• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.14 no.12
• /
• pp.2622-2627
• /
• 2010

The accuracy of face recognition used unmanned security system is very important and necessary. However, face recognition is a lot of restriction due to the change of distortion of face image, illumination, face size, face expression, round image. We propose a hybrid neural network for improve the performance of the face recognition. The proposed method is consisted of SOM and LVQ. In order to verify usefulness of the proposed method, we make a comparison between eigenface method, hidden Markov model method, multi-layer neural network.

• Journal of Korea Multimedia Society
• /
• v.9 no.1
• /
• pp.33-40
• /
• 2006

This paper presents an efficient face recognition method using both first moment of image and eigenvector. First moment is a method for finding centroid of image, which is applied to exclude the needless backgrounds in the face recognitions by shitting to the centroid of face image. Eigenvector which are the basis images as face features, is extracted by principal component analysis(PCA). This is to improve the recognition performance by excluding the redundancy considering to second-order statistics of face image. The proposed methods has been applied to the problem for recognizing the 60 face images(15 persons *4 scenes) of 320*243 pixels. The 3 distances such as city-block, Euclidean, negative angle are used as measures when match the probe images to the nearest gallery images. In case of the 45 face images, the experimental results show that the recognition rate of the proposed methods is about 1.6 times and its the classification is about 5.6 times higher than conventional PCA without preprocessing. The city-block has been relatively achieved more an accurate classification than Euclidean or negative angle.

• Proceedings of the IEEK Conference
• /
• 2000.09a
• /
• pp.861-864
• /
• 2000

본 논문에서는 물체의 고유 칼라 정보 복원을 통하여 조명의 영향을 받지 않는 칼라 기반 얼굴검출 기법을 제안한다. 즉 주위 조명 영향으로부터 RGB 성분 계수를 파악하여 조명 성분에 영향을 받은 성분을 상쇄시키고, 색포화도와 밝기값 보상을 통해 고유 칼라를 복원(color recover)하는 실험을 하였고, 복원된 영상을 YCbCr 좌표계로 변환시킨 후, CbCr 각각에 대해 살색 성분이 나타내는 일정한 범위내의 부분을 검출하였다. 또한 이 진화 과정에서 생긴 잡음들을 형태학적인 모폴로지 필터를 통해 제거하였으며, 살색 후보 영역 중 같은 영역들은 레이블링하여 얼굴 후보 영역을 생성하였다. 그러나 칼라 정보만으로는 검출된 영역이 얼굴인지를 판단하기가 매우 어렵다. 그러므로 본 연구에서는 인간시각에 기반한 Gabor 필터를 사용하여, 검출된 살색 영역이 최종적으로 얼굴인지를 판별하는 효율적인 알고리즘을 제안한다.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2006.05a
• /
• pp.800-803
• /
• 2006

In this paper, we propose a method for improving the performance of the face recognition using a hybrid neural network. The propose method focused on improving face recognition technique using SOM and LVQ. In order to verify the effectiveness of the proposed method, we performed simulations on face database supplied ORL. The results show that the proposed method considerably improves on the performance of the eigenface, hidden markov model, multilayer neural network.

• Proceedings of the Korea Contents Association Conference
• /
• 2004.11a
• /
• pp.348-351
• /
• 2004

In this paper, we propose the face recognition method using Harr wavelet transform and 2D PCA. While previous PCA computed the covariance matrix by using one dimensional vectors, 2D PCA computed the covarinace matrix by using direct two dimensional image and extracted feature vector by solving eigenvalue problem. To gain the face image having the low dimension and robust property, the proposed method uses wavelet transformation. We apply the LL band image data to 2D PCA for face recognition. The experimental results indicate that our method improves recognition rate than 2D PCA into original image.

• Journal of the Institute of Convergence Signal Processing
• /
• v.13 no.3
• /
• pp.150-155
• /
• 2012

POC(Phase-Only Correlation) is a useful method that can conduct face recognition without using feature extraction or eigenface, but uses Fourier transformation for square areas. In this paper, we propose an effective face area to increase the performance of face recognition using POC. Specifically, three areas are experimented for POC. The frist area is the square area that includes head and space. The second area is the square area from ear to ear horizontally and from the end of chin to the forehead vertically. The third area is the square area from the line under the lips to the forehead vertically and from cheek to cheek horizontally. Experimental results show that the second face area has the best advantage among the three types of areas to define the threshold for POC.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.16 no.10
• /
• pp.2121-2128
• /
• 2012

This paper proposed the real time face area detection using Wavelet transform and the strong detection algorithm that satisfies the efficiency of computation and detection performance at the same time was proposed. The detected face image recognizes the face by configuring the low-dimensional face symbol through the principal component analysis. The proposed method is well suited for real-time system construction because it doesn't require a lot of computation compared to the existing geometric feature-based method or appearance-based method and it can maintain high recognition rate using the minimum amount of information. In addition, in order to reduce the wrong recognition or recognition error occurred during face recognition, the input symbol of Hidden Markov Model is used by configuring the feature values projected to the unique space as a certain symbol through clustering algorithm. By doing so, any input face will be recognized as a face model that has the highest probability. As a result of experiment, when comparing the existing method Euclidean and Mahananobis, the proposed method showed superior recognition performance in incorrect matching or matching error.

• Journal of Intelligence and Information Systems
• /
• v.11 no.1
• /
• pp.95-105
• /
• 2005

As the security system using facial recognition, the recognition performance depends on the environments (e. g. face expression, hair style, age and make-up etc.) For the revision of easily changeable environment, it's generally used to set up the threshold, replace the face image which covers the threshold into images already registered, and update the face images additionally. However, this usage has the weakness of inaccuracy matching results or can easily active by analogous face images. So, we propose the genetic algorithm which absorbs greatly the facial similarity degree and the recognition target variety, and has excellence studying capacity to avoid registering inaccuracy. We experimented variable and similar face images (each 30 face images per one, total 300 images) and performed inherent face images based on ingredient analysis as face recognition technique. The proposed method resulted in not only the recognition improvement of a dominant gene but also decreasing the reaction rate to a recessive gene.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.7 no.3
• /
• pp.499-505
• /
• 2003

This paper proposes methods to extract face elements and facial characteristics points for face recognition. We select a candidate region of the face elements with geometrical information between them inside the extracted face region with skin color and extract them using their inherent features. The facial characteristics to be applied to face recognition is expressed with geometrical relation such as distance and angle between the extracted face elements. Experiment results shows good performance to extract of face elements.

• Proceedings of the Korean Information Science Society Conference
• /
• 1998.10c
• /
• pp.452-454
• /
• 1998

본 논문에서는 MPEG 압축 비디오 상에서 얼굴 영역을 추출하고 이를 인식하는 방법에 대하여 제안한다. 제안된 방법은 크게 MPEG 압축 비디오의 처리를 위한 축소된 DC 영상의 구성 단계, 축소된 DC 영상에서의 얼굴 영역 추출 단계, 그리고 얼굴 영역이 추출된 프레임에 대한 압축 복원 및 얼굴 인식의 3단계로 구성되어있다. DC 영상의 구성 단계에서는 압축 복원 없이 DCT 계수의 DC 값과 2개의 AC 값만을 사용하여 부분적인 2차원 역 DCT 변환을 이용한 방법을 사용하였으며, 얼굴 영역 추출 단계에서는 DC 영상에 대해 얼굴의 색상 및 형태 정보를 이용한 얼굴 후보 영역 추출 방법과 K-L 변환 및 역 변환의 오차에 의한 얼굴 영역 추출 방법을 사용하였다. 얼굴 인식 단계에서는 얼굴 영역이 추출된 프레임에 대하여 GOP 단위의 압축 복원을 수행한 후 고유 얼굴 영상을 이용한 방법으로 얼굴 인식을 수행하였다. 제안된 방법의 성능을 검증하기 위하여 뉴스와 드라마 MPEG 비디오를 대상으로 실험을 수행하였으며, 실험 결과 제안된 방법이 효율적임을 알 수 있었다.