• 한국산업정보학회논문지
• /
• 제7권3호
• /
• pp.23-31
• /
• 2002

본 논문에서는 화상자료의 특성인 이웃 화소간의 종속성을 표현하는데 적합한 깁스분포를 바탕으로 얼굴 표정을 인식을 위한 특징벡터를 추출하는 새로운 방법을 제안하였다. 추출된 특징벡터는 얼굴 이미지의 크기, 위치, 회전에 대하여 불변한 특성을 갖는다. 얼굴 표정을 인식하기 위한 알고리즘은 특징벡터 추출하는 과정과 패턴을 인식하는 두 과정으로 나뉘어진다. 특징벡터는 얼굴 화상에 대하여 추정된 깁스분포를 바탕으로 수정된 2-D 조건부 모멘트로 구성된다. 얼굴 표정인식 과정에서는 패턴인식에 널리 사용되는 이산형 HMM를 사용한다. 제안된 방법에 대한 성능평가를 위하여 4가지의 얼굴 표정 인식 실험을 Workstation에서 실험한 결과, 제안된 얼굴 표정 인식 방법이 95% 이상의 성능을 보여주었다.

• Journal of Information Processing Systems
• /
• 제9권1호
• /
• pp.173-188
• /
• 2013

In Facial Expression Recognition Systems (FERS), only particular regions of the face are utilized for discrimination. The areas of the eyes, eyebrows, nose, and mouth are the most important features in any FERS. Applying facial features descriptors such as the local binary pattern (LBP) on such areas results in an effective and efficient FERS. In this paper, we propose an automatic facial expression recognition system. Unlike other systems, it detects and extracts the informative and discriminant regions of the face (i.e., eyes, nose, and mouth areas) using Haar-feature based cascade classifiers and these region-based features are stored into separate image files as a preprocessing step. Then, LBP is applied to these image files for facial texture representation and a feature-vector per subject is obtained by concatenating the resulting LBP histograms of the decomposed region-based features. The one-vs.-rest SVM, which is a popular multi-classification method, is employed with the Radial Basis Function (RBF) for facial expression classification. Experimental results show that this approach yields good performance for both frontal and near-frontal facial images in terms of accuracy and time complexity. Cohn-Kanade and JAFFE, which are benchmark facial expression datasets, are used to evaluate this approach.

• Journal of Information Processing Systems
• /
• 제17권2호
• /
• pp.337-351
• /
• 2021

Automatically recognizing facial expressions in video sequences is a challenging task because there is little direct correlation between facial features and subjective emotions in video. To overcome the problem, a video facial expression recognition method using spatiotemporal recurrent neural network and feature fusion is proposed. Firstly, the video is preprocessed. Then, the double-layer cascade structure is used to detect a face in a video image. In addition, two deep convolutional neural networks are used to extract the time-domain and airspace facial features in the video. The spatial convolutional neural network is used to extract the spatial information features from each frame of the static expression images in the video. The temporal convolutional neural network is used to extract the dynamic information features from the optical flow information from multiple frames of expression images in the video. A multiplication fusion is performed with the spatiotemporal features learned by the two deep convolutional neural networks. Finally, the fused features are input to the support vector machine to realize the facial expression classification task. The experimental results on cNTERFACE, RML, and AFEW6.0 datasets show that the recognition rates obtained by the proposed method are as high as 88.67%, 70.32%, and 63.84%, respectively. Comparative experiments show that the proposed method obtains higher recognition accuracy than other recently reported methods.

• 한국실내디자인학회논문집
• /
• 제13권6호
• /
• pp.175-183
• /
• 2004

As the information-oriented society makes progress, the role of Image has much influence on human being in space as a medium for information delivery and a means of artistic communication. These influences are appeared as expressional characteristics of the image such as the reproducibility of reality and unreality in the real world, the synchrony of expression of time, visual formality, a sign and the transmission of moaning. For these, the investigator examined the meaning of image in aspace, taking into consideration of the interrelationship of image, space, and human being. As study findings show, the expressional characteristics of image in space have such visual effects as a space in which pictorial formality and object exists, in which the mutual understanding of communication exists, and that realizes immaterial membrane in the aspect of time and space, according to the electronic light, color, and formation of the image media. In addition, it become clear that the characteristic could be staged on various circumstances by constructing the relationship between an object and a point of time interactively with bidirectional communication through combining technology and art. This suggests that the image develops as the form of sensory communication via interacting of space and human being.

• 한국의류학회지
• /
• 제25권2호
• /
• pp.327-338
• /
• 2001

The purpose of this thesis is to study the aspects of sex identity in the contemporary mens fashion expressed through sexual image and taste in the half of twentieth century. The aspects of sex identity expression in the contemporary mens fashion can be classified by image, that is, homosexual, heroic, bisexual, and fetish. 1. The homosexual image has shown the tendency to emphasize the masculinity since 1950.60s. It can be found in \"Cowboys costume\" which is typical of American traditional fashion, and jeans and underwear fashion expressed by muscular men has homosexual characteristics which contain narcissism. 2. Based on mens traditional gender role, the heroic image emphasizes mens physical characteristics and expresses tough and offensive masculine beauty in mens suit which is free from the authority and formality. 3. The bisexual image denied the division of gender role by costume and destroyed the traditional sex model by resolutely applying womens costume such as skirts to mens fashion. 4. The fetish image is similar to bisexual image in that they wear womens costume, but different in that it expresses sexual desire or fantasy. It is expressed through brilliant color, leather and metal ornaments, and sensual element of women which emphasizes \"body\". This shows the sex identity of contemporary men who want more sensible and free life.sible and free life.

• 제어로봇시스템학회논문지
• /
• 제13권8호
• /
• pp.754-759
• /
• 2007

In this paper, we proposed the Bi-Modal Sensor Fusion Algorithm which is the emotional recognition method that be able to classify 4 emotions (Happy, Sad, Angry, Surprise) by using facial image and speech signal together. We extract the feature vectors from speech signal using acoustic feature without language feature and classify emotional pattern using Neural-Network. We also make the feature selection of mouth, eyes and eyebrows from facial image. and extracted feature vectors that apply to Principal Component Analysis(PCA) remakes low dimension feature vector. So we proposed method to fused into result value of emotion recognition by using facial image and speech.

• 복식문화연구
• /
• 제24권1호
• /
• pp.27-40
• /
• 2016

This study investigated the preference of sensibility expression factors regarding fashion materials, such as the color, pattern and texture of fabric. Moreover, this study analyzed the relationship between the preference of sensibility expression factors and the preference of fashion images by identifying the preference of fashion images. The survey subjects were 312 women ranging in age from 20 to 40 years old. This study utilized a questionnaire as a measurement tool. First, this study performed a factorial analysis on the preference of sensibility expression factors of fashion materials. In regards to color preference, this study considered color depth such as light tone color, moderate tone color, dark tone color and vivid tone color. In regards to pattern preference, this study examined: geometric pattern, floral pattern, animal skins pattern, check pattern and symbolical pattern. In regard to preference of the texture, this study assessed: roughness, luster, flatness and lightness. Second, this study performed a factorial analysis on the preference of fashion images. This study examined five factors: dignity, uniqueness, femininity, activity and simplicity. Third, this study analyzed the effects of the preference of sensibility expression factors of fashion materials on the preference of fashion images. As a result, the color preference was related to the image preference associated with dignity, femininity and simplicity, whereas the pattern preference was related to the images of uniqueness, femininity, activity and simplicity. Moreover, the preference of texture image was related to the images of dignity, uniqueness, femininity and activity.

• 한국지능시스템학회논문지
• /
• 제20권6호
• /
• pp.806-813
• /
• 2010

표정은 사람의 감정을 표현하는 대표적인 수단이다. 이러한 이유로 표정은 사람의 의도를 컴퓨터에 전하는데 효과적인 방법으로 사용될 수 있다. 본 논문에서는 2D 영상에서 사람의 표정을 보다 빠르고 정확하게 인식하기 위해 Discrete Adaboost 알고리즘과 신경망 알고리즘을 통합하는 방법을 제안한다. 1차로 Adaboost 알고리즘으로 영상에서 얼굴의 위치와 크기를 찾고, 2차로 표정별로 학습된 Adaboost 강분류기를 이용하여 표정별 출력 값을 얻으며, 이를 마지막으로 Adaboost 강분류기 값으로 학습된 신경망 알고리즘의 입력으로 이용하여 최종 표정을 인식한다. 제안하는 방법은 실시간이 보장된 Adaboost 알고리즘의 특성과 정확성을 개선하는 신경망 기반 인식기의 신뢰성을 적절히 활용함으로서 전체 인식기의 실시간성을 확보하면서도 정확성을 향상시킨다. 본 논문에서 구현된 알고리즘은 평온, 행복, 슬픔, 화남, 놀람의 5가지 표정에 대해 평균 86~95%의 정확도로 실시간 인식이 가능하다.

• 한국디자인학회:학술대회논문집
• /
• 한국디자인학회 2004년도 춘계 학술발표대회 논문집
• /
• pp.178-179
• /
• 2004

• 한국멀티미디어학회:학술대회논문집
• /
• 한국멀티미디어학회 2000년도 추계학술발표논문집
• /
• pp.163-166
• /
• 2000

This paper presents an image processing algorithm which gives an sketch effect on the real image taken by CCD camera. The essence of the technique is the expression of shading by the line touch. Several line patterns are developed to emulate pencil drawings, fusains, Indian ink paintings. This algorithm is expected to be used for computer portraits, character development, and souvenirs.