• 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.

• 사상체질의학회지
• /
• 제24권4호
• /
• pp.17-27
• /
• 2012

Objectives : This study proposes developing Sasang Medical Diagnosis Program using Facial form for increase in Sasang Constitution Diagnosis objectivity and putting the Diagnosis Program into practical use. The author presents a review of extant research on Sasang constitution diagnosis utilizing facial feature analysis and suggests an agenda for further research. Methods : For this thesis, a collection of dissertations on the subject of 'Usage of facial form for constitution diagnosis' published until September of 2012 such as RISS4U, OASIS, KISTI, Korean TK were reviewed. The final 33 dissertations were classified into two categories, basic or clinical research and then analyzed. Results : 9 out of 33 dissertations were of basic research and 24 were of clinical research. 1) As result of review of references, a uniform tendency was found in facial form according to Sasang Constitution. 2) In the grade of practical use, facial element is repeatedly used and the facial element of important use has constitutional differences. 3) Standard faces per Sasang Constitution were derived as result of 2-dimensional research. 4) 3-dimensional research focused on improvement of accuracy and reliability of 3D-AFRA, and there has been an attempt to develop a prototype for identification. Conclusions : For practical use of facial feature in Sasang Constitution Diagnosis, 1) Standardization of diagnosis through establishing Sasang Medical Diagnosis clinical protocol must be preceded. After the standardization, practical purpose and direction of facial form in general may be decided. 2) Information on high quality facial form of constitutional and conditional patients must be collected to form extensive database. 3) Subdivided symptomatology, as well as Sasang Constitution must be considered for diagnosis in order for diagnosis technique to acquire clinical practicality.

• 대한치과교정학회지
• /
• 제37권3호
• /
• pp.192-203
• /
• 2007

본 연구는 비폐쇄를 보이는 III급 부정교합아동에서, 기도 공간의 형태와 안모 골격 형태 사이의 상관관계를 평가하였다. III급 부정교합을 보이며, 비폐쇄 소견을 보여 이비인후과로 의뢰된 환아 100명의 초진 측모 두부 규격 방사선 사진에서 상 인두기도 공간(upper PAS), 하 인두기도 공간(lower PAS), 편도의 크기, 구개-혀 공간을 측정하여 기도 공간 형태를 분석하였으며, 통상적인 계측점을 사용하여 안면 골격 분석을 시행하여 기도 공간 형태 계측항목과 안면 골격분성 항목간의 상관관계를 연구하여 다음과 같은 결과를 얻었다. Upper PAS는 ramal height, SNA, SNB, PFH, FHR, facial plane angle 항목과 양의 상관관계를 보였으며, saddle angle, articular angle, gonial angle의 sum, SN-GoGn, Y-axis to SN, FMA 항목과 음의 상관관계를 보였다. Lower PAS는 genial angle, FMA와 양의 상관관계를 보였으며, articular angle, facial depth, PFH, FHR와 음의 상관관계를 보였다. 편도의 크기는 PCBL, ramal height, Mn. body length, Mn. body length to ACBL, facial depth, facial length, PFH, AFH와 양의 상관관계를 보였다. 구개-혀 공간은 saddle angle, articular angle, genial angle의 합, facial length, AFH, FMA, LFH와 양의 상관관계를 보였으며, IMPA, overbite와 음의 상관관계를 보였다.

• Maxillofacial Plastic and Reconstructive Surgery
• /
• 제33권6호
• /
• pp.490-496
• /
• 2011

Purpose: Dr. Marquardt made the facial 'phi' mask using golden ratio. Most class III patients have bulky faces and want a smaller face. Using a facial golden mask, this study estimated and compared frontal photographs before and after operation for soft tissue measurement. The golden mask can be considered as a reference tool for facial esthetic analyses especially in lower face. Methods: Forty patients who had undergone orthognathic surgery at Samsung Medical Center from January 2006 to December 2009 were included in this study. These patients had Class III malocclusion. Lateral cephalometric radiographs, frontal clinical photos of pre-op and 8~12 month later post-op, and the facial 'phi' mask using golden ratio, were used for analysis. Reduction of the lower face area, occlusal plane changes, amounts of mandible setback and amounts of maxilla posterior impaction were estimated. Results: Lower facial reduction ratio and mandibular setback amounts were significantly different between 1-jaw and 2-jaw groups. Average postoperative changes in the area of lower face between bilateral sagittal split ramus osteotomy (BSSRO) and BSSRO combined maxilla posterior impaction were compared by using an independent simple t-test and $P$ value was 0.016. Therefore, the lower facial reduction ratio and mandibular setback amount were significantly different in maxilla posterior impaction. Conclusion: The two-jaw surgery group showed more reduction of the lower facial area than the 1-jaw surgery group. The amount of lower facial reduction was more related with the amount of mandibular setback. There was no significant relation in lower facial reduction with amount of maxilla posterior impaction, pre-op occlusal plane, post-op occlusal plane and the mandibular angle. A relationship between the change in the lower facial area and the amount of maxilla posterior impaction or the change of mandibular angle occlusal plane at pre-op could not be found because of the difference in the amount of setback between two groups.

• 한국지능시스템학회논문지
• /
• 제13권3호
• /
• pp.371-376
• /
• 2003

본 논문에서는 효과적인 정보를 표현하는 Independent Component Analysis(ICA)-factorial 표현방법을 이용하여 얼굴감정 인식을 수행한다. 얼굴감정인식은 두 단계인 특징추출 과정과 인식과정에 의해 이루어진다. 먼저 특징추출방법은 주성분 분석(Principal Component Analysis)을 이용하여 얼굴영상의 고차원 공간을 저차원 특징공간으로 변환한 후 ICA-factorial 표현방법을 통해 좀 더 효과적으로 특징벡터를 추출한다. 인식단계는 최소거리 분류방법인 유클리디안 거리에 근거한 K-Nearest Neighbor 알고리즘으로 얼굴감정을 인식한다. 6개의 기본감정(기쁨, 슬픔, 화남, 놀람, 공포, 혐오)에 대해 얼굴 감정 데이터베이스를 구축하고 실험해본 결과 기존의 방법보다 좋은 인식 성능을 얻었다.

• Maxillofacial Plastic and Reconstructive Surgery
• /
• 제30권6호
• /
• pp.546-553
• /
• 2008

Purpose: The purpose of this study was to estimate validity of posterior anterior cephalometric and 3D-CT for orbital canting analysis. Materials and methods: Three trained observers classified two patients group using standardized frontal photographs of facial asymmetry patients. Group A consisted of patients with facial asymmetry and orbital canting(n=19), and group B consisted of patients with only facial asymmetry(n=43). Orbital canting was measured with line of bilateral inferior orbitale. Orbital canting measurement was done with posterior anterior cephalometric and 3D-CT. Each horizontal reference line was established by bilateral GWSO(cephalometric), FZS(3D-CT). Maxillary canting and mandibular deviation angle were also measured and analyzed with orbital canting. Results: The mean orbital canting was $3.03{\pm}1.00^{\circ}$ in Group A and $1.11{\pm}0.76^{\circ}$ in Group B in frontal photograph. The mean orbital canting was $1.20{\pm}0.74^{\circ}$ in group A and $1.22{\pm}0.65^{\circ}$ in group B by cephalometric analysis(p>0.05). In 3D-CT, orbital canting was almost paralleled with horizontal reference line. The orbital canting, maxillay canting and mandibular deviation between two groups showed no significant differences except madibular deviation in 3D-CT. Conclusion: Common analysis of posterior anterior cephalometric and 3D-CT is not valide method to evaluate orbital canting for facial asymmetry patients with orbital canting.

• 복식문화연구
• /
• 제21권3호
• /
• pp.401-412
• /
• 2013

The purpose of this study was to investigate the effects of female applicant's facial attractiveness and feminine-masculine clothing image on job performance evaluation and hiring decision. The research design of study consisted of 3(facial attractiveness high, middle, low)${\times}$2(feminine and masculine clothing image) factorial design. The subject consisted of 243 persons whose occupation were mid-sized companies' administrator in Gwangju and Seoul City. The data were analyzed by factor analysis, Duncan test, ANOVA, t-test. The results of this study were as follows. First, three factors emerged to account for the job performance evaluation. These factors were given the titles of task performance, cooperation and self-management factors. Second, applicant's facial attractiveness exerted significant positive effect on self-management and significant negative effect on cooperation. Third, applicant's facial attractiveness exerted significant effect on hiring decision. Finally, the interaction effect of female applicant's facial attractiveness and feminine-masculine clothing image on job performance evaluation and hiring decision were not significant.

• 전자공학회논문지B
• /
• 제31B권9호
• /
• pp.101-111
• /
• 1994

The human face can provide a great deal of information in regard to his/her race, age, sex, personality, feeling, psychology, mental state, health condition and ect. If we pay a close attention to the aging process, we are able to find out that there are recognizable phenomena such as eyelid drooping, cheek drooping, forehead furrowing, hair falling-out, the hair becomes gray and etc. This paper proposes that the method to estimate the age by analyzing these feature components for the facial image. Ang we also introduce the method of facial image synthesis in accordance with the cange of age. The feature components according to the change of age can be obtainec by dividing the facial image into the 3-dimensional shape of a face and the texture of a face and then analyzing the principle component respectively using 3-dimensional model. We assume the age of the facial image by comparing the extracted feature component to the facial image and synthesize the resulted image by adding or subtracting the feature component to/from the facial image. As a resurt of this simulation, we have obtained the age changed ficial image of high quality.

• Safety and Health at Work
• /
• 제11권1호
• /
• pp.71-79
• /
• 2020

Background: Protection from yellow dust and particulate matter is ensured by the use of respirators among the Korean citizens and workers. However, the manikins used to test the performance of the same were manufactured considering western facial specifications owing to which they do not represent Korean facial characteristics. Methods: Analysis of the data from the 6th 3D anthropometric survey of Koreans (Size Korea; 2010-2013) of 4,583 people aged 7 to 69 years was performed to obtain their facial dimensions. We subsequently clustered 44 facial measurements using Design X software, followed by the creation of the cluster centroid. Results: Three 3D head forms were developed-small, medium, and large, and their images were stored in ".stl" format for 3D printing. The facial widths and lengths of the three head forms were 127.1 mm × 90.6 mm, 143.2 mm × 104.0 mm, and 149.1 mm × 120.2 mm, respectively. Conclusion: We developed manikin head forms according to the facial dimensions of the Korean population, which was essential in evaluating respiratory protective equipment. These head forms can be used to test the performance of respirators considering the facial dimensions of the Korean population.

• 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.