• 대한전자공학회:학술대회논문집
• /
• 대한전자공학회 2001년도 하계종합학술대회 논문집(4)
• /
• pp.231-234
• /
• 2001

This paper proposes a face recognition technique that effectively combines elastic graph matching (EGM) and Fisherface algorithm. EGM as one of dynamic lint architecture uses not only face-shape but also the gray information of image, and Fisherface algorithm as a class specific method is robust about variations such as lighting direction and facial expression. In the proposed face recognition adopting the above two methods, the linear projection per node of an image graph reduces dimensionality of labeled graph vector and provides a feature space to be used effectively for the classification. In comparison with a conventional method, the proposed approach could obtain satisfactory results in the perspectives of recognition rates and speeds. Especially, we could get maximum recognition rate of 99.3% by leaving-one-out method for the experiments with the Yale Face Databases.

• 한국방송∙미디어공학회:학술대회논문집
• /
• 한국방송공학회 2009년도 IWAIT
• /
• pp.446-449
• /
• 2009

In this paper, we present a novel method for reconstructing a super-resolution image using multi-view low-resolution images captured for depth varying scene without requiring complex analysis such as depth estimation and feature matching. The proposed method is based on the iterative back projection technique that is extended to the 3D volume domain (i.e., space + depth), unlike the conventional superresolution methods that handle only 2D translation among captured images.

• 한국방송∙미디어공학회:학술대회논문집
• /
• 한국방송공학회 2009년도 IWAIT
• /
• pp.737-741
• /
• 2009

This paper addresses the factorization method to estimate the projective structure of a scene from feature (points) correspondences over images with occlusions. We propose both a column and a row space approaches to estimate the depth parameter using the subspace constraints. The projective depth parameters are estimated by maximizing projection onto the subspace based either on the Joint Projection matrix (JPM) or on the the Joint Structure matrix (JSM). We perform the maximization over significant observation and employ Tardif's Camera Basis Constraints (CBC) method for the matrix factorization, thus the missing data problem can be overcome. The depth estimation and the matrix factorization alternate until convergence is reached. Result of Experiments on both real and synthetic image sequences has confirmed the effectiveness of our proposed method.

• 한국통신학회논문지
• /
• 제21권8호
• /
• pp.1874-1884
• /
• 1996

In this paper, we proposed automatic searching algorithm of thresholding value using multilevel thresholding for face segmentation from input bust image effectively. The proposed algorithm extracted the thresholding value of brightness that is formed background region, face region and hair region without illumination, background and face size from input image. The statistical projection analysis project the brightness of multilevel thresholding image into horizontal and vertical direction and decide the thresholding value of face. And the algorithm extracted elliptical type block of face from input image in order to reduce the back ground region and hair region efficiently. The proposed algorithm can reduce searching area of feature extraction and processing time for face recognication.

• 한국자기공명학회논문지
• /
• 제17권2호
• /
• pp.92-97
• /
• 2013

We study the electron spin resonance line-width (ESRLW) of $Fe^{3+}$ in crystalline $LiNbO_3$ ; the ESRLW is obtained using the projection operator method (POM) developed by Argyres and Sigel. The ESRLW is calculated to be axially symmetric about the c-axis and is analyzed by the spin Hamiltonian with an isotopic g factor at a frequency of 9.5 GHz. The ESRLW increases exponentially as the temperature increases, and the ESRLW is almost constant in the high-temperature region (T>8000 K). This kind of temperature dependence of the ESRLW indicates a motional narrowing of the spectrum when $Fe^{3+}$ ions substitute the $Nb^{5+}$ ions in an off-center position. It is clear from this feature that there are two different regions in the graph of the temperature dependence of the ESRLW.

• Journal of Biosystems Engineering
• /
• 제24권2호
• /
• pp.159-166
• /
• 1999

A system to extract continuously the real 3-D geometric fearture information from 2-D image of an object, which is fed randomly via conveyor has been developed. Two sets of structured laser lightings were utilized. And the laser structured light projection image was acquired using the camera from the signal of the photo-sensor mounted on the conveyor. Camera coordinate calibration matrix was obtained, which transforms 2-D image coordinate information into 3-D world space coordinate using known 6 points. The maximum error after calibration showed 1.5 mm within the height range of 103mm. The correlation equation between the shift amount of the laser light and the height was generated. Height information estimated after correlation showed the maximum error of 0.4mm within the height range of 103mm. An interactive 3-D geometric feature extracting software was developed using Microsoft Visual C++ 4.0 under Windows system environment. Extracted 3-D geometric feature information was reconstructed into 3-D surface using MATLAB.

• 한국멀티미디어학회논문지
• /
• 제6권1호
• /
• pp.89-96
• /
• 2003

본 논문은 카오스 이론의 프랙탈 차원과 수정된 에농 어트랙터를 이용하여 인쇄체 숫자를 인식하는 새로운 방법을 제안한다. 먼저 숫자 영상으로부터 망 특징 투영 특징, 교차거리 특징을 1차 구한 후, 이 특징들을 시계열 데이터로 변환한다. 그리고 본 논문에서 제안한 수정된 에농 시스템을 이용하여 프랙탈 차원을 나타내는 자연 척도 및 정보 비트값을 구한다. 마지막으로 표준패턴 데이터베이스와 비교하여, 최소 거리값을 이용하여 숫자 인식을 행한다. 실험 결과 10가지 숫자에 대하여 100%의 분류율을 나타내었고, 또한 실제 문서를 대상으로 실험한 결과 90%의 인식률과 초당 26자의 인식속도를 보임으로써 제안된 방법의 유효성을 보였다.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제12권3호
• /
• pp.1243-1263
• /
• 2018

The two-stage linear discrimination analysis (TSLDA) is a feature extraction technique to solve the small size sample problem in the field of image recognition. The TSLDA has retained all subspace information of the between-class scatter and within-class scatter. However, the feature information in the four subspaces may not be entirely beneficial for classification, and the regularization procedure for eliminating singular metrics in TSLDA has higher time complexity. In order to address these drawbacks, this paper proposes an improved two-stage linear discriminant analysis (Improved TSLDA). The Improved TSLDA proposes a selection and compression method to extract superior feature information from the four subspaces to constitute optimal projection space, where it defines a single Fisher criterion to measure the importance of single feature vector. Meanwhile, Improved TSLDA also applies an approximation matrix method to eliminate the singular matrices and reduce its time complexity. This paper presents comparative experiments on five face databases and one handwritten digit database to validate the effectiveness of the Improved TSLDA.

• 한국정보처리학회논문지
• /
• 제7권5호
• /
• pp.1361-1369
• /
• 2000

For the implementation of 3D Sound Localization system, the binaural filtering by HRTFs is generally employed. But the HRTF filter is of high order and its coefficients for all directions have to be stored, which imposes a rather large memory requirement. To cope with this, research works have centered on obtaining low dimensional HRTF representations without significant loss of information and synthesizing the original HRTF efficiently, by means of feature extraction methods for multivariate dat including PCA. In these researches, conventional linear PCA was applied to the frequency domain HRTF data and using relatively small number of principal components the original HRTFs could be synthesized in approximation. In this paper we applied neural network based nonlinear PCA model (NLPCA) and the nonlinear PLS repression model (NLPLS) for this low dimensional HRTF modeling and analyze the results in comparison with the PCA. The NLPCA that performs projection of data onto the nonlinear surfaces showed the capability of more efficient HRTF feature extraction than linear PCA and the NLPLS regression model that incorporates the direction information in feature extraction yielded more stable results in synthesizing general HRTFs not included in the model training.

• 한국정보과학회논문지:데이타베이스
• /
• 제33권4호
• /
• pp.384-395
• /
• 2006

항공 및 위성 사진과 같은 지형 / 지물 이미지는 대부분 비슷한 색상과 질감을 갖는다. 따라서 지형 / 지물 이미지 데이타베이스에서 질의 이미지를 정확하게 검색하기 위해서는 이미지의 형태 특정을 추출하는 것이 매우 중요하다. 본 논문에서는 지형 / 지물 이미지 검색을 위한 형태 특정 추출 방법을 제안한다. 이 방법은 이미지내 홀(hole) 이 있거나 이미지가 연결되지 않은 영역들로 구성되어 있을 경우에도 형태 특징들을 효과적으로 추출할 수 있다. 그리고 세그멘테이션 (segmentation)과 같은 특정 추출의 전처리 과정 (pre-processing) 에서 발생할 수 있는 오류에 강인하다. 본 논문에서 제안하는 방법이 기존의 방법들 보다 우수함을 다양한 이미지 검색 실험을 통해 보인다.