• Fashion & Textile Research Journal
• /
• v.6 no.5
• /
• pp.549-557
• /
• 2004

The purpose of this study is to investigate design properties affecting in evaluated image by adult women's town wear in un-limited circumstance. The stimulus, adult women's town wear, were collected from shopping mall, department stores and churches and evaluated by 20's 150 people. Through the estimations, the 76 pictures of 'good image' and 65 pictures of 'bad image', were selected and analyzed by classification categories. The results were as follows : 1) 'Good Image' is classified 6 groups which are like active casual, feminine casual, adult casual, modern, sporty casual, and elegance. 2) "Bad Image' is classified 5 groups which are like easy casual, active casual, soft casual, modern casual, and feminine casual. 3) Central code of adult women's town wear from 'Good Image' are simple, bright, and harmony and 'Bad Image' are complicate, heavy, and inharmony. The coordination, how to wear, is very important to evaluate image of women's town wear with other properties. Also, body shape appeared by important variable in evaluation.

• Proceedings of the Korean Information Science Society Conference
• /
• 2006.10c
• /
• pp.218-221
• /
• 2006

본 논문은 인터넷의 대표적인 문제점중의 하나인 Adult Image 분류 연구에 대해 기술한다. 특히 우리는 이러한 Adult Image를 분류하기 위한 Data Set을 5가지 타입으로 구성한다. 이러한 각 Image에 대해 Color, Gradient, Edge Direction 특성의 Feature들을 추출하고 이를 Histogram으로 구성한다. 이렇게 구성된 Histogram을 Support Vector Machine에 적용하여 Adult Image를 분류한다. 그 결과, 우리는 8250개의 Test Set에 대하여 Recall(96.53%), Precision(97.33%), False Positive(2.96%), F-Measure(96.93%)의 성능 결과를 보여준다.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.4 no.5
• /
• pp.910-924
• /
• 2010

As Internet usage has increased, the risk of adolescents being exposed to adult content and harmful information on the Internet has also risen. To help prevent adolescents accessing this content, a novel detection method for adult images is proposed. The proposed method involves three steps. First, the Image Of Interest (IOI) is extracted from the image background. Second, the IOI is distinguished from the segmented image using a novel weighting mask, and it is determined to be acceptable or unacceptable. Finally, the features (color and texture) of the IOI or original image are compared to a critical value; if they exceed that value then the image is deemed to be an adult image. A Receiver Operating Characteristic (ROC) curve analysis was performed to define this optimal critical value. And, the textural features are identified using a gray level co-occurrence matrix. The proposed method increased the precision level of detection by applying a novel weighting mask and a receiver operating characteristic curve. To demonstrate the effectiveness of the proposed method, 2850 adult and non-adult images were used for experimentation.

• The Journal of the Korea Contents Association
• /
• v.9 no.4
• /
• pp.1-11
• /
• 2009

To classifying and filtering of adult images, in recent the computer vision techniques are actively investigated because rapidly increase for the amount of adult images accessible on the Internet. In this paper, we investigate and develop the tool filtering of adult images using skin color model. The tool is consisting of two steps. In the first step, we use a skin color classifier to extract skin color regions from an image. In the nest step, we use a region feature classifier to determine whether an image is an adult image or not an adult image depending on extracted skin color regions. Using histogram color model, a skin color classifier is trained for RGB color values of adult images and not adult images. Using SVM, a region feature classifier is trained for skin color ratio on 29 regions of adult images. Experimental results show that suggested classifier achieve a detection rate of 92.80% with 6.73% false positives.

• Proceedings of the IEEK Conference
• /
• 2006.06a
• /
• pp.795-796
• /
• 2006

Recently, blocking harmful contents such as adult images has been widely researched. However, most of adult image blocking methods use flush color model without shape information. In this paper, we present a new adult image blocking methods based on shape and color information. In the first step, the shape is considered. In the second step, adult images are detected using skin color model. Considering both shape and color, the detection rate can be increased. We evaluate adult image detection performance using sample images.

• Journal of Fashion Business
• /
• v.8 no.2
• /
• pp.69-90
• /
• 2004

The purpose of this study is to find the recognition of Korean image fashion design by U.K fashion specialists. U.K is one of the important countries in design field including fashion design since 1960. For this purpose, literature research and focus individual interview research were carried. First, through the researches precedent, it was found that a national image is related with it's design industry and what is Korean image fashion design, were studied. Second, for individual interviews to 13 U.K fashion specialists, who are teaching and researching in University that has postgraduate course over M.A and had industrial experiences from 7years to 22years, were progressed using open questions and visual image stimulus. The open questions were consisted with four parts : personal educational and industrial background, recognitions about oriental fashion, recognitions about Korean image and Korean fashion design before seeing the visual stimulus, recognition about Korean image fashion design and the characteristics of Korean after seeing the visual stimulus. The results are as follows; First, the 12 U.K specialists have recognized 'oriental fashion' is one of important fashion trends now a days. Japan and Japanese designers are recognized as a represented nation and designer in oriental fashion by them. Two of the specialists referred to need changing the term 'oriental' because the term has been used in the sights of western from colonial age and Japanese is not included the oriental any more. Secondly, 11 interviews have recognized nothing about the Korean national image some of them has negative image due to political situation in Korean Peninsula. However 2 interviews who had been Korea before has positive image. In the questions about Korean fashion and Korean fashion designers, 10 of 13 interviews have nothing and negative recognitions. So it was founded that Korean fashion design was recognized as a lower level by U. K. fashion specialists. Thirdly, in the questions about Korean fashion image and the design characteristics of Korean fashion after seeing the visual stimulus, the response was represented two directions. One is about over decorative image through ethnic design and the other is about simple image differ from Japanese. The 13 interviews felt the Korean Image fashion design such like traditional, decorative, opulent, flat cutting, fresh proportion, loose, layering, natural, simplicity, complicate, adventure, easy, stylish, soft, feminine, young image, adult sexy image. The images were analyzed five image groups : adult sexy image, adult ethnic image, natural image, young avant-garde image, young simple casual image. No one preferred the adult sexy image, adult ethnic image and natural image. However 10 interviews preferred young avant-garde group and 13 interviews preferred the young simple casual image. So this group can be understanded and useful informed as one of competitive power in global fashion industry.

• Journal of the Korean Society for Precision Engineering
• /
• v.19 no.12
• /
• pp.100-108
• /
• 2002

This paper describes the development of adult authentication system using numerical recognition. Nowadays, the automats are very popular and they are dealing in many item suck as coffee, soft drinks, alcoholic drinks and cigarettes, etc. Among these items, some are harmful to the minor, and so the sale of these to the minor must be prohibited. In relation to this, adult authentication system is required to be equipped to the automat which deals in items harmful to minor. According to these demands, we develop the adult authentication system. This system capture the image of a residence certificate card by the identification card-reader, and recognize its numbers and identify it as adult or minor by main computer, where numeral recognition is accomplished by using image processing methods and neural network recognizer. The characteristic test of the system is carried out, and its result reveals that the system has the error of less than 1%. Thus, It is thought that the system can be used for identifying adult in the automats.

• Fashion & Textile Research Journal
• /
• v.8 no.1
• /
• pp.48-54
• /
• 2006

This study was designed to find out how the sociocultural attitudes toward appearance, BMI(Body Mass Index) of adult women affect their body image and body satisfaction. A questionnaire was prepared in the survey and a total of 456 adult women were selected by way of stratified random sampling. The research findings are as follows. The sociocultural attitudes toward appearance were shown in two factors of 'internalization' and 'recognition' The body image was shown in three dimensions of 'care for appearance,' 'concern about weight,' and 'appearance attractiveness'. Internalization, recognition, and BMI affected care for appearance. Concern about weight was affected by internalization, and BMI. Appearance attractiveness was affected only by BMI. Most people were not satisfied with their weight and overall body shape in spite that they in fact had normal or lean figures except 3.3% of the total samples. From the above research findings, adult women's satisfaction with their body was shown high when they were subjectively satisfied with their appearance attractiveness. And the results indicate that adult women are under pressure of socioculturally distorted image of beauty.

• Annual Conference of KIPS
• /
• 2007.11a
• /
• pp.121-123
• /
• 2007

현재까지 유해한 컨텐츠(Contents)를 차단하기 위한 활발한 연구가 있었으나, 사람의 사운드(sound)와 이미지(image)를 통합한 필터링(filtering) 기법에 대한 연구는 활발히 이루어지지 않은 측면이 있다. 본 논문은 이미지(image) 데이터 중 피부색 분포 비율과 사운드(sound) 데이터 중 주파수 분석을 통한 심층적인 기법을 활용하여 현재까지 진행되고 있는 이미지 필터링(image filtering)방법에 대한 수행 결과보다 획기적으로 개선된 성능을 보이고자 한다. 즉, 사운드와 이미지의 특징 정보를 이용한 성인 컨텐츠(Adult Contents)분류 기법을 활용하는 것으로 성인 컨텐츠(Adult Contents)에서 두드러지는 특징을 보이는 사운드 패턴을 분석하여 현재까지 한정된 자원인 이미지만을 활용한 기법보다는 현저한 향상된 수행능력을 예측해 볼 수 있다.

• The Journal of Pediatrics of Korean Medicine
• /
• v.22 no.2
• /
• pp.155-170
• /
• 2008

Objectives : The purpose of this study was to have better data and to make efficient clinical reviews about children's growth based on two methods; X-ray of hand, and ultrasound of calcaneus. Methods : The study was composed of 100 patients(50 of boys and 50 of girls) who visited in department of pediatrics, O O university oriental hospital considering growth and their development. Bone age was measured by the TW3 method's RUS score in simple X-ray image of hand and bone density through ultrasonic image of calcaneus. To predict children's estimated height, their parent's height, bone age, and present height's percentile was measured. Results and Conclusions : 1. The bone age results from X-ray image of hand and ultrasonic image of calcaneus were correlated. Younger children have older bone age from hand's X-ray than the one from calcaneus's ultrasound. Older children have older bone age from calcaneus's ultrasound than one from hand's X-ray. 2. Predicted adult height by bone age(BH) of hand's X-ray and BH of calcaneus's ultrasound were correlated, but predicted adult height by inheritance(IH) was not correlated with others. 3. Bone age and predicted adult height were correlated with age, height, weight, BMI, and especially with age and height. 4. Measuring bone age by X-ray image of hand and ultrasonic image of calcaneus was simple and effective way of estimating children's growth and development.