• 제어로봇시스템학회:학술대회논문집
• /
• 제어로봇시스템학회 2005년도 ICCAS
• /
• pp.1034-1038
• /
• 2005

This paper presents a novel method to speed up neural network (NN) based face detection systems. NN-based face detection can be viewed as a classification and search problem. The proposed method formulates the search problem as an integer nonlinear optimization problem (INLP) and develops an improved genetic algorithm (IGA) to solve it. Each individual in the IGA represents a subwindow in an input image. The subwindows are evaluated by how well they match a NN-based face filter. A face is indicated when the filter response of the best particle is above a given threshold. Experimental results show that the proposed method leads to a speedup of 83 on $320{\times}240$ images compared to the traditional exhaustive search method.

• 패션비즈니스
• /
• 제14권1호
• /
• pp.43-56
• /
• 2010

In the modern 21st century society, the personal image is considered to be very important. As a result, the importance of presenting one's personal image through personal color in fashion and beauty related fields are increasing, and is the most realistic and practical field of color. When the color of the wardrobe and the skin color are in disharmony, that disharmony becomes the source of the lines and wrinkles that appear on one's face, resulting in shades. The boundary that is created when the color of the wardrobe and the skin color are in disharmony, it works negatively on one's image. When color arrangements are close or similar (in harmony) or are in complementary color arrangements or in strong contrasting state (contrasting harmony), it is generally believed to be beautifully harmonious. Personal color assessment is finding colors, through systematic and scientific methods, that improve the personal image by reaching harmony with skin colors that each and every individual are uniquely born with. In this study, one was able to learn the improved visual effects of the face image through creating harmony with the wardrobe and color shade make up and complementary colors that were selected based on personal colors. The base make up, through using the contrasting effects of the complementary colors which represents the supplementing, correcting, and complementing of the face image by contrasting with complementary colors, brings positive changes through correcting the base skin color. It is believed that this study finds its importance in that the improved image that is created by the overall harmony of the wardrobe and body can be used as valuable data in marketing and new product development efforts in the related industries.

• 한국컴퓨터정보학회논문지
• /
• 제17권9호
• /
• pp.49-56
• /
• 2012

본 논문은 얼굴 검출을 이용한 숏의 유형을 판단하는 방법론을 제시한다. 클로즈 업 숏이나 미디엄 숏, 롱 숏과 같은 숏의 유형은 영화의 서사 구조를 파악하는 주요한 단서이다. 클로즈 업을 통해 감독은 등장인물의 감정 상태를 묘사하고 롱 숏을 통해 인물이 처한 상황이나 배경을 묘사하게 된다. 인물의 심리나 감정의 변화, 인물이 처한 상황을 묘사하는 숏의 여러 유형은 인물과 카메라의 거리에 의해 결정된다. 따라서 화면에 등장하는 인물의 얼굴 크기를 알아내어 숏의 유형을 판단할 수 있다. 이를 위해 본 논문에서는 얼굴 검출을 통해 숏의 유형을 감지하는 방법론을 제시하고 시스템으로 구현하여 성능을 평가한다. 평가실험에서 클로즈 업 숏과 미디엄 숏의 감지 성능은 95%와 90%로 비교적 높게 나타났지만 얼굴의 윤곽이 불분명한 롱 숏의 경우 53.3%로 측정되었다.

• 전자공학회논문지CI
• /
• 제48권3호
• /
• pp.46-53
• /
• 2011

본 논문에서는 대표적인 특징점 추출 알고리즘인 SURF (Speeded Up Robust Features)를 이용한 얼굴 인식 방법을 소개한 다. 일반적으로, SURF를 이용한 물체 인식은 특징점 추출 및 정합만을 수행하지만, 본 논문에서 제안하는 SURF를 이용한 얼굴 인식 방법은 특징점 추출 및 정합뿐만 아니라 얼굴 영상 회전 및 특징점 검증을 추가로 수행한다. 얼굴 영상 회전은 특징점의 수를 증가시키기 위해 수행되며, 특징점 검증은 정확하게 정합된 특징점들을 찾기 위해 수행된다. 비록 본 논문에서 제안한 SURF를 이용한 얼굴 인식 방법은 PCA를 이용한 방법보다 연산 시간이 더 요구되었지만, 인식률은 보다 더 높았다. 이러한 실험 결과를 통해, 특징점 추출 알고리즘도 얼굴 인식에 적용할 수 있음을 확인할 수 있었다.

• 한국지능시스템학회논문지
• /
• 제13권4호
• /
• pp.461-468
• /
• 2003

In a sequence of images obtained by surveillance cameras, facial regions appear very small and their colors change abruptly by lighting condition. This paper proposes a new face detection scheme, robust on complex background, small size, and lighting conditions. The proposed method is consisted of three processes. In the first step, the candidates for the face regions are selected using face color distribution and motion information. In the second stage, the non-face regions are removed using face color ratio, boundary ratio, and average of column-wise intensity variation in the candidates. The face regions containing eyes and mouth are segmented and classified, and then they are scored using their topological relations in the last step. To speed up and improve a performance the above process, a block based image segmentation technique is used. The experiments have shown that the proposed algorithm detects faced regions with more than 91% of accuracy and less than 4.3% of false alarm rate.

• 대한전자공학회:학술대회논문집
• /
• 대한전자공학회 2000년도 하계종합학술대회 논문집(4)
• /
• pp.218-221
• /
• 2000

In this paper, we propose a new algorithm to detect human faces for controling a camera used in video conference. We model the distribution of skin color and set up the standard skin color in YIQ color space. An input video frame image is segmented into skin and non-skin segments by comparing the standard skin color and each pixels in the input video frame. Then, shape filler is applied to select face segments from skin segments. Our algorithm detects human faces in real time to control a camera to capture a human face with a proper size and position.

• 한국소성가공학회:학술대회논문집
• /
• 한국소성가공학회 1994년도 춘계학술대회 논문집
• /
• pp.218-222
• /
• 1994

The chassis of T.V, cassette, radio and telephone are made up thin-iron plates that makes a lot of female screws to assemble various parts. Then makes hole in the iron plate and forming female screws of burring face after the burring. Female screws forming in thin-iron plate is very difficult working then rolled forming process use of plastic deformation instead of cut forming. In this study goals are solve the trouble problem of industrial field and develop new model tap for prevent chips.

• 한국콘텐츠학회논문지
• /
• 제16권11호
• /
• pp.776-783
• /
• 2016

영화 속 얼굴 이미지에 대한 논의가 시작된 것은 영화가 하나의 예술임을 인정받기 시작한 무성 영화 시기부터 인데, 이 시기는 클로즈 업을 인식하는 두 관점에 따라 얼굴 이미지는 다른 의미와 기능을 갖게 된다. 첫째는 클로즈 업을 영화의 새로운 미학적 가능성을 지닌 독립체로 보는 관점인데, 이 경우 클로즈 업은 얼굴을 재현하는 독자적 수단으로 이해되어, 얼굴 이미지와 클로즈 업 모두에게 특권적 지위가 부여된다. 둘째, 클로즈 업 또한 서사를 완성하는 구성체 중 하나로 보는 관점인데, 이 경우 대상의 지표 성을 부곽 시키는 클로즈 업의 특성이 이해되지 못하기 때문에, 얼굴 이미지 또한 서사의 완결성을 와해시키지 않는 선에서, 다른 쇼트와 변별점이 없는 하나의 쇼트로 이해된다. 이 시기의 논의는 영화의 미장센을 구성하는 다양한 요소와의 관계를 통해 다양한 기능과 의미를 갖게 될 얼굴 이미지의 미학적 가능성에 대한 논의가 클로즈 업의 자장 안에서만 이루어진다는 한계가 있지만. 그럼에도 불구하고 영화 속 얼굴 이미지에 대한 미학적 탐구의 가능성을 발견 했다는 점과, 이 시기의 비평적 담론이 현재까지도 유효하다는 점에서 큰 의의가 있다.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제7권11호
• /
• pp.2720-2736
• /
• 2013

Multi-view face detection has become an active area for research in the last few years. In this paper, a novel multi-view human face detection algorithm based on improved real Adaboost is presented. Real Adaboost algorithm is improved by weighted combination of weak classifiers and the approximately best combination coefficients are obtained. After that, we proved that the function of sample weight adjusting method and weak classifier training method is to guarantee the independence of weak classifiers. A coarse-to-fine hierarchical face detector combining the high efficiency of Haar feature with pose estimation phase based on our real Adaboost algorithm is proposed. This algorithm reduces training time cost greatly compared with classical real Adaboost algorithm. In addition, it speeds up strong classifier converging and reduces the number of weak classifiers. For frontal face detection, the experiments on MIT+CMU frontal face test set result a 96.4% correct rate with 528 false alarms; for multi-view face in real time test set result a 94.7 % correct rate. The experimental results verified the effectiveness of the proposed approach.

• 전기학회논문지
• /
• 제64권6호
• /
• pp.900-911
• /
• 2015

In this paper, 3D face recognition model is designed by using Polynomial based RBFNN(Radial Basis Function Neural Network) and PNN(Polynomial Neural Network). Also recognition rate is performed by this model. In existing 2D face recognition model, the degradation of recognition rate may occur in external environments such as face features using a brightness of the video. So 3D face recognition is performed by using 3D scanner for improving disadvantage of 2D face recognition. In the preprocessing part, obtained 3D face images for the variation of each pose are changed as front image by using pose compensation. The depth data of face image shape is extracted by using Multiple point signature. And whole area of face depth information is obtained by using the tip of a nose as a reference point. Parameter optimization is carried out with the aid of both ABC(Artificial Bee Colony) and PSO(Particle Swarm Optimization) for effective training and recognition. Experimental data for face recognition is built up by the face images of students and researchers in IC&CI Lab of Suwon University. By using the images of 3D face extracted in IC&CI Lab. the performance of 3D face recognition is evaluated and compared according to two types of models as well as point signature method based on two kinds of depth data information.