• 방송공학회논문지
• /
• 제26권1호
• /
• pp.61-69
• /
• 2021

이미지 아웃페인팅은 이미지의 맥락을 고려하여 주어진 이미지의 외부를 지속적으로 채울 수 있다는 점에서 매우 흥미로운 문제이다. 이 작업에는 두 가지 주요 과제가 있다. 첫 번째는 생성된 영역의 내용과 원래 입력의 공간적 일관성을 유지하는 것이다. 두 번째는 적은 양의 인접 정보로 고품질의 큰 이미지를 생성하는 것이다. 기존의 이미지 아웃페인팅 방법은 일관되지 않고 흐릿하며 반복되는 픽셀을 생성하는 등 어려움을 겪고 있다. 하지만 최근 딥러닝 기술의 발달에 힘입어 기존의 전통적인 기법들에 비해 높은 성능을 보여주고 있는 알고리즘들이 소개되었다. 딥러닝 기반 아웃 페인팅은 현재까지도 다양한 네트워크가 제안되며 활발히 연구되고 있다. 본 논문에서는 아웃 페인팅 분야의 최신 기술 현황 및 동향을 소개하고자 한다. 딥러닝 기반의 아웃페인팅 알고리즘 중 대표적인 네트워크들을 분석하고 다양한 데이터 셋과 비교 방법을 통한 실험 결과를 보여줌으로써 최근 기법들을 비교하고자 한다.

• 한국콘텐츠학회논문지
• /
• 제10권7호
• /
• pp.169-178
• /
• 2010

본고는 영화와 관련된 개념과 이론에 대한 들뢰즈의 사유를 바탕으로 히치콕 영화를 이론적으로 재고찰했다. 이를 통해 영화가 사회과학과 예술학의 학문적 분파로서는 물론 철학으로서의 학문적 의의도 가질 수 있음을 논증하고자 했다. 연구결과에 의하면, 들뢰즈는 히치콕이 지각, 행동, 감화-이미지를 틀 지우고 변형시킬 수 있는 정신-이미지를 영화에 활용함으로써 인물과 인물, 인물과 사건, 사건과 사건 사이의 추상적이고 자연적인 관계를 포함해 그 관계가 전개되는 방식도 이미지로 대상화했음을 알 수 있었다. 또한 들뢰즈는 히치콕 영화를 통해 정신-이미지가 영화의 의미를 전환시키고, 영화적 본질 또는 정신적 관계항으로서의 카메라가 이미지를 특정 형식으로 틀 지우는 동시에 그 이미지에 침투해 변형시킨다는 점을 강조하고 있었다. 본고는 들뢰즈의 영화에 대한 사유를 영화이론에 일반화하는데 적지 않은 문제가 있음을 논의의 한계로 제한하고 있지만, 영화와 철학의 관계에 대한 실천적 논의를 위한 이론적 성찰의 계기로서 새로운 연구문제를 탐색하는데 의의를 두고 있다.

• 패션비즈니스
• /
• 제12권4호
• /
• pp.114-130
• /
• 2008

The purpose of this study is to investigate the effect of Dot Pattern Size(0.8, 1.8, 2.5, 5, 8), color combination(W/Bk, Bk/Gr, Gr/W), Area-Ratio(Background/Dot, Dot/Background) on wearing dot-printed dresses image. Sets of stimulus and response scales(7 point semantic) were used as experimental materials. The stimuli were 30 color pictures manipulated with the combination of Dot Pattern Size, color combination, and Area-Ratio using computer simulation. The subjects were 180 female undergraduates living in Gyeongnam-do. The data was analyzed by using SPSS program. Analyzing methods were ANOVA and LSD test. Image factor of the stimulus was composed of 5 different components, visibility, chastity, attractiveness, cuteness and feminity. Among them, the visibility and chastity were important. Each dimensional image was affected by dot pattern size, color combination and Area-Ratio. In the visibility image, color combination(W/Bk is the most effective) is more influential, the larger size is effective pattern. In the cuteness and feminity image, area ratio(low-brightness dot pattern is the more effective) is more effective than color combination or dot pattern size. Even the same dot pattern size and area was recognized as different image depending on the area ratio. According to the variation of dot pattern size, color combination and area-ratio, it was investigated that the images for a dress wearer were expressed diversely, were shown differently in image dimensions, and could be produced to different images.

• 대한안전경영과학회지
• /
• 제23권2호
• /
• pp.65-77
• /
• 2021

This paper conducted an empirical study to identify the causal relationship of factors affecting the purchase intention of automobiles from the customer's point of view. This study sets the purchase intention as a result variable and constructs a causal model with brand image, product attributes (exdogenous variable), and customer value (endogenous variable) as a cause that affects purchase intention. The results of this study are summarized as follows. First, the symbolic image of the brand was found to have a very significant effect on customer value (p=0.01), and the external attribute of the product also had a significant effect on customer value (p=0.1). Second, customer value was found to have a very significant effect on purchase intention (p=0.01), and the functional image of the brand also had a significant effect on purchase intention (p=0.1). Third, there is a strong positive (+) correlation between the functional image of the brand and the symbolic image of the brand, the intrinsic attribute of the product, and the external attribute of the product, and also between the symbolic image of the brand and the intrinsic attribute of the product and There was also a positive (+) correlation between extrinsic attributes. Therefore, in order to increase customer value, automobile manufacturing companies have a functional value of products from a customer-oriented perspective. It is judged that every effort should be made to maintain a lasting relationship by grasping the values of customers, which are social values, emotional values, situational values, and cognitive values.

• 전기전자학회논문지
• /
• 제22권4호
• /
• pp.1079-1087
• /
• 2018

최근 딥러닝(deep learning) 인공지능 기반의 컴퓨터 비전 분야는 각종 영상분석 분야에서 화제로 떠오르고 있다. 본 연구에서는 딥러닝 기반의 여러 이미지 인식 알고리즘 중 이미지 내에서 객체를 검출하는 데 사용되는 Faster R-CNN 알고리즘을 이용하여 화재 이미지에서 불꽃을 검출하고자 한다. 학습 과정에서 소량의 데이터셋을 통한 화재검출 정확도 향상을 위해 이미지 오그멘테이션(image augmentation) 기법을 이용하고, 이미지 오그멘테이션을 6가지 유형별로 나누어 학습하여 정확도, 정밀도, 검출률을 비교하였다. 그 결과, 이미지 오그멘테이션의 종류가 늘어날수록 검출률이 상승하지만, 다른 객체 검출 모델들의 일반적인 정확도와 검출률의 관계와 마찬가지로 오검출율 또한 10%에서 최대 30%까지 증가하게 됨을 확인하였다.

• International Journal of Control, Automation, and Systems
• /
• 제3권4호
• /
• pp.542-551
• /
• 2005

A new distance measurement method with the use of a single camera and a rotating mirror is presented. A camera in front of a rotating mirror acquires a sequence of reflected images, from which distance information is extracted. The distance measurement is based on the idea that the corresponding pixel of an object point at a longer distance moves at a higher speed in a sequence of images in this type of system setting. Distance measurement based on such pixel movement is investigated. Like many other image-based techniques, this presented technique requires matching corresponding points in two images. To alleviate such difficulty, two kinds of techniques of image tracking through the sequence of images and the utilization of multiple sets of image frames are described. Precision improvement is possible and is one attractive merit. The presented approach with a rotating mirror is especially suitable for such multiple measurements. The imprecision caused by the physical limit could be improved through making several measurements and taking an average. In this paper, mathematics necessary for implementing the technique is derived and presented. Also, the error sensitivities of related parameters are analyzed. Experimental results using the real camera-mirror setup are reported.

• 복식문화연구
• /
• 제16권5호
• /
• pp.841-851
• /
• 2008

The purpose of the study was to examine the effect of motif categories and motif colors on evaluating sensory image of fashion materials with the gestalt theory as the background. The research was conducted on a quasi experimental basis, with subjects numbering 187 male and 207 female college students. Data were collected in the period from march 19th to march 31st, 2007. A set of fabric stimuli and semantic differential scales were developed. The stimuli were thirteen fabric species(each measuring 12 by 13cm). Variables included; (a) motif colour(white, grey, pink and blue) (b) motif categories(plain, paisley, flower, stripes and zebra effect). The semantic differential scale to measure sensory image of fabric stimuli included 23 sets of bi-polar adjectives. The data were analysed by factor analysis and ANOVA and the major finding were as follows. 1) Four sensory dimensions emerged of importance: salience, attractiveness, comfort and softness. 2) The motif category effected on the four sensory image dimensions while the motif colour effected on salience, comfort and softness sensory dimensions. 3) An interaction effect was founded between motif category and motif colour. 4) Motif category showed significant effects on the preference and liking of the fashion, however the motif colour did not show any significant effects on the preference and liking. As a whole the results supported the gestalt theory and the results can be used for the marketing strategy for developing fashion fabrics.

• 대한의용생체공학회:학술대회논문집
• /
• 대한의용생체공학회 1991년도 춘계학술대회
• /
• pp.18-21
• /
• 1991

In this paper, the effect of background grey levels on the visual perception of target image displayed on CRT monitor has been investigated. The purpose of this study is to investigate the efficacy of CRT monitor as a display medium of image information especially in medical imaging field. Three sets of experiments have been performed in this study; the first was to measure the luminance response of CRT monitor and to find the best fitting equation, and the second was the psychophysical experiment measuring the threshold grey level difference between the target image and the background required for visual discrimination for various background grey levels, and the third was to develop a visual model that is predictable of the threshold grey level difference measured in the psychophysical experiment. The result of psycophysical experiment shows that the visual perception performance is significantly degraded in the range of grey levels lower than 50, which is turned out due to the low luminance change of CRT monitor in this range while human eye has been adapted to relatively bright ambient illumination.

• 대한의용생체공학회:의공학회지
• /
• 제28권2호
• /
• pp.294-299
• /
• 2007

For efficient and accurate diagnosis of ultrasound images, appropriate time gain compensation(TGC) and dynamic range(DR) control of ultrasound echo signals are important. TGC is used for compensating the attenuation of ultrasound echo signals along the depth, and DR controls the image contrast. In recent ultrasound systems, these two factors are automatically set by a system and/or manually adjusted by an operator to obtain the desired image quality on the screen. In this paper, we propose an algorithm to find the optimized parameter values far TGC and DR automatically. In TGC optimization, we determine the degree of attenuation compensation along the depth by dividing an image into vertical strips and reliably estimating the attenuation characteristic of ultrasound signals. For DR optimization, we define a novel cost function by properly using the characteristics of ultrasound images. We obtain experimental results by applying the proposed algorithm to a real ultrasound(US) imaging system. The results verify that the proposed algorithm automatically sets values of TGC and DR in real-time such that the subjective quality of the enhanced ultrasound images may be sufficiently high for efficient and accurate diagnosis.

• 한국정보처리학회논문지
• /
• 제7권9호
• /
• pp.3029-3036
• /
• 2000

영상분할은 컴퓨터비전 시스템에서 영상정보추출의 중요한 과정 중의 하나이다. 이중에서 퍼지 클러스터링 방법은 영상분할에 광범위하게 사용되고 있다. 대부분의 퍼지 클러스터링 방법으로는 FCM 알고리즘이 사용된다. 그러나 FCM 알고리즘은 클러스터의 중심과 데이터간의 거리에 의존하기 때문에 클러스터 크기가 다를 경우에는 데이터가 오분류될 수 있다. 본 논문에서는 클러스트 크기에 상관없이 데이터를 분류할 수 있는 평균내부거리를 이용한 퍼지 클러스터링 알고리즘을 제안하였다. 평균내부거리는 각 데이터로부터 해당 클러스터 중심까지의 거리를 평균한 값으로 클러스터의 크기와 밀도에 비례한다. 실험 결과를 통하여 제안된 방법이 분류 엔트로피와 적합도 함수에 의해서 좋은 결과를 보여주고 있음을 증명하였다.