• 한국의류산업학회지
• /
• 제22권6호
• /
• pp.739-751
• /
• 2020

This study compares and analyzes bridal wedding styles in Korean and Chinese SNS. For this, 715 photos (362 in Korea and 353 in China) collected on social media (Instagram and Xiaohongshu) were used for analysis. The bridal wedding style was divided by item into wedding dress, makeup, and hairstyle, and detailed characteristics of images and designs were examined through content analysis and statistical analysis. First, in the case of Korea, many wedding dresses showed ball gowns and mermaid silhouettes, white colors, and no details and trimmings. As for the makeup, transparent and natural skin expression, straight eyebrows, and pink and peach lip colors were highlighted; and as for hairstyles, many long hair with half-covered ears appeared. Second, in China, a ball gown, mermaid silhouette, and white color are often seen in wedding dresses, and point colors and use of various details and trimmings are noted. The makeup showed a lot of bright skin expression, brown and black eyebrows, and red and brown lip colors. As for hairstyle, a lot of long hair with completely exposed ears appeared. Third, the comparison of bridal wedding styles between Korea and China indicated that China used relatively more details and trimmings in wedding dresses. As for makeup, Korea has a natural image, and China has a classic image. As for the hairstyle, there were many styles in which Koreans had half the ears covered, and Chinese had no bangs and completely exposed ears.

• 한국콘텐츠학회논문지
• /
• 제6권4호
• /
• pp.11-18
• /
• 2006

본 논문에서는 두 가지 새로운 임펄스 잡음 검출기를 설계하고 총변량(total variation) 최적화를 통하여 영상에 존재하는 임펄스 잡음을 제거하는 방법을 제안한다. 설계된 임펄스 잡음 검출기는 적응 미디언 필터(AMF:Adaptive Median Filter)를 기반으로 하고 있으며 기존의 검출기에 비해 잡음검출의 정확도가 높고 영상의 디테일 정보를 잡음으로 오인하는 확률을 줄였다. 또한 제안하는 검출기는 잡음발생 확률에 무관하게 우수한 성능을 유지한다. 영상에서의 잡음제거는 제안된 검출기에 의해 얻어진 잡음후보 화소에 대해서만 총변량 최적화를 적용하므로 불필요한 계산을 줄이고 영상의 경계선을 잘 보존하면서 잡음을 제거할 수 있다. 실험결과 제안하는 2단 구조의 잡음제거 알고리즘은 다양한 형태의 잡음 밀도에 대해서 기존의 알고리즘에 비해 약 2dB 정도의 화질개선 효과를 얻었다.

• 한국멀티미디어학회논문지
• /
• 제22권11호
• /
• pp.1242-1250
• /
• 2019

Super resolution technique aims to convert a low-resolution image with coarse details to a corresponding high-resolution image with refined details. In the past decades, the performance is greatly improved due to progress of deep learning models. However, universal solution for various objects is a still challenging issue. We observe that learning super resolution with a general dataset has poor performance on faces. In this paper, we propose a super resolution fusion scheme that works well for both general- and face datasets to achieve more universal solution. In addition, object-specific feature extractor is employed for better reconstruction performance. In our experiments, we compare our fusion image and super-resolved images from one- of the state-of-the-art deep learning models trained with DIV2K and FFHQ datasets. Quantitative and qualitative evaluates show that our fusion scheme successfully works well for both datasets. We expect our fusion scheme to be effective on other objects with poor performance and this will lead to universal solutions.

• 한국정보과학회논문지:컴퓨팅의 실제 및 레터
• /
• 제16권3호
• /
• pp.366-370
• /
• 2010

본 논문에서는 한 장의 2-D 이미지를 입력으로 받아 자동적으로 동양화처럼 보이는 그림으로 바꾸어 주는 비 사실적 랜더링(NPR) 시스템을 제안한다. 동양화의 선은 두께가 다양하고 끝이 갈라지는 특성이 있고, 농담의 변화를 주어 대상 물체의 입체감을 표현하는 경향이 있다. 본 논문에서는 이미지의 외곽선 추출을 바탕으로 curvature를 사용하여 선을 따라 변하는 두께를 가진 선을 생성한다. 또한 기존에 수채화 스타일에서 제안된 농담 표현방법과는 반대로 지역적인 세부표현을 증가시키고 광역적인 농담을 줄여준다. 본 논문은 외곽선 추출, 선 그리기를 통하여 다양한 두께의 선을 표현해 주는 방법과, 이미지의 필터링과 intensity 변환을 통한 농담을 표현하는 일련의 연속적인 방법을 제안하다.

• International journal of advanced smart convergence
• /
• 제9권3호
• /
• pp.85-96
• /
• 2020

It is important to extract salient object image and to solve the invariance problem for image recognition. In this paper we propose a new hybrid algorithm for invariance and improved classification performance in image recognition, whose algorithm is combined by FT(Frequency-tuned Salient Region Detection) algorithm, Guided filter, Zernike moments, and a simple artificial neural network (Multi-layer Perceptron). The conventional FT algorithm is used to extract initial salient object image, the guided filtering to preserve edge details, Zernike moments to solve invariance problem, and a classification to recognize the extracted image. For guided filtering, guided filter is used, and Multi-layer Perceptron which is a simple artificial neural networks is introduced for classification. Experimental results show that this algorithm can achieve a superior performance in the process of extracting salient object image and invariant moment feature. And the results show that the algorithm can also classifies the extracted object image with improved recognition rate.

• 제어로봇시스템학회:학술대회논문집
• /
• 제어로봇시스템학회 1996년도 한국자동제어학술회의논문집(국내학술편); 포항공과대학교, 포항; 24-26 Oct. 1996
• /
• pp.105-108
• /
• 1996

In this paper, we present an interpolation schema for image resolution enhancement using fuzzy logic. Proposed algorithm can recover both low and high frequency information in image data. In general, interpolation techniques are based on linear operators which are essentially details in the original image. In our fuzzy approach, the operator itself balances the strength of its sharpening and noise suppressing components according to the properties of the input image data. The proposed interpolation algorithm is performed in three step. First logic reasoning is applied to coarsely interpret the high frequency information. These results are combined to obtain the optical output. Using our approach, resolution of the original image can be applied to various kind of image processing topics such as image enhancement, subpixel edge detection, and filtering.

• 대한가정학회지
• /
• 제49권8호
• /
• pp.13-24
• /
• 2011

This paper purports to find the characteristics of blue costumes as expressed in modern fashion. As such, we made an empirical analysis, using 436 works from 2006 S/S through 2010 F/W collections. Result indicated that modernness was emphasized by applying a minimum expressive technique to items of a simple form, and an image of simplicity was emphasized by placing focus on the rational color image of blue. Second, images of blue, such as trust, stability, and success were expressed as refined elegance through the combination of tight silhouettes, glossy and translucent materials, and indirect exposure. Third, natural image was expressed through natural silhouettes, soft or natural materials, gradation of various tones, and natural motif decorations as well as the color itself. Fourth, dynamic image was expressed by introducing fashion items, mostly worn by men only, to simple and tight forms or through the inspiration of a masculine fashion image. Fifth, abstinent image was expressed by using an image of religious apparel or symbols, covering up the body, minimizing details, and using a dark blue tone.

• Journal of information and communication convergence engineering
• /
• 제20권3호
• /
• pp.195-203
• /
• 2022

Noise generated during image acquisition and transmission can negatively impact the results of image processing applications, and noise removal is typically a part of image preprocessing. Denoising techniques combined with nonlocal techniques have received significant attention in recent years, owing to the development of sophisticated hardware and image processing algorithms, much attention has been paid to; however, this approach is relatively poor for edge preservation of fine image details. To address this limitation, the current study combined a steering kernel technique with adaptive masks that can adjust the size according to the noise intensity of an image. The algorithm sets the steering weight based on a similarity comparison, allowing it to respond to edge components more effectively. The proposed algorithm was compared with existing denoising algorithms using quantitative evaluation and enlarged images. The proposed algorithm exhibited good general denoising performance and better performance in edge area processing than existing non-local techniques.

• 복식
• /
• 제52권1호
• /
• pp.103-115
• /
• 2002

The purpose of this study is understanding modern woman's various aesthetic values and aesthetic sense through observing expression aspects of gender images in the 1990s military fashion and analyzing their aesthetic characteristics and formative factors. Gender images in military fashion have expressed the masculine image of authoritative image and offensive image and offensive image, and the feminine image of conservative image and ostentatious image, sensual image. The authoritative image showed authority and dignity of military uniform by using the classic military uniform's image. This spoke for desire of women to rise their position. Heroism and androcentrism affected as its formation factor. The offensive image destroyed original dignified image of military uniform by that resistance to authoritative image and existing gender identity appeared as way-out form and deconstructive expression. Anti-establishment spirit, resistance to gender identity, and deconstructionism affected as its formation factor. The conservative image expressed military fashion only with slim and soft silhouette, curved line, color, and simple details by magnifying feminity. Fallen man's authority affected as its formation factor. The ostentatious image expressed military fashion with magnifying accessories such as gold button and belt by women who wanted to display ostentatiously their social position and charm. The aspiration for class of elite affected as its formation factor. The sensual image intended to show erotic voluptuous beauty of woman's body by indirect and direct body exposure. Narcissism, desire to show, and sexual amusement affected as its formation factor.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제13권7호
• /
• pp.3620-3637
• /
• 2019

Since the birth of Synthetic Aperture Radar (SAR), it has been widely used in the military field and so on. However, the existence of speckle noise makes a good deal inconvenience for the subsequent image processing. The continuous development of sparse representation (SR) opens a new field for the speckle suppressing of SAR image. Although the SR de-noising may be effective, the over-smooth phenomenon still has bad influence on the integrity of the image information. In this paper, one novel SAR image de-noising method based on residual image fusion and sparse representation is proposed. Firstly we can get the similar block groups by the non-local similar block matching method (NLS-BM). Then SR de-noising based on the adaptive K-means singular value decomposition (K-SVD) is adopted to obtain the initial de-noised image and residual image. The residual image is processed by Shearlet transform (ST), and the corresponding de-noising methods are applied on it. Finally, in ST domain the low-frequency and high-frequency components of the initial de-noised and residual image are fused respectively by relevant fusion rules. The final de-noised image can be recovered by inverse ST. Experimental results show the proposed method can not only suppress the speckle effectively, but also save more details and other useful information of the original SAR image, which could provide more authentic and credible records for the follow-up image processing.