables the usage of an existing dataset to represent new image data without time-consuming data collection and annotation. Comparisons with state-of-the-art methods carried out on benchmark datasets demonstrate promising performance of our method.

• Journal of Electrical Engineering and Technology
• /
• 제12권1호
• /
• pp.363-372
• /
• 2017

This paper presents a novel face recognition algorithm based on the deep convolution neural network and key point detection jointed local binary pattern methodology to enhance the accuracy of face recognition. We firstly propose the modified face key feature point location detection method to enhance the traditional localization algorithm to better pre-process the original face images. We put forward the grey information and the color information with combination of a composite model of local information. Then, we optimize the multi-layer network structure deep learning algorithm using the Fisher criterion as reference to adjust the network structure more accurately. Furthermore, we modify the local binary pattern texture description operator and combine it with the neural network to overcome drawbacks that deep neural network could not learn to face image and the local characteristics. Simulation results demonstrate that the proposed algorithm obtains stronger robustness and feasibility compared with the other state-of-the-art algorithms. The proposed algorithm also provides the novel paradigm for the application of deep learning in the field of face recognition which sets the milestone for further research.

• 한국컴퓨터정보학회논문지
• /
• 제10권6호
• /
• pp.137-152
• /
• 2005

본 논문에서는 사용자 프로파일을 기반으로 한 정보 필터링을 사용하여 개개인의 학생들에게 맞는 효율적인 웹 기반 비디오 학습 평가 시스템을 제시한다. 비디오 기반 문제들이 주어지면, 키 프레임들은 위치, 크기, 그리고 컬러 정보를 기반으로 추출되고, 문제 출제 구간은 시간 윈도우와 그레이-레벨 히스토그램의 차이에 의해 추출된다 게다가, 카테고리-기반 시스템과 키워드 기반 시스템을 합성하여 문제 출제가 효율적인 평가를 확실해 하도록 주어진다. 따라서 학생들은 관심 영역을 유지하면서 부족한 영역을 보충함으로서 학업 성취도를 향상시킬 수 있다.

• 한국의상디자인학회지
• /
• 제14권3호
• /
• pp.123-134
• /
• 2012

As a mirror of its time, fashion exactly reflects the trend of today' s society. In addition, as a tool to satisfy our desire and express images, it brings diverse image changes in people's minds. This kind of pattern is frequently found in wedding images, especially in wedding dresses. When it comes to a wedding dress, in general, we often think of a modest and pure-white one. In modem times, however, people prefer a differentiated image which reflects their personality and ego. In other words, it is an important means of expressing their own personality by focusing on their active and personal aspects and getting away from the passive, stable and formative aspects in the past. These days, a combination of narcissism and eroticism is found on wedding dresses. This kind of trend has been expressed into romanticism. In terms of the latest sensitive wedding dress trend, unique and creative couture wedding dresses have appeared. Therefore, this study has classified this kind of trend into two categories; Formative and Design aspects. According to an analysis on the formative and design aspects of romanticism in wedding dresses, diverse characteristics were observed in shape, color, material and ornament in the Formative Aspect. In terms of design, the characteristics of sensuality, ornament, exoticism and compromise were observed.

• 한국의류산업학회지
• /
• 제17권4호
• /
• pp.604-612
• /
• 2015

This study investigates preferred bicycle wear designs that can satisfy active senior consumers. A survey was conducted on 50-60's women who periodically rode bicycles. The results indicated a preference for slim designs and red colors. Jacket designs preferences were for a tight fit for size tolerance, stand collar style, and elastic band details for cuff styles. Pants design preferences were for a whole band waist belt type with a tight fit style such as leggings in pants silhouette, zipper details on the side line and ankle length. They also preferred styles with pads attached to underpants in the pad style and the part of the back waist in the pocket position. The survey showed four kinds of jacket design drawings on an ordinal scale rating. Results indicated a preference for set-in variation jackets with the red and gray color combination. Finally, we demonstrated bicycle wear design suggestions. The jacket applied different armhole line colors connected to the sleeve to make the waist slimmer; in addition, stretchable material helped improve armpit part functionality.

• 공학교육연구
• /
• 제21권1호
• /
• pp.77-89
• /
• 2018

The chromosomes of all the world are the same in all 24 pairs, but the key, skin color and appearance are different. Also, it is the resistance of adult disease, diabetes, cancer. In 1953, Watson, Crick of Cambridge University experimentally discovered a DNA double helix structure, and in 1962, They laureates the Nobel Prize. In 1964, Temin, University of Wisconsin, USA, experimentally identified the ability to copy gene information from RNA to DNA and received the Nobel Prize in 1975. In this paper, we analyzed 24 pairs of DNA chromosomes using mathematical matrices based on the combination order sequence of four groups, and designed the Taegeuk pattern genetic code for the first time in the world. In the case of normal persons, the middle Yin-Yang taegeuk is designed as a block circulant Jacket matrix in DNA, and the left-right and upper-lower pairs of east-west and north-south rulings are designed as pair complementary matrices. If (C U: A G) chromosomes are unbalanced, that is, people with disease or inheritance become squashed squirming patterns. In 2017, Professor Michel Young was awarded a Nobel by presenting a biological clock and experimentally explained the bio-imbalance through a yellow fruit fly experiment.This study proved mathematical matrices for balanced and unbalanced RNA.

• 패션비즈니스
• /
• 제13권5호
• /
• pp.13-22
• /
• 2009

The purpose of this paper is to understand some methods to provide women's shoes and their design images from the 1920s to the 1940s, and to benefit from the new methods for creative design and new materials in the future. It also examined the features with the academic background of women's shoes. To target on analysis items and shoes design factors, moreover, it made reference to the paper by former research. As a result, 28 shoes (30.04%) in the 1920s, 47 shoes(51.09%) in the 1930s, and 17 shoes (18.47%) in the 1940s have been collected. The strap shoes, combination color, bright texture, and buckle trimmings are first on the list. The most popular shoes were pumps and strap, in terms of design and shape. In particular, black was one of the common and popular colors. Most of the shoes had buckles. Texture also was brilliant. This study was meaningful to provide understanding and information about a variety of women's shoes designs from the 1920s to the 1930s.