• 한국멀티미디어학회논문지
• /
• 제19권12호
• /
• pp.1909-1918
• /
• 2016

This paper presents a unified framework for joint Convolutional Neural Network (CNN) based vehicle detection by leveraging multi-spectral image pairs. With the observation that under challenging environments such as night vision and limited light source, vehicle detection in a single color image can be more tractable by using additional far-infrared (FIR) image, we design joint CNN architecture for both RGB and FIR image pairs. We assume that a score map from joint CNN applied to overall image can be considered as confidence of vehicle existence. To deal with various scale ratios of vehicle candidates, multi-scale images are first generated scaling an image according to possible scale ratio of vehicles. The vehicle candidates are then detected on local maximal on each score maps. The generation of overlapped candidates is prevented with non-maximal suppression on multi-scale score maps. The experimental results show that our framework have superior performance than conventional methods with a joint framework of multi-spectral image pairs reducing false positive generated by conventional vehicle detection framework using only single color image.

• 한국정보통신학회논문지
• /
• 제6권5호
• /
• pp.789-795
• /
• 2002

컬러 이미지는 Gray Scale 이미지와는 달리 3가지 채널의 조합으로 이루어지고 방대한 정보량 때문에 효과적인 이미지 분할이 어렵다. 본 논문에서는 범용성 있는 Color Morphological Pyramids(CMP)구조를 제안하고, 그를 이용한 이미지 분할을 보인다. 이미지 피라미드 구조는 최초 이미지의 반복적인 필터링과 샘플링에 의해 면적비가 $2^{\int}({\int}=1,2,....,N)$이 되는 순차적 이미지 계열이다. 본 방법에서는 CMP를 이용하여 RGB, HSI, CMY 등의 컬러 공간에서 연속적인 필터링 처리로 불필요한 크기의 물체 및 잡음을 제거하고, 다운샘플링과정으로 해상도를 낮춰준다. 생성된 CMP에서 인접 레벨 이미지간에는 이웃한 픽셀 벡터간의 상대거리를 이용한 연결식을 사용하여 새 레벨의 이미지를 생성한 후 이를 이미지 분할한다. 이미지 분할실험을 통하여 본 방법의 유효성을 검증한다.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제11권1호
• /
• pp.570-595
• /
• 2017

Robust and perceptually-adaptive image watermarking algorithms have mainly targeted gray-scale images either at the modeling or embedding levels despite the widespread availability of color images. Only few of the existing algorithms are specifically designed for color images where color correlation and perception are constructively exploited. In this paper, a new perceptual and high-capacity color image watermarking solution is proposed based on the extension of Tsui et al. algorithm. The $CIEL^*a^*b^*$ space and the spatio-chromatic Fourier transform (SCFT) are combined along with a perceptual model to hide watermarks in color images where the embedding process reconciles between the conflicting requirements of digital watermarking. The perceptual model, based on an emerging color image model, exploits the non-uniform just-noticeable color difference (NUJNCD) thresholds of the $CIEL^*a^*b^*$ space. Also, spread-spectrum techniques and semi-random low-density parity check codes (SR-LDPC) are used to boost the watermark robustness and capacity. Unlike, existing color-based models, the data hiding capacity of our scheme relies on a game-theoretic model where upper bounds for watermark embedding are derived. Finally, the proposed watermarking solution outperforms existing color-based watermarking schemes in terms of robustness to standard image/color attacks, hiding capacity and imperceptibility.

• 한국생활과학회지
• /
• 제16권5호
• /
• pp.987-996
• /
• 2007

The purpose of this study is to find out the effect of three clothing cues(jacket color, skirt color, and tone of Korean dress) on the image of Korean dress wearer using tone-in-tone coloration. The experimental materials developed for this study are a set of stimuli and response scales. The stimuli are 16 color pictures manipulated with three clothing cues by drawing. The 7-point scale designed for visual evaluation of female impression formation includes 24 bipolar adjectives. The subjects were 192 undergraduate female students in Chinju city. The results of this study are as follow: As analyzing the image of female figure by the jacket color, skirt color, and tone, three factors including youth activity, attractiveness, and visibility were identified. Among these factors, youth activity and attractiveness were proved to be more important. Each dimensional image was affected by color coordination of jacket color, skirt color and tone of korean dress. Therefore, the image of Korean dress wearer could be varied by color coordination of jacket color, skirt color and tone of Korean dress.

• 패션비즈니스
• /
• 제17권6호
• /
• pp.111-124
• /
• 2013

The purpose of this research was to conduct a comparison analysis of color in the characteristics and image of Korea/China/Japan airline uniforms. Research subjects for this research included 19 Korean, Chinese and Japanese airlines servicing the Incheon International Airport in South Korea. The analysis methods are based on the Munsell Color Order System and PCCS (Practical Color Coordinate System) tone classification in order to examine the color characteristics. For the color image analysis, the present research performed a positioning on Shigenobu Kobayashi's color images scale with adjectives in order to compare the resulting differences. As a result of the analysis, this research discovered the following; First, achromatic colors were found to be used most frequently in flight attendant uniforms of Korea/China/Japan. In Korean flight attendant uniforms, YR/Y, GY and B/PB/P; in Chinese, R/PB, RP, YR/Y/GY and BG/P; in Japanese, RP, R/P/PB and Y/BG were found in order. As for the main uniform colors, Korean flight attendant uniforms were found to be in the colors YR, and R/GY/B/P; Chinese flight attendant uniforms, R, PB, and P/B; and Japanese flight attendant uniforms, R, BG, B, RP and N. Second, Korean flight attendant uniforms used W and It most frequently; China flight attendant uniforms, W; and Japanese flight attendant uniforms, W and v. Regarding the main colors, Korean flight atteddant uniforms used lt/g and v/p; Chinese flight attendant uniforms, v, dp and s/d/dkg; and Japanese flight attendant uniforms, v/dkg and Bk. Third, after positioning each country's uniform color combination bars on the Kobayashi image scale, Korean flight attendant uniforms showed classic images along with casual/pretty/elegant/chic images; Chinese flight attendant uniforms displayed, casual images as well as, dynamic/gorgeous/chic/cool casual/dandy images; and finally, Japanese flight attendant uniforms converyed dandy images along with casual/gorgeous images. This research findings indicate that Korea/China/Japan airlines' flight attendant uniforms seek for differentiated image establishment by reflecting their own CIs and unique national cultures in the uniform color marketing.

• 디지털콘텐츠학회 논문지
• /
• 제10권3호
• /
• pp.423-431
• /
• 2009

영상을 검색하는 데에는 형태, 색상, 질감 등 여러 가지 특징을 사용 할 수 있다. 그 중 가장 활발한 연구가 이루어지고 있는 분야가 형태 정보와 색상 정보를 이용하는 분야이다. 형태 정보를 이용하는 검색 방법에는 2D 모멘트와 푸리에변환 등의 방법이 유명하다. 또 다른 방법으로는 CSS(Curvature Scale Space)가 있는데, 이는 윤곽선 정보를 이용하여 윤곽선의 굴곡을 2D 그래프로 표현하여 그 그래프의 극대점을 특징 값으로 사용하여 영상을 비교 검색하는 방법이다. 기존 CSS 방법에는 몇 가지 문제점이 있어서 본 논문에서는 기존 방법을 향상시킨 ICSS 방법을 사용하여 영상을 검색한다. 색상 정보를 이용하는 방법에는 RGB 색상정보를 이용하는 방법과 HSI 색상정보를 이용하는 방법 등이 있는데 본 논문에서는 HSI 색상정보를 이용하여 색상 히스토그램으로 표현한 후 영상의 비교 척도로 사용하였다. 영상의 유사도를 측정하는 방법으로는 유클리디언 디스턴스를 주로 사용하는데, 본 논문에서는 정확도와 검색 시간을 단축시키고자 DTW를 사용하여 영상의 유사도를 측정하였고, 유클리디언 디스턴스를 사용했을 때와 비교하여 성능 향상 결과를 보인다.

• 대한가정학회지
• /
• 제41권8호
• /
• pp.159-178
• /
• 2003

The purpose of this research is to analyze the characteristics of colors and color images on fashion web sites, and to provide efficient color informatiom which is useful in color planning and is suitable for brand image on fashion web sites.216 color sample used by 48 fashion web sites were collected and analyzed. The results of this study are as follows; First, dominant colors on fashion web sites are neutral color such as white, gray, black and warm colors such as red, yellow red, yellow. Except blue, cool colors show low usage ratio. Second, the color tones of vivid, strong, light, dull, dark, dark grayish were shown in order. Third, the color images on fashion web sites are casual, modem, romantic, natural, elegant, chic, classic, dandy, pretty, clear, cool casual, dynamic, gorgeous, ethnic and formal.

• 복식
• /
• 제55권3호
• /
• pp.19-32
• /
• 2005

Internet shopping has transformed our daily lives as well as the pattern of consumption. In the word, the development and the growth of online shopping site have led to new pattern of consumption. This applies in particular to clothing, among the product on sale, on the internet. The purpose of this research is to analyze the characteristic of colors and images on internet fashion soho mall web site, and to provide efficient color information which is usefull in color planning and suitable for brand image on fashion web site. 147 color sample used by 40 fashion soho mall web site were collected and analyzed. The results of this study are as follows. First, dominant color on fashion web site is static color as black and it's ratio is 33$\%$. Second, G color is not used. Third, Hue and tone mainly used It tone of P color except V tone. And the color image on internet fashion soho mall web site are modern, chic, dandy, formal. Results from analyzing the fashion soho mall Web site. it is important to unity the company's image but its more important to make a color plan considering the sites feature and the customers's sensitivity.

• 한국정보통신학회:학술대회논문집
• /
• 한국정보통신학회 2012년도 춘계학술대회
• /
• pp.649-651
• /
• 2012

본 논문에서는 슈퍼픽셀특성을 이용한 칼라영상분할을 연구한다. 슈퍼픽셀은 특성이 비슷한 인접화소들을 묶어서 하나의 큰 화소로 취급하는 것으로 고속영상처리 및 영상인식을 위해 사용될 수 있다. 본 연구에서는 슈퍼픽셀특성이 비교적 우수한 $La^*b^*$ 칼라특징공간에서 슈퍼픽셀을 구하고 클러스터링 및 기울기기반 분할 알고리즘을 적용한 영상분할을 연구한다.

• 감성과학
• /
• 제15권2호
• /
• pp.231-238
• /
• 2012

본 연구는 하나의 감각으로 인해 다른 영역의 감각을 불러일으키는 공감각 현상을 기초로 하여 사운드에서 컬러이미지를 유추하여 생성하는 응용 시스템의 구현을 최종 목표로 한다. 이를 위해 사운드의 특징정보인 기본주파수(F0, Fundamental Frequency)에서 음계(Scale) 및 옥타브(Octave) 성분을 추출한 후, HSI 컬러모델인 색상(Hue), 명도(Intensity) 성분에 각각 매핑한다. 본 논문에서 채도(saturation)값은 고정된 값을 사용한다. 이를 다시 RGB 컬러모델로 변환한 후 최종 BMP 포맷으로 컬러 이미지를 출력한다. 본 연구에서 제시한 사운드-컬러이미지 변환 방법을 토대로 기본 변환 시스템을 구현해 본 결과, 기본주파수에서 추출된 음계 및 옥타브 성분의 변화에 따라 색상 및 명도가 상이한 다양한 컬러가 나오는 것을 확인할 수 있었다. 또한 하드웨어적 구현을 위해 TMS320C6713 DSP Board에 포팅하여 실험한 결과 제안된 시스템의 시뮬레이션 결과와 동일한 컬러 이미지가 출력됨을 확인하였다.