• 한국방송∙미디어공학회:학술대회논문집
• /
• 한국방송공학회 1999년도 KOBA 방송기술 워크샵 KOBA Broadcasting Technology Workshop
• /
• pp.159-164
• /
• 1999

Homogeneous but distinct visual objects having low-contrast boundaries are usually merged in most of the segmentation algorithms. To alleviate this problem, an efficient image segmentation algorithm based on a bottom-up approach is proposed by using spatial domain information only. For initial image segmentation, we adopt an efficient marker extraction algorithm conforming to the human visual system. Then, two region-merging algorithms are successively applied so that homogeneous visual objects can be represented as simple as possible without destroying low-contrast real boundaries among them. The resultant segmentation describes homogeneous visual objects with few regions while preserving semantic object shapes well. Finally, a size-based region decision procedure may be applied to represent complex visual objects simpler, if their precise semantic contents are not necessary. Experimental results show that the proposed image segmentation algorithm represents homogeneous visual objects with a few regions and describes complex visual objects with a marginal number of regions with well-preserved semantic object shapes.

• 한국주거학회:학술대회논문집
• /
• 한국주거학회 2006년도 추계학술발표대회 논문집
• /
• pp.109-114
• /
• 2006

Since 1970, concentrated development in Gang-nam caused serious differences with Gang-buk, such as housing environment, education, culture, industry, economy, and consumption behavior as well as Urban Infrastructure. In this result, the government has executed New Town project to solve these matters, but the residents' opinion wasn't reflected. The purpose of this study is to research New Town project district residents' consciousness and prefered image for housing and this research will be applied to basic study for housing environment improvement policy in future. The results of this study are summerized as follows : First, living satisfaction is revealed almost low level, and the most demands for improvement are to solve lacks of park and green yard. Second, image of Gang-nam, which of housing environment is better than Gang-buk, is expressed as the words, 'Complicated', 'Complex', 'Polished', 'Convenient', 'Rich'. Third, prefered types of housing are high-rise Residental Complex, Apartment mixed with detached housing, and tower-type apartment. And These are connected with there values

• International Journal of Aerospace System Engineering
• /
• 제2권2호
• /
• pp.79-82
• /
• 2015

The handicap for investigating the aero-optical effect focuses on the accurate prediction on the index refraction fluctuation or density fluctuation. In recent years, with the development of CFD techniques and optical experimental techniques, the comprehension have developed on the aero-optical transmitting effect in many kinds of complex flow. This study mainly introduces the optical aberration in compressible mixing layer. And then the debates about the mechanism of aero-optical effects and assessment of image blur also present.

• Nutrition Research and Practice
• /
• 제13권6호
• /
• pp.521-528
• /
• 2019

BACKGROUND/OBJECTIVES: The aim of this study was to develop Korean food image detection and recognition model for use in mobile devices for accurate estimation of dietary intake. MATERIALS/METHODS: We collected food images by taking pictures or by searching web images and built an image dataset for use in training a complex recognition model for Korean food. Augmentation techniques were performed in order to increase the dataset size. The dataset for training contained more than 92,000 images categorized into 23 groups of Korean food. All images were down-sampled to a fixed resolution of $150{\times}150$ and then randomly divided into training and testing groups at a ratio of 3:1, resulting in 69,000 training images and 23,000 test images. We used a Deep Convolutional Neural Network (DCNN) for the complex recognition model and compared the results with those of other networks: AlexNet, GoogLeNet, Very Deep Convolutional Neural Network, VGG and ResNet, for large-scale image recognition. RESULTS: Our complex food recognition model, K-foodNet, had higher test accuracy (91.3%) and faster recognition time (0.4 ms) than those of the other networks. CONCLUSION: The results showed that K-foodNet achieved better performance in detecting and recognizing Korean food compared to other state-of-the-art models.

• 한국정보통신학회:학술대회논문집
• /
• 한국해양정보통신학회 2007년도 춘계종합학술대회
• /
• pp.209-212
• /
• 2007

디지털 영상은 획득 및 전송과정에서 다양한 잡음에 의해 훼손되며, 이러한 잡음들 중, AWGN과 임펄스 잡음이 가장 대표적이다. 기존의 영상복원 알고리즘들은 대부분 단일 잡음환경하에서 처리되며, 상이한 잡음모델에 대해서는 우수하지 못한 처리결과를 나타내었다. 따라서 본 논문에서는 임펄스 잡음과 AWGN을 분리하여 제거하기 위한 영상복원 알고리즘을 제안하였다 이때, 공간영역에서 마스크 내의 중심화소와 인접한 화소들의 크기차와 공간거리를 파라미터로 사용하여 복합적인 잡음성분을 제거하였다. 시뮬레이션 결과로부터, 제안한 방법은 임펄스 잡음과 AWGN을 제거하는 동시에 에지와 같은 영상의 상세정보를 우수하게 보존하였다.

• 한국의상디자인학회지
• /
• 제19권2호
• /
• pp.145-157
• /
• 2017

This study aimed to investigate the relationship between values orientation and fashion image orientation with a focus on the new senior generation. The study conducted data analysis by using the SPSS 17.0 program for the final 397 copies excluding incomplete questionnaires. The results of this study are as follows. First, values orientation was drawn as the four factors of social orientation, dependent orientation, family-like orientation, and individual orientation. The fashion image orientation was compartmentalized as three factors: urban image, individual image, and feminine image. Second, social orientation had a significant effect on feminine image as shown in Photo 1, while only individual orientation exerted a significant influence on individual image (Photo 2). Social orientation and family-like orientation respectively exercised a significant influence on individual image and feminine image, and feminine image (Photo 3). Third, the four groups did not show a significant difference in the relationship with the sub-elements of fashion image orientation. It appeared that all four groups were felt as a complex image in which modern, feminine, and individual images were mixed. Fourth, the four groups showed a significant difference in the unusual, elegant, refined, youthful, feminine, and modern images among the factors of fashion image orientation.

• 한국인터넷방송통신학회논문지
• /
• 제23권5호
• /
• pp.109-114
• /
• 2023

본 논문에서는 Ka-대역 소형 레이더에 적용하기 위한 65-nm CMOS 기반 2채널 이미지 제거 수신기를 기술하였다. 설계된 수신기는 Low-Noise Amplifier (LNA), IQ mixer 및 Analog Baseband (ABB) 회로로 구성된다. ABB 내에 complex filter를 포함하여 원하지 않는 이미지 성분을 억제할 수 있으며, RF 및 ABB의 가변 이득 증폭기 (VGA)에서 이득을 4.5-56 dB 범위에서 조절할 수 있어 수신기의 동적 영역을 확보할 수 있다. 이득 조절은 수신기에 내장된 SPI 컨트롤러를 통해 수행된다. 수신기 칩은 Ka-대역 목표주파수 내 이득 36 dB에서 잡음지수 <15 dB, OP1dB >4 dBm, 이미지 제거비 >30 dB, 채널 간 격리도 >45 dB 특성을 보였다. 본 수신기는 1.2 V 공급전압에서 420 mA를 소모하며, 칩 면적은 4000×1600 ㎛ 이다.

• 방송공학회논문지
• /
• 제21권3호
• /
• pp.380-390
• /
• 2016

화질 평가 방법은 그동안 많은 방법이 소개되어 왔다. 특히 우수한 성능을 보여주는 무참조 평가에서 기법에서 발전이 지속되어 왔다. 본 논문에서는 쉬어렛 영역에서 자연영상의 통계적 특성에 기반한 무참조 영상화질 평가 방법을 제안한다. 제안 방법은 쉬어릿 계수의 통계 특성으로부터 왜곡에 민감한 특징을 추출한다. 쉬어렛 변환의 복소수 계수로부터 위상과 크기 특징을 얻어낸다. 또한 쉬어렛 변환은 다양한 스케일로 영상을 분석할 수 있기 때문에, 스케일간의 계수의 의존성에 대한 왜곡의 영향을 분석한다. 화질 예측을 위해서 특징들은 SVM(support vector machine)을 이용하여 영상 왜곡 분류 및 화질 예측에 활용된다. 실험결과는 제안 방법이 주관적 평가와의 높은 상관도를 보여주고, 또한 기존 참조 및 무참조 방법보다 우수한 성능을 보여준다.

• 한국멀티미디어학회논문지
• /
• 제13권6호
• /
• pp.833-840
• /
• 2010

In this paper a method for detecting and extracting the face from the image in YCbCr spaceis proposed. The face region is obtained from the complex original image by using the difference method and the face color information is taken from the reduced face region throughthe Ellipse clustering method. The experimental results showed that the proposed method can efficiently detect and extract the face from the original image under the general light intensity except for low luminance.

• 전자공학회논문지B
• /
• 제33B권7호
• /
• pp.60-69
• /
• 1996

To retrieve complex data such as images in multimedia information, we need the content-based retrieval methods based on the visual properties rather than keywords. In this paper, a contrent-based image retrieval system is desinged and implemented to retrieve images using the features of images such as colors, lines and intensity vetor features when a visual query inputs. The contents for image retrievals are the color features extracted from the color component of 16 blocks of the image, th eline features extracted form 4 lines in the image and the shape features extracted from the intensity vectors of the 16 blocks. We can either use a whole image or a sketch image for query. As the experimental results demonstrate the precision 91% the recall 33% and the average rank 3.1 the retrieval performance is found to be high. The experimental results indicate that the retrieval using the weighted features have led to substantial improvement in the percision and performance of system.