• Journal of KIISE:Software and Applications
• /
• v.31 no.2
• /
• pp.213-223
• /
• 2004

In this paper, we propose a new face detection method for real-time applications. It is based on the template-matching and appearance-based method. At first, we apply Min-max normalization with histogram equalization to the input image according to the variation of intensity. By applying the PCA transform to both the input image and template, PC components are obtained and they are applied to the LDA transform. Then, we estimate the distances between the input image and template, and we select one region which has the smallest distance. SVM is used for final decision whether the candidate face region is a real face or not. Since we detect a face region not the full region but within the $\pm$12 search window, our method shows a good speed and detection rate. Through the experiments with 6 category input videos, our algorithm shows the better performance than the existing methods that use only the PCA transform. and the PCA and LDA transform.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.16 no.3
• /
• pp.383-388
• /
• 2006

SBR is one of the most general sewage/wastewater treatment processes and, particularly, has an advantage in high concentration wastewater treatment like sewage wastewater. A Kernel PCA based fault diagnosis system for biological reaction in full-scale wastewater treatment plant was proposed using only common bio-chemical sensors such as ORP(Oxidation-Reduction Potential) and DO(Dissolved Oxygen). During the SBR operation, the operation status could be divided into normal status and abnormal status such as controller malfunction, influent disturbance and instrumental trouble. For the classification and diagnosis of these statuses, a series of preprocessing, dimension reduction using PCA, LDA, K-PCA and feature reduction was performed. Also, the diagnosis result using differential data was superior to that of raw data, and the fusion data show better results than other data. Also, the results of combination of K-PCA and LDA were better than those of LDA or (PCA+LDA). Finally, the fault recognition rate in case of using only ORP or DO was around maximum 97.03% and the fusion method showed better result of maximum 98.02%.

• Journal of IKEEE
• /
• v.6 no.1 s.10
• /
• pp.72-79
• /
• 2002

This paper presents an efficient face detection algorithm for color images with a complex background. The presented algorithm utilizes the color information and eigenface that is calculated by PCA-LDA (Principle Component Analysis - Linear Discriminant Analysis). The method of using the color information is faster than any other methods. Eigenface includes average information of the whole test faces. Therefore eigenface can decide that the candidate region is a face. The whole process is composed of two steps. First, it finds first face candidates region of skin tone using a color information in image. We can get a size and position of face candidate region. Second, we compare first face candidate region with eigenface, so decide that an image whether include a face or not. The advantages of the proposed approach include that increasing the detection speed by deciding a size and position of first face candidates region. Also, Betting 97% of the detection rate by comparing the eigenfaces calculated in PCA-LDA.

• MALSORI
• /
• no.65
• /
• pp.67-80
• /
• 2008

We have been building a text-independent speaker recognition system that is robust to unknown channel and noise environments. In this paper, we propose a linear transformation to obtain robust features. The transformation is optimized to maximize the distances between the Gaussian mixtures. We use rotation of the axes, to cope with the problem of scaling the transformation matrix. The proposed transformation is similar to PCA or LDA, but can achieve better result in some special cases where PCA and LDA can not work properly. We use YOHO database to evaluate the proposed method and compare the result with PCA and LDA. The results show that the proposed method outperforms all the baseline, PCA and LDA.

• Proceedings of the KIEE Conference
• /
• 2011.07a
• /
• pp.1960-1961
• /
• 2011

얼굴 인식에서 가장 많이 사용되고 있는 PCA(Principal Component Analysis)는 고차원의 얼굴 데이터를 낮은 차원으로 표현할 수 있다는 장점이 있다. LDA(Linear Discriminant Analysis)는 서로 다른 데이터를 잘 분리할 수 있으며, 얼굴 인식에서 우수한 성능을 보인다. 본 연구에서는 서로의 장점을 결합하여 PCA와 LDA를 혼합, 적용하였다. 고차원의 얼굴데이터를 PCA로 차원 축소한 후 LDA를 이용해 더욱 효과적인 분류가 되어 얼굴 인식률을 향상시킨다. 인식 모듈로는 pRBFNN(Polynomial Based Radial Basis Function Neural Networks) 모델을 구축하여 고차원 패턴인식 문제에 대한 해결책을 제시하고자 한다. 그리고 제안된 패턴분류기는 얼굴 데이터를 사용하여 성능을 확인한다.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2017.05a
• /
• pp.515-518
• /
• 2017

I used Principal Component Analysis(PCA) and Linear Discriminant Analysis(LDA) to utilize biometric technology for security. I used 14 korean consonants(ㄱ to ㅎ). And It has both information of gestures for each consonants and identity of user. So this experiment is set for this two aspects. I used database including 20 people's images. Each person did 140 action for every consonant with 10 trials. PCA and LDA must be applied on self-collected database using MATLAB programming. Equal Error Rate (EER) is used for evaluate performance of this analysis.

• The KIPS Transactions:PartB
• /
• v.11B no.6
• /
• pp.645-652
• /
• 2004

Among the biometrics, signature shows more larger variation than the other biometrics such as fingerprint and iris. In order to overcome this problem, we propose a robust offline signature verification method based on PCA and LDA. Signature is projected to vertical and horizontal axes by new grid partition method. And then feature extraction and decision is performed by PCA and LDA. Experimental results show that the proposed offline signature verification has lower False Reject Rate(FRR) and False Acceptance Rate(FAR) which are 1.45% and 2.1%, respectively.

• Proceedings of the Korean Information Science Society Conference
• /
• 2002.10d
• /
• pp.538-540
• /
• 2002

본 논문에서는 실시간 카메라 입력 영상에 적합한 얼굴 검출을 위해 다양한 외부적 환경에 덜 민감한 새로운 알고리즘을 제안한다. 빛이나 조명의 영향에 의한 오류를 방지하기 위해 전처리 과정을 포함시키고 형판 정합방법의 단점을 개선하기 위해 얼굴 인식에서 주로 쓰이는 방법인 주성분 분석(PCA :Principal Component Analyses) 변환을 적용하고. 생성된 주성분(Principal Component)을 선형 판별 분석(LDA: Linear Discriminant Analysis)의 입력으로 사용하는 방법을 통해 얼굴을 검출하도록 하였다. 실험을 위해 실제 환경과 같은 6개 카테고리의 동영상을 중심으로 실험한 결과, 본 논문에서 제안하는 방법이 기존의 PCA만을 이용한 방법보다 좋은 성능을 보여줌을 알 수 있었다.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.15 no.4
• /
• pp.492-497
• /
• 2005

Feature extraction is important to classify data with large dimension such as image data. The representative feature extraction methods lot feature extraction ate PCA, ICA, LDA and MLP, etc. These algorithms can be classified in two groups: unsupervised algorithms such as PCA, LDA, and supervised algorithms such as LDA, MLP. Among these two groups, supervised algorithms are more suitable to extract the features for classification because of the class information of input data. In this paper we suggest a new feature extraction algorithm PCA-FX which uses class information with PCA to extract ieatures for classification. We test our algorithm using Yale face database and compare the performance of proposed algorithm with those of other algorithms.

• Proceedings of the KIEE Conference
• /
• 2007.07a
• /
• pp.1924-1925
• /
• 2007

본 논문에서는 부정맥 판별을 위한 전처리 과정으로 PCA, LDA, ICA를 바탕으로 하여 정확도를 비교하여 보았다. 각각의 전처리는 고유의 특성을 가지고 있으며 본 논문의 목적은 부정맥 판별상 어떤 전처리가 더욱 정확성의 면에서 효과적인지를 알아보는 것이다. 본 논문의 데이터는 MIT-BIH에 기반하고 있으며, Beat의 분류는 정상(Normal), 좌각차단(Left Bundle Branch Block, LBBB), 우각차단(Right Bundle Branch Block, RBBB), 조기심실수축(Premature Ventricular Contraction, PVC), 조기심방수축(Atrial Premature Beat, APB), paced Beat, 심실보충수축(Ventricular Escape Beat)로 나누었다. 실험적 결과는 PCA-BPNN의 경우 95.53%, ICA-BPNN의 경우 93.95%, LDA-BPNN의 경우 96.42%로 LDA가 가장 ECG 부정맥 판별 응용에 있어 가장 효율적인 방법으로 나타났다.