• Journal of Information Processing Systems
• /
• 제15권1호
• /
• pp.7-21
• /
• 2019

In order to improve the edge segmentation effect of the level set image segmentation and avoid the influence of the initial contour on the level set method, a saliency level set image segmentation model based on local Renyi entropy is proposed. Firstly, the saliency map of the original image is extracted by using saliency detection algorithm. And the outline of the saliency map can be used to initialize the level set. Secondly, the local energy and edge energy of the image are obtained by using local Renyi entropy and Canny operator respectively. At the same time, new adaptive weight coefficient and boundary indication function are constructed. Finally, the local binary fitting energy model (LBF) as an external energy term is introduced. In this paper, the contrast experiments are implemented in different image database. The robustness of the proposed model for segmentation of images with intensity inhomogeneity and complicated edges is verified.

• 한국정보통신학회:학술대회논문집
• /
• 한국정보통신학회 2022년도 춘계학술대회
• /
• pp.367-369
• /
• 2022

본 논문에서는 훈련된 모델이 분류에 실패한 이미지들에 대한 정확한 결과를 확인하기 위해 다중 클래스의 이미지 분류를 구현하면서 각각의 클래스에 맞게 레이블 매핑을 하여 예측 결과를 확인했다. Kaggle의 Intel Image Classification 데이터셋을 사용하여 CNN 모델을 구축하고 훈련을 진행하였으며, 테스트 데이터셋의 이미지들을 레이블 매핑을 통해 다중 클래스의 이미지들이 매핑된 레이블 값과 모델이 분류한 값을 비교하였다.

• 대한간호학회지
• /
• 제39권4호
• /
• pp.539-548
• /
• 2009

Purpose: This study was designed to construct and test the structural equation model on sexual satisfaction in patients with a colostomy. Methods: The model construction was based upon Roy's adaptation model. Stoma-related discomfort (SRD), age, frequency of sexual intercourse, treatment modality, and gender affect sexual satisfaction and are mediated by physical, psychosocial, and interdependence modes. Each mode was conceptualized as sexual function, body image/ depression, and marital intimacy. The patients were 112 colostomates with colorectal cancer who were asked to complete a mail-back survey on their demographic data, SRD, body image, depression, marital intimacy, sexual function, and sexual satisfaction. Data were analyzed using SPSS WIN 15.0 and AMOS WIN 7.0. Results: Significant variables for sexual satisfaction in the final model were body image affected by SRD, depression affected by body image and SRD, marital intimacy affected by depression, and sexual function affected by marital intimacy. Conclusion: The results of this study suggest that specific guidelines for SRD are necessary to improve sexual satisfaction among colostomates. Nurses should be vigilant in monitoring depression and body image disturbance, and providing appropriate interventions to increase marital intimacy. Treatment modality, gender, and age should be considered in developing education programs pertaining to sexuality.

• 한국방송∙미디어공학회:학술대회논문집
• /
• 한국방송공학회 2009년도 IWAIT
• /
• pp.368-372
• /
• 2009

Pin-hole model has been widely used as a robust tool for easily understanding how to obtain a stereo image and how to present the depth-cue to an observer in stereoscopy. However, most of the processes to analyze depth cue in stereoscopy are performed that a stereo image is taken by camera model practically but depth cue of the image is analyzed by pin-hole model. Therefore, the result of depth cues by the process to be uncorrected. The reason of the uncorrected depth cue is led to the image distances of camera model due to variable focused object distances, and it makes a depth distortion. In this paper, we tried to show the contradiction such as occurring depth distortion in the process which the pin-hole model is used to analyze depth cue despite practical camera model is used in stereoscopy, and we presents the method to overcome the contradiction.

• Journal of information and communication convergence engineering
• /
• 제19권2호
• /
• pp.79-83
• /
• 2021

This paper proposes a novel image classification method based on few-shot learning, which is mainly used to solve model overfitting and non-convergence in image classification tasks of small datasets and improve the accuracy of classification. This method uses model structure optimization to extend the basic convolutional neural network (CNN) model and extracts more image features by adding convolutional layers, thereby improving the classification accuracy. We incorporated certain measures to improve the performance of the model. First, we used general methods such as setting a lower learning rate and shuffling to promote the rapid convergence of the model. Second, we used the data expansion technology to preprocess small datasets to increase the number of training data sets and suppress over-fitting. We applied the model to 10 monkey species and achieved outstanding performances. Experiments indicated that our proposed method achieved an accuracy of 87.92%, which is 26.1% higher than that of the traditional CNN method and 1.1% higher than that of the deep convolutional neural network ResNet50.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제14권1호
• /
• pp.366-381
• /
• 2020

Steganography has been successfully employed in various applications, e.g., copyright control of materials, smart identity cards, video error correction during transmission, etc. Deep learning-based steganography models can hide information adaptively through network learning, and they draw much more attention. However, the capacity, security, and robustness of the existing deep learning-based steganography models are still not fully satisfactory. In this paper, three models for different cases, i.e., a basic model, a secure model, a secure and robust model, have been proposed for different cases. In the basic model, the functions of high-capacity secret information hiding and extraction have been realized through an encoding network and a decoding network respectively. The high-capacity steganography is implemented by hiding a secret image into a carrier image having the same resolution with the help of concat operations, InceptionBlock and convolutional layers. Moreover, the secret image is hidden into the channel B of carrier image only to resolve the problem of color distortion. In the secure model, to enhance the security of the basic model, a steganalysis network has been added into the basic model to form an adversarial network. In the secure and robust model, an attack network has been inserted into the secure model to improve its robustness further. The experimental results have demonstrated that the proposed secure model and the secure and robust model have an overall better performance than some existing high-capacity deep learning-based steganography models. The secure model performs best in invisibility and security. The secure and robust model is the most robust against some attacks.

• 한국간호교육학회지
• /
• 제21권1호
• /
• pp.5-15
• /
• 2015

Purpose: This study, on the basis of a structural model that includes nurse image, image determinants, and self-esteem of Cambodian nursing students influenced by the Korean nursing education system, demonstrates distinctions between senior and junior according to experience in clinical practice. Methods: Data were collected via a questionnaire from 194 nursing students in Cambodia and subsequently analyzed using the STATA IC 12 program. Results: First, image determinants and their sub-factors-subjective, institutional, and media-effect both the nurse image and self-esteem of nursing students. Second, the study confirms that nurse image has no significant effect on self-esteem. Also, the individual factor, a sub-factor of nurse image, possesses a weak relationship with nurse image. Third, the structural model mediating between senior and junior reveals differences resulting from experiences derived from clinical practice. Conclusion: The study has significance in that Cambodian nursing students, who are rarely studied in terms of nurse image, have been systemically analyzed via a structural model incorporating image determinants and self-esteem.

• 한국측량학회지
• /
• 제26권6호
• /
• pp.561-569
• /
• 2008

현재, 3D 영상모형을 구축하기 위하여 위성영상, 항공영상 및 항공레이저 측량데이터가 주로 이용되고 있으나, 상기 데이터만을 이용하여 구축된 모형은 지형과 지물을 세밀하고 사실감 있게 표현하는데 제한적이기 때문에 고품질의 영상모형 구축이 필요한 실정이다. 3D 영상모형 구축시 3D 모형을 제작하고, 지상사진으로부터 텍스쳐를 제작하여 3D 모형에 사실감을 부여한다. 본 연구에서는 지형과 인공 시설물을 대상으로 사진측량 및 레이저측량 데이터를 이용하여 공간정확도, 세밀묘사 및 현실감이 강조된 실세계에 근접한 3D 영상모형을 구축하기 위한 기법을 분석하였다. 연구결과 위치정확도와 함께 사진과 같은 사실감을 확보한 3D 영상모형을 구축하였고 인터넷 3D 영상지도 서비스를 통하여 제공 할 수 있도록 하였다. 3D 영상모형을 이용하여 일조권분석, 경관분석, 시설물관리 등 다양한 분야에 활용할 수 있었다.

• 한국의상디자인학회지
• /
• 제16권2호
• /
• pp.31-40
• /
• 2014

The purpose of the study was to examine model postures at fashion shows with respect to expressing fashion images including elegance and neutral-gender images. Data were gathered from the fashion shows held 2000 S/S through 2009 F/W, when elegance and neutral-gender fashion images were obvious in fashion collections. Three designer brands representing elegance and neutral-gender fashion images were selected by the researcher and fashion specialists including graduate students majoring in fashion. The fashion collection photos representing each image were selected from style.com, a website which contains four world's biggest fashion collections. The results showed different hand positions as a model posture according to fashion images. In the neutral-gender image, 16 photos (47%) showed a hand position at pockets, in the elegance image, 24 photos (82.3%) showed a hand position laying down by the sides. Also, walking pose was shown to be different between two fashion images. In the neutral-gender fashion image, 16 photos (52.9%) revealed a pose of 'natural walk', while 29 photos (100%) showed a pose of 'walk in a straight line' in the elegance imaged fashion. In conclusion, the neutral-gender image photos showed the pocket-positioned hand and the 'natural walk' poses more than elegance image photos, and elegance image photos revealed the hand position laying down by the sides and the 'walk in a straight line' poses than the photos of the neutral-gender image.

• 복식문화연구
• /
• 제6권2호
• /
• pp.203-216
• /
• 1998

In the study, conceiving that signifying processes like model, setting, advertisement and description are important to promote the purchase of clothes that would satisfy consumer's desire by their expressing mode, I considered the processes of components of which fashion ads consist. As for the methods to study, I regarded the results presented from prior researches of clothing & textiles and other disciplines for the components of fashion ads and objectified their image that may be interpreted subjectively: and then, I adopted to analyse them using advertisement-semiological method to make clear the signifying processes. The results are as follow: 1. Fashion ad, one of visual symbols to transfer brand image, conveys the image with which various components are combined like model, clothes, setting and description as signs. ① the image of clothes amy be differently expressed according to social, cultural norm and individual characteristics, in the case of clothes, therefore, the signified can be regarded as the transferred image by design of the clothes① sign, and the abstract conception which may be rise to mind by the image in a ceratin culture. ② Each signifier such as countenance, line of vision, attitude and hairstyle of a model conveys different image, or the signified, respectively, and it amy operate as a sign that can express the brand image symbolically. ③ The signifiers like background, color and property symbolize the advertised merchandise of clothes and define it attribute.. 2. In the case of fashion ads, key referent systems are fashion phenomena, contemporary role image, social psychology, common morality, and social, economical and milieu.