• ETRI Journal
• /
• 제37권6호
• /
• pp.1199-1210
• /
• 2015

Automatic emotion recognition based on facial cues, such as facial action units (AUs), has received huge attention in the last decade due to its wide variety of applications. Current computer-based automated two-phase facial emotion recognition procedures first detect AUs from input images and then infer target emotions from the detected AUs. However, more robust AU detection and AU-to-emotion mapping methods are required to deal with the error accumulation problem inherent in the multiphase scheme. Motivated by our key observation that a single AU detector does not perform equally well for all AUs, we propose a novel two-phase facial emotion recognition framework, where the presence of AUs is detected by group decisions of multiple AU detectors and a target emotion is inferred from the combined AU detection decisions. Our emotion recognition framework consists of three major components - multiple AU detection, AU detection fusion, and AU-to-emotion mapping. The experimental results on two real-world face databases demonstrate an improved performance over the previous two-phase method using a single AU detector in terms of both AU detection accuracy and correct emotion recognition rate.

• 한국의류산업학회지
• /
• 제13권3호
• /
• pp.398-408
• /
• 2011

This study shows that the self-satisfaction individually is rising and social life is attracted effective and successful in image making field by structuring the facial image improvement program through experimental study in order to improve unbalance of women's face. Experiment is conducted by electing 3 samples for 12 weeks and analyzing the measurement and visual analysis, infrared thermography, and evaluation of experts in order to check the facial unbalance. Subject 1 had the effect at approximately in 4 weeks with the severely distorted chin line and mouth appendage. The facial outline became softer to turn entire image to be softer and more feminine. Subject 2 had the severe distortion of location and size of eyes and nose. But the skin was getting better at first, followed by eyes getting clearer with the location changed in left and right. Subject 3 had the twisted nose and lower chin, but after two weeks, the eye area and skin were better and the width of left and right chin was similarly changed. On the basis of the above research result, the program to effectively improve the image was structured and presented with the resolution of facial unbalance. Program is consist of the training of breathing method, face washing method, facial muscle exercise.

• 전기전자학회논문지
• /
• 제25권4호
• /
• pp.613-618
• /
• 2021

표정 인식은 다양한 분야에서 지속적인 연구의 주제로서 자리 잡아 왔다. 본 논문에서는 얼굴 이미지 랜드마크 간의 거리를 계산하여 추출된 특징을 사용해 각 랜드마크들의 관계를 분석하고 5가지의 표정을 분류한다. 다수의 관측자들에 의해 수행된 라벨링 작업을 기반으로 데이터와 라벨 신뢰도를 높였다. 또한 원본 데이터에서 얼굴을 인식하고 랜드마크 좌표를 추출해 특징으로 사용하였으며 유전 알고리즘을 이용해 상대적으로 분류에 더 도움이 되는 특징을 선택하였다. 본 논문에서 제안한 방법을 이용하여 표정 인식 분류를 수행하였으며 제안된 방법을 이용하였을 때가 CNN을 이용하여 분류를 수행하였을 때 보다 성능이 향상됨을 볼 수 있었다.

• Maxillofacial Plastic and Reconstructive Surgery
• /
• 제23권4호
• /
• pp.366-375
• /
• 2001

Bone shaving for surgical correction is general method in facial asymmetrical patient with fibrous dysplasia. Therefore, decision of bone shaving amount on the preoperative planning is very difficult for improvement of ideal occlusal relationship and harmonious face. Preoperative planning of facial asymmetry with fibrous dysplasia is generally confirmed by the simulation surgery based on evaluation of clinical examination, radiographic analysis and analysis of facial study model. However, the accurate postoperative results can not be predicted by this method. By using the computed tomography based RP(rapid prototyping) model, simulation of facial skeleton can be duplicated and 3-dimensional simmulation surgery can be perfomed. After fabrication of postoperative study model by preoperactive bone shaving, preoperative and postoperactive surgical index was made by omnivaccum and clear acrylic resin. Amount of bone shaving is confirmed by superimposition of surgical index at the operation. We performed the surgical correction of facial asymmetry patients with fibrous dysplasia using surgical index and prototyping model and obtained the favorable results.

• International Journal of Fuzzy Logic and Intelligent Systems
• /
• 제8권2호
• /
• pp.105-110
• /
• 2008

Human being recognizes emotion fusing information of the other speech signal, expression, gesture and bio-signal. Computer needs technologies that being recognized as human do using combined information. In this paper, we recognized five emotions (normal, happiness, anger, surprise, sadness) through speech signal and facial image, and we propose to method that fusing into emotion for emotion recognition result is applying to multimodal method. Speech signal and facial image does emotion recognition using Principal Component Analysis (PCA) method. And multimodal is fusing into emotion result applying fuzzy membership function. With our experiments, our average emotion recognition rate was 63% by using speech signals, and was 53.4% by using facial images. That is, we know that speech signal offers a better emotion recognition rate than the facial image. We proposed decision fusion method using S-type membership function to heighten the emotion recognition rate. Result of emotion recognition through proposed method, average recognized rate is 70.4%. We could know that decision fusion method offers a better emotion recognition rate than the facial image or speech signal.

• 융합정보논문지
• /
• 제11권7호
• /
• pp.1-6
• /
• 2021

본 연구의 목적은 이미지 랜드마크 선정 기법을 기반으로, 인공신경망 안면 영상분석 시스템의 성능을 향상하기 위한 내용이다. 랜드마크 선정을 위하여 안면 이미지 연령을 분류를 위한 CNN 기반의 다층 ResNet 모델의 구성이 필요하며, ResNet 모델에서 입력 노드의 변화에 따른 출력 노드의 변화를 감지하는 히트 맵을 추출한다. 추출된 다수의 히트 맵을 결합하여 연령 구분 예측과 관계된 안면 랜드마크를 구성한다. 이를 통하여, 안면 랜드마크를 통하여 픽셀의 위치별 중요도를 분석할 수 있으며, 가중치가 낮은 픽셀의 제거함으로서 상당량의 입력 데이터 감소가 가능해졌다. 이러한 기법은 인공신경망 시스템의 연산 성능 향상에 기여하게 된다.

• Journal of Acupuncture Research
• /
• 제33권1호
• /
• pp.57-68
• /
• 2016

Objectives : The purpose of this study was to investigate the correlation between patients' characteristics and the beginning period of improvement, as well as contribute to the efficient management of Bell's palsy patients. Methods : The subjects were 94 patients with Bell's palsy. This study was carried out through the use of an administrative database that included patients' characteristics and clinical information. The analysis of the beginning period of improvement by gender, hypertension, diabetes, drinking history, smoking history and facial palsy history was conducted by independent sample t-test. The analysis of the beginning period of improvement by age, House-Brackmann grade, Yanagihara scale and period receiving Korean medical treatment was conducted by Pearson's correlation analysis. Further analysis of the beginning period of improvement by associated symptoms and seasons was conducted by one-way analysis of variance. Results : 1. Significant correlations were not found between the beginning period of improvement and gender, age, season, smoking history, drinking history, facial palsy history, House-Brackmann grade, Yanagihara scale, hypertension, diabetes or associated symptoms. 2. There was significant correlation between the period of receiving Korean medical treatment and the beginning period of improvement. Conclusion : In this study, the earlier that patients received korean medicine treatment after onset, the earlier that the beginning period of improvement could be seen. Therefore, for the efficient management of facial paralysis patients, it is expected to help secure a baseline.

• 한국멀티미디어학회논문지
• /
• 제15권1호
• /
• pp.40-50
• /
• 2012

Face recognition is a biometric technology used to identify individuals based on facial feature information. Previous studies of face recognition used features including the eye, mouth and nose; however, there have been few studies on the effects of using other facial components, such as the eyebrows and chin, on recognition performance. We measured the recognition accuracy affected by these facial components, and compared the differences between computer-based and human-based facial recognition methods. This research is novel in the following four ways compared to previous works. First, we measured the effect of components such as the eyebrows and chin. And the accuracy of computer-based face recognition was compared to human-based face recognition according to facial components. Second, for computer-based recognition, facial components were automatically detected using the Adaboost algorithm and active appearance model (AAM), and user authentication was achieved with the face recognition algorithm based on principal component analysis (PCA). Third, we experimentally proved that the number of facial features (when including eyebrows, eye, nose, mouth, and chin) had a greater impact on the accuracy of human-based face recognition, but consistent inclusion of some feature such as chin area had more influence on the accuracy of computer-based face recognition because a computer uses the pixel values of facial images in classifying faces. Fourth, we experimentally proved that the eyebrow feature enhanced the accuracy of computer-based face recognition. However, the problem of occlusion by hair should be solved in order to use the eyebrow feature for face recognition.

• 대한두개안면성형외과학회지
• /
• 제16권3호
• /
• pp.136-142
• /
• 2015

Background: The incidence and etiology of facial bone fracture differ widely according to time and geographic setting. Because of this, prevention and management of facial bone fracture requires ongoing research. This study examines the relationship between socioeconomic status and the incidence of facial bone fractures in patients who had been admitted for facial bone fractures. Methods: A retrospective study was performed for all patients admitted for facial bone fracture at the National Medical Center (Seoul, Korea) from 2010 to 2014. We sought correlations amongst age, gender, fracture type, injury mechanism, alcohol consumption, and type of medical insurance. Results: Out of the 303 patients meeting inclusion criteria, 214 (70.6%) patients were enrolled in National Health Insurance (NHI), 46 (15.2%) patients had Medical Aid, and 43 (14.2%) patients were homeless. The main causes of facial bone fractures were accidental trauma (51.4%), physical altercation (23.1%), and traffic accident (14.2%). On Pearson's chi-square test, alcohol consumption was correlated significantly with accidental trauma (p<0.05). And, the ratio of alcohol consumption leading to facial bone fractures differed significantly in the homeless group compared to the NHI group and the Medical Aid group (p<0.05). Conclusion: We found a significant inverse correlation between economic status and the incidence of facial bone fractures caused by alcohol consumption. Our findings indicate that more elaborate guidelines and prevention programs are needed for socioeconomically marginalized populations.

• 한국멀티미디어학회논문지
• /
• 제7권10호
• /
• pp.1478-1484
• /
• 2004

본 논문은 사용자로 하여금 얼굴표정들의 공간으로부터 일련의 표정을 실시간적으로 선택하게 함으로써 3차원 아바타의 실시간적 얼굴 표정을 제어하는 기법을 기술한다. 본 시스템에서는 약 2400여개의 얼굴 표정 프레임을 이용하여 표정공간을 구성하였다. 본 기법에서는 한 표정을 표시하는 상태표현으로 얼굴특징점들 간의 상호거리를 표시하는 거리행렬을 사용한다. 이 거리행렬의 집합을 표정공간으로 한다. 3차원 아바타의 얼굴 표정은 사용자가 표정공간을 항해하면서 실시간적으로 제어한다. 이를 도와주기 위해 표정공간을 PCA투영 기법을 이용하여 2차원 공간으로 가시화했다. 본 시스템이 어떤 효과가 있는지를 알기 위해 사용자들로 하여금 본 시스템을 사용하여 3차원 아바타의 얼굴 표정을 제어하게 했는데, 본 논문은 그 결과를 평가한다.