• 사물인터넷융복합논문지
• /
• 제9권6호
• /
• pp.125-134
• /
• 2023

본 논문은 MediaPipe Face Mesh 모델을 이용해 일련의 프레임 시퀀스에서 얼굴 제스처를 인식해 해당 사용자 이벤트를 처리하는 얼굴 제스처 기반의 사용자 인터페이스 선행 연구의 성능 개선 방안을 제안함에 그 목적이 있다. 선행 연구는 MediaPipe Face Mesh 모델에서 선택한 7개의 랜드마크의 3차원 좌표들로부터 얼굴 제스처를 인식해 해당 사용자 이벤트를 발생시키고 이에 대응하는 명령을 수행하는 것이 특징이다. 제안된 방법은 그 과정에서 커서 위치들에 적응형 이동 평균 처리를 적용해 미세 떨림을 완화함으로써 커서 안정화를 도모하고, 양안 동시 개폐 시에 양안의 일시적 개폐 불일치를 차단해 그 성능을 개선하였다. 제안된 얼굴 제스처 인터페이스의 사용성 평가 결과, 얼굴 제스처의 평균 인식률이 선행 연구에서 95.8%였던 것에 비해 98.7%로 상향되는 것이 확인되었다.

• Maxillofacial Plastic and Reconstructive Surgery
• /
• 제35권6호
• /
• pp.360-367
• /
• 2013

Purpose: The purpose of this study was to perform three-dimensional (3D) assessment of facial soft tissue in patients with skeletal Class III and mandibular asymmetry after orthognathic surgery. Methods: Samples consisted of 3D facial images obtained from five patients with A point-nasion-B point angle less than 2 degrees, and more than 5 mm of menton deviation. All patients had been treated at Gangneung-Wonju National University Dental Hospital from 2009 to 2012. They had undergone orthognathic surgery of Lefort I, and sagittal split osteotomy for correction of skeletal deformity, and orthodontic treatment. Facial scanning was performed before treatment (T1) and post-surgical orthodontic treatment (T2). Linear and angle variables of soft tissue landmarks, antero-posterior facial depth, and facial volume were measured. Results: No significant differences in width of the alar base, mouth width, and nasal canting were observed between T1 and T2. However, lip deviation, menton deviation, alar canting, lip canting, and menton deviation angle were significantly reduced at T2. Antero-posterior facial depth on the axial plane parallel to the left cheilion was significantly reduced on the deviated side and significantly increased on the non-deviated side at T2. Volume of the lower lateral and lower medial parts of the face was reduced on the deviated side, and volume of upper lateral and lower lateral parts on the non-deviated side was significantly increased at T2. Conclusion: After orthognathic surgery, facial asymmetry of soft tissue was improved following skeletal changes, especially the mandibular region. Although the length of the alar base and mouth width did not change, lip and soft tissue menton were displaced to the medial side after treatment. Facial depth also became symmetric after treatment. Facial volume showed a decrease on the lower part of the deviated side and that on lateral parts of the non-deviated side showed an increase after treatment.

• 대한두경부종양학회지
• /
• 제22권1호
• /
• pp.29-32
• /
• 2006

Background: The location of parotid gland tumors can influence the duration and the difficulty of the operation. If the information about tumor location was available preoperatively, it would allow accurate operative planning and counseling of patients in terms of the length of the operation and the potential morbidity. Methods: This study was based on a retrospective review of 100 patients with parotid gland tumors underwent parotidectomy from January 2000 to October 2005 at Yong-Dong Severance Hospital. Based on computed tomographic(CT) scan findings, 4 landmarks such as facial nerve(FN) line, Utrecht(U) line, Conn's are(CA), and retromandibular vein (RV) were drawn on the scans in same plane. The location of tumors were determined by the landmarks and confirmed by the operative findings. The accuracy of each landmarks was evaluated. To find out the accuracies according to tumor size, the tumors were divided into 2 groups; less than 2 cm and larger than 2 cm in diameter. Results: U line was the most accurate(94%), sensitive(89.3%) and specific(97.7%) in predicting tumor location of the parotid gland. However, in small tumors less than 2cm, FN line (p=0.022) and RV criteria (p=0.028) were more reliable in accuracy. Conclusion: CA, FN line, U line, and RV are all useful landmarks in preoperative prediction for the location of parotid gland tumors. However, U line was the most accurate, but we must consider that proper landmark should be used in prediction according to the size of tumor because the accuracy of landmark may change.

• Journal of the Korean Association of Oral and Maxillofacial Surgeons
• /
• 제34권6호
• /
• pp.622-627
• /
• 2008

In orthognathic surgery, precise analysis and diagnosis are essential for successful results. In facial asymmetric patient, traditional 2D image analysis has been used by lateral and P-A Cephalometric view, Skull PA, Panorama, Submentovertex view etc. But clinicians sometimes misdiagnose because they cannot find exact landmark due to superimposition, moreover image can be magnified and distorted by projection technique or patient's skull position, when using these analysis and method. For overcome these defects, analysis by using of 3D CT has been introduced. In this way we can analysis precisely by getting the exact image free of artifact and finding exact landmark with no interruption of superimposition. So we want to review of relationship between various skeletal landmarks of mandible or cranial base and facial asymmetry by predictable analysis using 3D CT. We select the cases of the patients who visited our department for correction of facial asymmetry during 2003-2007 and who were taken image of 3D CT for diagnosis. 3D CT images were reconstructed to 3D image by using V-Work program (Cybermed Inc., Seoul, Korea). And we analysis the relationship between facial asymmetry and various affecting factor of skeletal pattern. The mandibular ramus hight difference between right and left was most affecting factor that express facial asymmetry. And in this research, there was no relationship between cranial base and facial asymmetry. The angulation between facial midline and mandibular ramus divergency has significant relationship with facial asymmetry

• Journal of Korean Neurosurgical Society
• /
• 제51권5호
• /
• pp.253-261
• /
• 2012

Objective : This study was performed to determine the anatomical landmarks and optimal dissection points of the facial nerve (FN) and the hypoglossal nerve (HGN) in the submandibular region to provide guidance for hypoglossal-facial nerve anastomosis (HFNA). Methods : Twenty-nine specimens were obtained from 15 formalin-fixed adult cadavers. Distances were measured based on the mastoid process tip (MPT), common carotid artery bifurcation (CCAB), and the digastric muscle posterior belly (DMPB). Results : The shortest distance from the MPT to the stylomastoid foramen was $14.1{\pm}2.9$ mm. The distance from the MPT to the FN origin was $8.6{\pm}2.8$ mm anteriorly and $5.9{\pm}2.8$ mm superiorly. The distance from the CCAB to the crossing point of the HGN and the internal carotid artery was $18.5{\pm}6.7$ mm, and that to the crossing point of the HGN and the external carotid artery was $15.1{\pm}5.7$ mm. The distance from the CCAB to the HGN bifurcation was $26.6{\pm}7.5$ mm. The distance from the digastric groove to the HGN, which was found under the DMPB, was about $35.8{\pm}5.7$ mm. The distance from the digastric groove to the HGN, which was found under the DMPB, corresponded to about 65.5% of the whole length of the DMPB. Conclusion : This study provides useful information regarding the morphometric anatomy of the submandibular region, and the presented morphological data on the nerves and surrounding structures will aid in understanding the anatomical structures more accurately to prevent complications of HFNA.

• 대한치과교정학회지
• /
• 제19권1호
• /
• pp.107-122
• /
• 1989

McNamara씨 분석법에 의한 악안면골의 성장변화 평가 및 한국인 기준치를 얻고자 정상적인 안모 및 교합을 가졌다고 생각되는 $9\~21$세의 388명(9세, 11세, 13세는 2년 간격으로 누년적으로 얻은 것임)을 대상으로 두부방사선규격 사진을 분석하여 다음과 같은 결론을 얻었다. 남녀별 각 계측항목의 연령에 따른 평균치 및 표준편차를 산출하였다. $9\~13$세 사이에서 effective maxillary length, effective mandibular length, lower anterior facial height는 연평균 각각 1.90, 2.89, 0.99mm 증가하였다. 주어진 effective maxillary length에 대한 적절한 effective mandibular length, lower anterior facial height를 추정 할 수 있는 회귀방정식을 얻었다.

• 대한치과교정학회지
• /
• 제45권3호
• /
• pp.105-112
• /
• 2015

Objective: A recently developed facial scanning method uses three-dimensional (3D) surface imaging with a light-emitting diode. Such scanning enables surface data to be captured in high-resolution color and at relatively fast speeds. The purpose of this study was to evaluate the accuracy and precision of 3D images obtained using the Morpheus 3D$^{(R)}$ scanner (Morpheus Co., Seoul, Korea). Methods: The sample comprised 30 subjects aged 24.34 years (mean $29.0{\pm}2.5$ years). To test the correlation between direct and 3D image measurements, 21 landmarks were labeled on the face of each subject. Sixteen direct measurements were obtained twice using digital calipers; the same measurements were then made on two sets of 3D facial images. The mean values of measurements obtained from both methods were compared. To investigate the precision, a comparison was made between two sets of measurements taken with each method. Results: When comparing the variables from both methods, five of the 16 possible anthropometric variables were found to be significantly different. However, in 12 of the 16 cases, the mean difference was under 1 mm. The average value of the differences for all variables was 0.75 mm. Precision was high in both methods, with error magnitudes under 0.5 mm. Conclusions: 3D scanning images have high levels of precision and fairly good congruence with traditional anthropometry methods, with mean differences of less than 1 mm. 3D surface imaging using the Morpheus 3D$^{(R)}$ scanner is therefore a clinically acceptable method of recording facial integumental data.

• 정보처리학회논문지:소프트웨어 및 데이터공학
• /
• 제10권3호
• /
• pp.99-108
• /
• 2021

차량 내부에는 조명 변화, 부분적인 가림 및 운전자의 상태 변화와 같은 제한되지 않은 조건들이 존재한다. 본 논문에서는 비 제약적인 차량 환경에서의 운전자 얼굴 인증 시스템을 제안한다. 제안한 방법은 차량 내부 및 외부의 조명 변화에 따라 발생하는 얼굴 이미지의 변화를 최소화하기 위해서 근적외선(NIR) 카메라를 사용한다. 특히 정면에서의 강한 빛에 노출된 얼굴 이미지를 처리하기 위해서, 학습 이미지의 평균과 분산을 사용하여 정상적인 얼굴 이미지로부터 빛에 과다하게 노출된 이미지로 변환하여 사용한다. 따라서 정상적인 조명에서의 얼굴 분류기와 강한 정면광에서의 얼굴 분류기를 각각 동시에 만들어진다. 제안하는 얼굴 분류기는 얼굴 랜드마크를 추출하고 각 랜드마크의 신뢰도 점수를 합산하여 얼굴을 최종적으로 식별한다. 특히 각 랜드마크를 인식하여 부분적인 얼굴 가림에 강하기 때문에 안경이나 선글라스를 착용하는 상황에서도 높은 성능 향상이 가능하다. 즉 가려지지 않은 남은 랜드마크의 점수를 사용하여 운전자를 인식할 수 있다. 또한 등록 운전자와 미등록 운전자 간의 관계를 고려한 새로운 인식 거부 방법과 새로운 평가 방법을 논문에서 제안한다. 자체 취득한 데이터 셋, 공인된 PolyU 및 ORL 데이터 셋으로 실험한 결과 제안한 방법이 효과적임을 확인할 수 있었다.

• Archives of Plastic Surgery
• /
• 제42권2호
• /
• pp.226-231
• /
• 2015

The aim of this study is to consider breast imagery in art as depicted through western painting. Twenty western art paintings were collated. Most of the sample paintings were created from the mid-nineteenth century to the late twentieth century and some are from the Renaissance period. Ten anthropometric items were used to measure 15 distances between two landmarks and 3 angles between three points. The distance from the nipple to the sternal notch and to the midclavicular point was the same and they were 0.46 of the distance from the sternal notch to the umbilicus. The shape of the projection of the breast was almost an isosceles triangle and the altitude of the triangle was at a proportion of 0.45 of the bottom length and 0.16 of the distance from the sternal notch to the umbilicus. The distance between the lateral ends of the breasts was 2.14 times the facial width and the distance between nipples was 1.36 times the facial width. Proportions from works of art are more ideal and attractive than clinically measured proportions. The desirable ratios measured from historical paintings might be useful in planning breast surgeries.

• 한국정보통신학회:학술대회논문집
• /
• 한국해양정보통신학회 2010년도 추계학술대회
• /
• pp.196-199
• /
• 2010

사람의 눈동자는 얼굴 크기와 비교해 볼 때 상대적으로 일정한 거리를 가지고 있기 때문에 이미지 정규화에 있어서 중요한 지표로 사용된다. 이 논문은 이러한 특징을 이용해 최적화된 세그멘테이션 방법을 사용하여 눈동자 검출의 새로운 접근방법을 소개한다. 눈 검출 방법은 세 가지 중요한 단계로 나눌 수 있다. (1)흑백 영상에서 눈 영역에 적합한 에지 추출 방법, (2)레이블링(labeling) 기법을 이용한 눈 영역 추출, (3)밝기값 정보를 이용한 눈동자 위치 검출. 실험 결과로는 다양한 조명 환경과 얼굴표정을 가진 2408장의 FERET 영상을 이용하여 98.9%의 검출 성능을 보였다.