• 한국해양공학회지
• /
• 제36권1호
• /
• pp.32-40
• /
• 2022

Underwater optical images face various limitations that degrade the image quality compared with optical images taken in our atmosphere. Attenuation according to the wavelength of light and reflection by very small floating objects cause low contrast, blurry clarity, and color degradation in underwater images. We constructed an image data of the Korean sea and enhanced it by learning the characteristics of underwater images using the deep learning techniques of CycleGAN (cycle-consistent adversarial network), UGAN (underwater GAN), FUnIE-GAN (fast underwater image enhancement GAN). In addition, the underwater optical image was enhanced using the image processing technique of Image Fusion. For a quantitative performance comparison, UIQM (underwater image quality measure), which evaluates the performance of the enhancement in terms of colorfulness, sharpness, and contrast, and UCIQE (underwater color image quality evaluation), which evaluates the performance in terms of chroma, luminance, and saturation were calculated. For 100 underwater images taken in Korean seas, the average UIQMs of CycleGAN, UGAN, and FUnIE-GAN were 3.91, 3.42, and 2.66, respectively, and the average UCIQEs were measured to be 29.9, 26.77, and 22.88, respectively. The average UIQM and UCIQE of Image Fusion were 3.63 and 23.59, respectively. CycleGAN and UGAN qualitatively and quantitatively improved the image quality in various underwater environments, and FUnIE-GAN had performance differences depending on the underwater environment. Image Fusion showed good performance in terms of color correction and sharpness enhancement. It is expected that this method can be used for monitoring underwater works and the autonomous operation of unmanned vehicles by improving the visibility of underwater situations more accurately.

• 한국의류학회지
• /
• 제24권1호
• /
• pp.3-14
• /
• 2000

The purpose of this study were 1) to segment 20's female apparel market using consumer's fashion image preference in formal wear, and 2) to identify the group differences in seasonal color (hue and tone) and color image (image associate with lightness and chroma) preference as well as in demographic variables. The subjects were 253 females in their late twenties living in Seoul, Korea. The data were collected using self-administred questionnaires and analyzed by factor analysis. Cluster analysis, $\chi$2 -test, one-way ANOVA, and Duncan test. The results of this study were as follows: 1) Four fashion image groups were identified through cluster analysis using consumer's fashion image preference: Elegant image group, Sexy image group, Lively image group, and Romantic image group. 2) There were significant differences among fashion image groups in hue preference for spring clothes, tone preferences for spring and fall clothes. Color images are associated with lightness for spring and summer, and are associated with chroma for spring, summer, and fall. Group differences in demographic variables were found in socio-economic status and average expenditure for formal jacket.

• 복식
• /
• 제55권5호
• /
• pp.28-42
• /
• 2005

Objectives of this study were to investigate about the avatar's fashion products efficiency on casual wear advertisements to study about the effect of avatar's clothing image provided by apparel industry and to investigate about the effort of avatar's fashion product on apparel advertisement for fundamental data on the avatar's fashion marketing. Following were the summaries of the results: First, in terms of the correlation between avatar's clothing and casual brand images of nate avatar's fashion marketing, avatar's clothing image of FUBU male, female and maru male, female that is reformative, characteristical, unique and sensitive to latest fashion has positive correlation with FUBU and maru brand images. Therefore, consumers' higher perception on avatar's clothing image that are 'reformative', 'characteristical' and 'unique', indicated higher casual brand image perception, proving avatar's clothing image is effective in suggesting the brand. Second, in terms of advertising the avatar by clothing them with garments of each brand and comparing avatar's clothing and casual brand images, active avatar's clothing image of FUBU male, female emphasized active brand image of FUBU. However, FUBU male avatar's clothing image did not emphasize 'reformative', 'characteristical', 'cool', or 'sensitive to latest fashion' images compared to FUBU female avatar's clothing image. Also, in case of maru, 'male', 'conseuative' and 'insensitive to latest fashion' image of male avatar clothing emphasized maru brand image. Maru female's 'unpractical' ,'female' and 'characteristical' images emphasized maru brand image.

• 대한가정학회지
• /
• 제46권6호
• /
• pp.49-59
• /
• 2008

The purpose of this study was to examine the role of consumers' self-image and pursued-image of clothes on the clothing purchase decision making according to the location. Data were obtained from a questionnaire filled out by 575 women living in Seoul and Jechon. For data comparative analysis, paired t-test, t-test, factor analysis and multiple regression analysis were used. The results of this study are as follows: 1. There were significant differences in self-image and pursued-image in terms of clothing purchases between women who live in Seoul and Jechon residents. 2. Demographic variables influenced to the self-image and pursued-image of clothes factor. Among them, size of the city was the most important factor which influence to the clothing purchase behavior. 3. Self-image, pursued-image of clothes, problem recognition and evaluative criteria factors significantly differed between Seoul and Jechon residents. In two cities, problem recognition factor which was arisen by external stimulus and all of the evaluative criteria factors showed significant differences. 4. When the cities were partitioned by size(large and small city), the influence of self-image and pursued-image of clothes on the clothing purchase behavior showed different phases. Generally, self image and pursued-image of clothes were more important to various problem recognition and evaluative criteria factors in large city(i.e. Seoul) than in small city(i.e. Jechon). However economic rational factor was the exception.

• 한국의류학회지
• /
• 제30권11호
• /
• pp.1608-1617
• /
• 2006

Due to the popularization of fashion, it is important for consumers to find out under what kinds of reasons consumers choose and prefer the clothing products as consumers are interested in clothing and have variety of their styles to choose This study is to classify the self-image of the middle-aged women and examine the characteristics of each group and also to inquire into the evaluation of clothing by each group. Data are collected through a self-administered questionnaire survey from 4 to October 31, 2005 from 350 middle-aged women in Daegu; 275 are used for the data analysis. Data analysis is performed using SPSS Package, which included cluster analysis, factor analysis, ANOVA, Duncan's multiple range test, and chi-square test. The results are as follows: 1. As a result of factor analysis of self-image, the five factors which are intelligent image, social image, fashionable image, female image, bold image are extracted. Besides, as a result of cluster analysis, the three types which are female-type, neuter-type, male-type are classified. 2. The middle-aged women regard the classic style as their best style for outgoing and then they like the casual style, elegant style, dramatic style in order. 3. As a result of factor analysis for clothing image, the four factors which are dignity, attraction, simplicity activity are extracted. 4. According to self-image types, there are differences for clothing image and preferring clothing styles. While female-type groups choose the elegant style, neuter-type groups regard the classic style as their best style and male-type groups regard the casual style as their best style. In case of daring style, the preference shows the lowest among all the types unrelated to self-image types.

• 한국정보과학회논문지:컴퓨팅의 실제 및 레터
• /
• 제7권5호
• /
• pp.510-521
• /
• 2001

영상은 대용량적인 특성과 비정형적인 특성을 가지고 있으므로 신속하고 효율적으로 영상을 검색하기 위해 영상의 정확한 특징정보를 추출하여 검색 시스템을 구축하여야 한다. 영상 검색 시스템은 텍스트 기반의 전통 데이타베이스와는 다른 모델링 방법과 검색방법을 사용한다. 따라서, 영상 검색 시스템에서의 검색속도와 정확도를 향상시키기 위해서는 새로운 영상 데이타베이스 생성기법과 효율적인 검색 기법이 필요하다. 본 논문에서는 입력 영상으로부터 검색에 상용되는 에지 특징정보 추출을 위해 라플라시 안마스크와 입력 영상을 컨벌루션하여 에지의 외곽선 데이타를 추출하였으며, 그리고 추출한 에지 특징정보와 메타데이타로 영상 데이타베이스를 생성하여 신속하고 효율적으로 영상을 검색할 수 있도록 주석기반 및 내용기반 영상 검색 시스템을 구현하였다. 주석기반 및 내용기반 영상 검색 시스템은 영상의 하위 레벨에 표현된 내용기반 에지 특징정보와 특징정보 추출이 어려운 상위레벨에 표현된 주석기반 에지 특징 정보를 영상의 색인으로 구성하여 사용하기 때문에 영상 컨텐츠 검색의 성능을 향상시킬 수 있다. 마지막으로 본 논문에서 제시한 영상 검색 시스템은 메타데이타에 의해 영상 데이타베이스를 구축하므로 정확한 영상 컨텐츠 정보의 축적관리와 영상의 정보공유 및 재이용이 가능하다.

• 복식
• /
• 제62권4호
• /
• pp.58-68
• /
• 2012

Personal image concerns a person's talent, expertise, as well as the internal and external image. It is a core value that differentiates one individual from another. As personal branding via personal image management has become more important, there is a fast-growing number of online systems that provide self-test programs to analyze one's style and habits and also provide expert advice for not only styles but lifestyles as well. This study develops a systematic and objective personal image consulting system and offers basic information for the research of personal image making. For that purpose, the study attempts to examine the present state of global companies that use online image consulting programs and analyze their digital content. The results are as follows: 1) two domestic companies, Colorz and Atzine, and seven foreign companies, notably Covet and Boutique, were brisk in business; 2) two types of personal image-diagnosis programs - Visual search and Virtual matching - are now in operation; and 3) mobile applications exist as an evolved personal image-diagnosis program. With an increased interest in such programs, various companies at home and abroad are establishing systematic and scientific analysis systems, which are needed for personal image-making online. Under these circumstances, domestic companies are also urged to enhance levels of image-diagnosis content and actual commercialization and utilization, to develop programs that enable objectified, systematic personal image-making. To this end, the results of this study may serve as a helpful tool to consider future directions.

• 전자공학회논문지 IE
• /
• 제48권1호
• /
• pp.51-56
• /
• 2011

본 연구에서는 흉부 CR 영상에 대하여 다양한 디지털 영상처리 기법을 가하여 영상의 질을 개선시키고 화질 평가를 수행하였다. 또한 흉부 CR 영상의 선예도를 개선하기 위해서 고주파 강조 필터링과 히스토그램 평활화를 MATLAB으로 구현하여 시뮬레이션 한 결과 고주파 강조 필터링과 히스토그램 평활화를 통해서 원영상의 대조도가 개선되었다. 디지털영상처리에 의해 화질의 개선된 정도를 평가하기 위해서 영상의 관찰에 의한 주관적 평가기법을 이용하였다. 신호 또는 병소가 있는 영상에서 신호 또는 병소를 발견할 확률로 감도를 계산하였다. 고주파 강조 필터링과 히스토그램 평활화가 수행된 영상의 감도는 원영상보다 개선되었고, 의료영상에서 수행된 디지털 영상처리는 영상의 질을 향상시켰다.

• 대한전기학회:학술대회논문집
• /
• 대한전기학회 1999년도 추계학술대회 논문집 학회본부 B
• /
• pp.781-783
• /
• 1999

A storage or transmission of image is difficult without image compression processing because the numbers of generated or reborned image data are very much. In case of the random signal, image compression efficiency is low doing without loss of image information, but compressibility by using JPEG is better. We used Huffman code of JPEG, it assigne the low bit value for data of a lot of generated frequency, assigne the high bit value for data of a small quantity. This paper improved image compression efficiency with transformming Q-factor and certified the results with compressed image. A proposed method is very efficience for continuos a color tone image.

• 한국산업융합학회 논문집
• /
• 제27권3호
• /
• pp.563-572
• /
• 2024

Image processing and computer vision technologies are becoming increasingly important in a variety of application fields that require techniques and tools for sophisticated image analysis. In particular, image segmentation is a technology that plays an important role in image analysis. In this study, in order to identify recent research trends on image segmentation techniques, we used the Web of Science(WoS) database to analyze the R&D topography based on the network structure of the author's keyword co-occurrence matrix. As a result, from 2015 to 2023, as a result of the analysis of the R&D map of research articles on image segmentation, R&D in this field is largely focused on four areas of research and development: (1) researches on collecting and preprocessing image data to build higher-performance image segmentation models, (2) the researches on image segmentation using statistics-based models or machine learning algorithms, (3) the researches on image segmentation for medical image analysis, and (4) deep learning-based image segmentation-related R&D. The scientometrics-based analysis performed in this study can not only map the trajectory of R&D related to image segmentation, but can also serve as a marker for future exploration in this dynamic field.