• 정보과학회 논문지
• /
• 제44권3호
• /
• pp.295-305
• /
• 2017

얼굴 외형은 조명의 영향을 크게 받기 때문에 조명 변화는 얼굴 인식 시스템의 성능을 저하시키는 요인 중 하나이다. 본 논문에서는 non-alpha Weberface(non-alpha WF)와 히스토그램 평활화를 결합하여 조명 변화에 강건한 얼굴 표현 방법을 제안한다. 먼저, 입력 얼굴 영상에 대해 명암 대비 조절 파라미터를 적용하지 않은 non-alpha WF를 생성한다. 이후, non-alpha WF의 히스토그램 분포를 전역적으로 균일하게 하고 명암 대비를 향상시키기 위해 히스토그램 평활화를 수행한다. 제안하는 방법을 통해 전처리된 얼굴 영상으로부터 저차원 판별 특징을 추출하기 위해 $(2D)^2PCA$를 적용한다. Extended Yale B 및 CMU PIE 얼굴 데이터베이스에 대해 실험한 결과, 제안하는 방법으로 각각 93.31%와 97.25%의 평균 인식률을 얻었다. 또한, 제안하는 방법은 기존 WF뿐만 아니라 여러 조명 처리 방법들과 비교하여 향상된 인식 성능을 보였다.

• 한국정밀공학회지
• /
• 제16권8호
• /
• pp.178-185
• /
• 1999

In tool condition monitoring systems, parameters should be set to a certain threshold. In many cases, however, the threshold is dependent on cutting conditions, especially the radial immersion ratio. In this presented is a method of on-line estimation of the radial immersion ratio in face milling. When a tooth finishes sweeping, a sudden drop of cutting force occurs. The force drop is equal to the cutting force that acting on a tooth at the swept angle of cut and can be acquired from cutting force signals in feed and cross-feed directions. Average cutting force per tooth period can also be calculated from cutting force signals in two directions. The ratio to cutting forces in two directions acting on a tooth at a certain swept angle of cut and the ratio of average cutting forces in two directions per tooth period are functions of the swept angle of cut and the ratio of radial to tangential cutting forces. Using these parameters, the radial immersion ratio is estimated. Various experiments are performed to verify the proposed method. The results show that the radial immersion ratio can be estimated by this method regardless of other cutting conditions.

• 한국생산제조학회지
• /
• 제9권6호
• /
• pp.57-64
• /
• 2000

In order to regulate cutting torque in milling, monitoring system should be set to a certain threshold. Radial immersion ratio is an important factor to determine the threshold and should be estimated in process for automatic regulation. In this paper, on-line estimation of the radial immersion ratio using spindle motor current in face milling is presented. When a tooth finishes sweeping, a sudden drop of cutting torque occurs. This torque drop is equal to the cutting torque acting on a single tooth at the swept angle of cut and can be acquired from cutting torque signals. Average cutting torque per revolution can also be calculated from cutting torque signals. The ratio of cutting torque acting on a single tooth at the swept angle of cut to the average cutting torque per revolution is a function of the swept angle of cut and the number of teeth. Using the magnitude of this ratio, the radial immersion ratio is estimated. Identical algorithm is adopted to estimate the immersion ratio based on the spindle motor current measurement. The experiments performed under different cutting conditions show that the radial immersion ratio can be estimated within 10% error range by the proposed method using spindle motor current. Varying immersion ratio is also estimated well using the presented algorithm.

• 대한전자공학회:학술대회논문집
• /
• 대한전자공학회 2001년도 하계종합학술대회 논문집(4)
• /
• pp.33-36
• /
• 2001

We introduce, in this paper, the face recognition method that improves recognition rate and training time in eigen system. To increase recognition rate we use Gabor filter. To reduce the increasing training time owing to use Gabor filtering, we extract new feature vectors that are made with average and standard deviation. In experimental results, we get higher recognition rate and shorter training time in improved system than it in original eigen system.

• Maxillofacial Plastic and Reconstructive Surgery
• /
• 제16권3호
• /
• pp.390-395
• /
• 1994

The author evaluated the nasal asymmetry after primary operations in the patients with unilateral cleft lip using full face photographs. The results are as follows : 1. Nasal deviation angle is average 2.98+3.01 degree(ranged from 0 to 10 degree), significantly different from control group(p<0.05). 2. Nasal deviation angle is average 5.1% when inter-medial canthal distance is 100%, significantly different from control group(p<0.05). 3. Difference in nostril size between cleft and noncleft side is 2.1% when inter-medial canthal distance is 100%, significantly different from control group(p<0.05). 4. Nasal attractiveness analysis shows higher points in difference in nostril size, nasal deviation, nasal form in that order. 5. Nasal asymmetry after primary operations in the patients with unilateral cleft lip using full face photographs is related with Nasal deviation angle, Nasal deviation distance and Difference in nostril size between cleft and noncleft side.

• 한국컴퓨터정보학회:학술대회논문집
• /
• 한국컴퓨터정보학회 2011년도 제43차 동계학술발표논문집 19권1호
• /
• pp.315-316
• /
• 2011

일반적으로 영상에서 얼굴 영상을 검출하고 인식하는 알고리즘은 패턴 인식 연구에 있어서 인간과 컴퓨터의 상호작용의 연구라는 면에서 아주 중요한 문제로 연구되어 왔다. 본 논문에서는 고유얼굴을 이용하여 유클리디언 거리법과 퍼지기법의 인식률을 비교해보고자 한다. PCA(Principal Component Analysis) 방식은 우수한 인식 결과를 보장하는 얼굴인식 기법중의 하나이며, 얼굴 영상을 이용하여 공분산 행렬을 계산하고, 공분산 행렬을 통해 생성된 저차원의 벡터, 즉 고유얼굴(Eigenface)을 이용하여 가중치를 계산하고, 이 가중치를 기준으로 인식을 수행하는 기법이다. 이를 기반으로 하여, 본 논문에서는 전처리 과정, 고유얼굴 과정, 유클리디언 거리법 및 퍼지 소속도 함수 설계 과정, 신경망 학습과정, 인식과정으로 구성된 5단계의 얼굴 인식 알고리즘을 제안한다.

• Journal of Electrical Engineering and Technology
• /
• 제9권1호
• /
• pp.136-142
• /
• 2014

A major problem in Switched Reluctance Motor (SRM) is torque ripple, which causes undesirable acoustic noise and vibration. This work focuses on reducing the undesirable torque ripple in SRM by modifying stator and rotor geometry. This paper presents a comparative study on torque ripple minimization in SRM with modified pole shapes such as stator pole taper, stator pole face with non-uniform air gap and pole shoe attached to rotor pole. Further this paper presents a detailed sensitivity analysis of the effect of different geometrical parameters that alter the pole face shapes on the performance of SRM. The analysis is performed using finite-element method considering average torque and torque ripple as performance parameters. Based on the analysis, a design combining stator pole taper with non-uniform air gap is proposed to improve the torque characteristics of SRM. The dynamic characteristics of the proposed design are simulated and the results show satisfactory reduction in torque ripple.

• 대한전자공학회논문지SP
• /
• 제42권1호
• /
• pp.9-16
• /
• 2005

본 논문에서는 스테레오 영상에서 좌ㆍ우측 영상을 입력받아 거리 변화에 따른 얼굴인식률을 PCA(Principal Component Analysis) 알고리듬으로 비교한다. 제안된 방법에서는 RGB컬러공간에서 YCbCr컬러공간으로 변환하여 얼굴영역을 검출한다. 또한 스테레오 영상을 이용하여 거리를 취득한 후 추출된 얼굴영상의 확대 및 축소하여 보다 강건한 얼굴영역을 추출하고, PCA 알고리듬으로 인식률을 실험하였다. 취득된 얼굴영상의 평균적인 인식결과로 98.61%(30cm), 98.91%(50cm), 99.05%(100cm), 99.90%(120cm), 97.31%(150cm), 96.71%(200cm)의 인식률을 얻을 수 있었다. 따라서 실험을 통하여 제안된 방법은 거리에 따라 확대 및 축소를 적용하면 높은 인식률을 얻을 수 있음을 보였다.

• International Journal of Computer Science & Network Security
• /
• 제21권8호
• /
• pp.288-296
• /
• 2021

Automated face recognition in a runtime environment is gaining more and more important in the fields of surveillance and urban security. This is a difficult task keeping in mind the constantly volatile image landscape with varying features and attributes. For a system to be beneficial in industrial settings, it is pertinent that its efficiency isn't compromised when running on roads, intersections, and busy streets. However, recognition in such uncontrolled circumstances is a major problem in real-life applications. In this paper, the main problem of face recognition in which full face is not visible (Occlusion). This is a common occurrence as any person can change his features by wearing a scarf, sunglass or by merely growing a mustache or beard. Such types of discrepancies in facial appearance are frequently stumbled upon in an uncontrolled circumstance and possibly will be a reason to the security systems which are based upon face recognition. These types of variations are very common in a real-life environment. It has been analyzed that it has been studied less in literature but now researchers have a major focus on this type of variation. Existing state-of-the-art techniques suffer from several limitations. Most significant amongst them are low level of usability and poor response time in case of any calamity. In this paper, an improved face recognition system is developed to solve the problem of occlusion known as FRS-OCC. To build the FRS-OCC system, the color and texture features are used and then an incremental learning algorithm (Learn++) to select more informative features. Afterward, the trained stack-based autoencoder (SAE) deep learning algorithm is used to recognize a human face. Overall, the FRS-OCC system is used to introduce such algorithms which enhance the response time to guarantee a benchmark quality of service in any situation. To test and evaluate the performance of the proposed FRS-OCC system, the AR face dataset is utilized. On average, the FRS-OCC system is outperformed and achieved SE of 98.82%, SP of 98.49%, AC of 98.76% and AUC of 0.9995 compared to other state-of-the-art methods. The obtained results indicate that the FRS-OCC system can be used in any surveillance application.

• 대한치과교정학회지
• /
• 제36권5호
• /
• pp.369-379
• /
• 2006

본 연구는 수직적 안모유형에 따른 치료 후 교합평면 변화양상을 조사하여 향후 적절한 치료계획과 기전의 설정에 도움이 되고자 시행하였다. 골격성 I급 부정교합으로 진단되어 비발치로 치료받은 성민 60명(남자 28명, 여자 32명)을 대상으로 하였으며, Ricketts법의 facial axis, facial depth, mandibular plane angle, lower face height, mandibular arc의 5개 항목을 이용하여 한국 성인의 정상교합자 통계치의 기준에 따라 short face type (1군), average face type (2군), long face type (3군)으로 분류하였다. 각 군의 치료 전, 치료 종료, 종료 후 1년의 측모두부 방사선사진 계측치를 비교 분석하였다. 1군은 치료 종료 시와 비교하여 유지기간에 일반적 교합평면각, 기능적 교합 평면각, L6/L1, MP-L6 항목이 유의하게 감소하였고(p < 0.01) L1-FOP 항목은 유의하게 증가하였다(p < 0.001). 2군은 유지기간에 유의한 변화를 보이지 않았으며 3군은 유지기간에 기능적 교합평면각이 유의하게 증가하였다(p < 0.05). 기능적 교합평면각의 치료 후 변화량은 각 군 간에 유의한 차이를 보였으며 특히, 1군과 3군 간에 매우 큰 유의성을 보였다. 따라서 치아의 압하, 정출 및 치료 후 전치부 피개 등에 있어 유지기간 중 교합평면각 변화의 고려가 필요하다고 생각된다.