• ETRI Journal
• /
• 제31권4호
• /
• pp.438-447
• /
• 2009

In this paper, a novel image projection technique for face recognition application is proposed which is based on linear discriminant analysis (LDA) combined with the relevance-weighted (RW) method. The projection is performed through 2-directional and 2-dimensional LDA, or $(2D)^2$LDA, which simultaneously works in row and column directions to solve the small sample size problem. Moreover, a weighted discriminant hyperplane is used in the between-class scatter matrix, and an RW method is used in the within-class scatter matrix to weigh the information to resolve confusable data in these classes. This technique is called the relevance-weighted $(2D)^2$LDA, or RW$(2D)^2$LDA, which is used for a more accurate discriminant decision than that produced by the conventional LDA or 2DLDA. The proposed technique has been successfully tested on four face databases. Experimental results indicate that the proposed RW$(2D)^2$LDA algorithm is more computationally efficient than the conventional algorithms because it has fewer features and faster times. It can also improve performance and has a maximum recognition rate of over 97%.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제12권1호
• /
• pp.368-391
• /
• 2018

Face recognition (FR) with a single sample per person (SSPP) is common in real-world face recognition applications. In this scenario, it is hard to predict intra-class variations of query samples by gallery samples due to the lack of sufficient training samples. Inspired by the fact that similar faces have similar intra-class variations, we propose a virtual sample generating algorithm called k nearest neighbors based virtual sample generating (kNNVSG) to enrich intra-class variation information for training samples. Furthermore, in order to use the intra-class variation information of the virtual samples generated by kNNVSG algorithm, we propose image set based multimanifold discriminant learning (ISMMDL) algorithm. For ISMMDL algorithm, it learns a projection matrix for each manifold modeled by the local patches of the images of each class, which aims to minimize the margins of intra-manifold and maximize the margins of inter-manifold simultaneously in low-dimensional feature space. Finally, by comprehensively using kNNVSG and ISMMDL algorithms, we propose k nearest neighbor virtual image set based multimanifold discriminant learning (kNNMMDL) approach for single sample face recognition (SSFR) tasks. Experimental results on AR, Multi-PIE and LFW face datasets demonstrate that our approach has promising abilities for SSFR with expression, illumination and disguise variations.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제9권11호
• /
• pp.4502-4518
• /
• 2015

Fisher linear discriminant analysis (LDA) is one of the most popular projection techniques for feature extraction and has been widely applied in face recognition. However, it cannot be used when encountering the single sample per person problem (SSPP) because the intra-class variations cannot be evaluated. In this paper, we propose a novel method called local similarity based linear discriminant analysis (LS_LDA) to solve this problem. Motivated by the "divide-conquer" strategy, we first divide the face into local blocks, and classify each local block, and then integrate all the classification results to make final decision. To make LDA feasible for SSPP problem, we further divide each block into overlapped patches and assume that these patches are from the same class. To improve the robustness of LS_LDA to outliers, we further propose local similarity based median discriminant analysis (LS_MDA), which uses class median vector to estimate the class population mean in LDA modeling. Experimental results on three popular databases show that our methods not only generalize well SSPP problem but also have strong robustness to expression, illumination, occlusion and time variation.

• 한국식품과학회지
• /
• 제28권2호
• /
• pp.298-304
• /
• 1996

참기름의 품질을 고소한 냄새, 탄 냄새, 전체적 인 품질 만족도의 세 항목으로 관능 평가하고 상온에서 휘발성 성분을 포집하여 정량 분석하였다 그리고 세 항목의 관능특성 변화에 가장 밀접한 휘발성 성분을 파악하고자 단계적 분별 분석을 실시하였으며 선정된 5개 peaks의 중요도를 정준 분별 분석, 분별 분석 그리고 주성분 분석으로 검토하였다. 5개 peaks 중에서도 가장 중요한 휘발성 성분은 2,5-dimethylpyrazine과 2-methylpyrazine이었으며 각각 고소한 냄새 및 탄 냄새의 척도로 적용할 수 있었다 관능적으로 좋은 품질의 참기름은 2,5-dimethylpyrazine과 2-methylpyrazine의 함량이 각각 $35.82{\sim}4.43$ppm,\;28.90{\sim}6.35ppm$인 것으로 밝혀졌다.

• International Journal of Human Ecology
• /
• 제16권2호
• /
• pp.21-31
• /
• 2015

This study explored determinants of family support that young renter households received to afford their housing costs. Microdata set of the 2014 Korea Housing Survey was used as secondary data for the study. Total 1,752,899 households headed by persons between 20 and 34 years of age and whose rental type was either Jeon-se or monthly rental with deposit in private rental units were selected as study subjects. For the data analysis, a series of discriminant analysis was conducted using IBM SPSS 21.0. Major findings were as follows. (1) Among the subjects, 28.2% were found to receive financial support from parents or other relatives. (2) To see the discriminant analysis results, a linear combination of seven household and housing characteristics (householder's gender, whether or not the householder worked in the previous week, whether or not the householders have a spouse, tenure type, structure type, location and deposit amount) could explain 44.6% of variance in young renter households' receipt of family support with a prediction accuracy of 77.2%. (3) To summarize the final discriminant model, Jeon-se renter households in location other than Incheon or Gyeonggi Province living in a unit in structure other than multifamily structure headed by younger householders that did not worked previous week or without spouse; with a greater deposit had the maximum tendency to receive family support to pay rental costs.

• 교육시설
• /
• 제4권4호
• /
• pp.29-41
• /
• 1997

The objective of this paper is to construct a model to be used in deciding whether to repair or rebuild school buildings is depending on their ages and other factors. The theme of this paper is the age is the main variable but other factors such as floor, innerwall, ceiling, door, inner window of the class room, outer window of the class room, inner window of the corridor, outer window of the corridor, middle window between the classroom and the corridor, light, heater, speaker, fire protection sensor, TV monitor, and telephone status would influence the final decisions. This paper employs an experimental case study method. Using the stepwise, statistical, classification method commonly used in discriminant analysis, it evaluates 12,766 rooms of 87 different high schools in Seoul. The result of this study indicates that some critical variables influencing the final decisions are the status of TV monitor, middle window between the classroom and the corridor, light, inner window of the corridor, fire protection sensor, innerwall, speaker utensil, outer window of the class room, and door of the class room. This paper also suggests a linear discriminant function will be used for this kind of studies. Finally the paper recommends policies with respect to the variables and discriminant functions evaluated.

• 정보보호학회논문지
• /
• 제24권6호
• /
• pp.1055-1063
• /
• 2014

전력소모량을 이용한 부채널 분석의 성능 향상을 위해 다양한 분석기법이 제안되고 있다. 이들 중, 사전처리 단계에서 적용 가능한 파형압축은 전력분석을 위한 소요시간을 단축하고 수집신호의 잡음성분을 줄이기 위해 널리 사용되는 방법이다. 본 논문에서는 영상처리 등에 많이 사용되고 있는 선형판별분석(Linear Discriminant Analysis)을 이용한 전력분석기법을 제안한다. 또한, 실험을 통해 기존의 파형압축방법 중 가장 성능이 좋은 것으로 알려진 주성분분석(Principal Component Analysis)을 이용한 방법과의 성능 비교를 통해 제안기법의 우수성을 증명한다.

• 한국조경학회지
• /
• 제29권1호
• /
• pp.11-21
• /
• 2001

This study investigates how motivations, preferences, and past experiences vary by each hikers trail choice at the Mt.Keyryong National Park. The purpose of this study is to find out the factors influencing behavioral choice in the recreation areas, and establish the fundamental theory for the efficient management of the resource and visitors. For this study, we have collected 472 respondents by on-site self-administrated questionnaire from the hikers in the park. The collected data were analyzed by the descriptive statistics and the discriminant analysis. The motivations variable of hiking participation on mountain trail were categorized three types; close-nature, escapism, and physical improvement. The preferences for trail environment were classified as four categories by factor analysis; preference for nature, safety, use density, and facilities. In descriptive statistics, the study showed that the experienced hikers prefer natural trials and hikers who have preference for close-nature select longer and deeper forest trails. The results of discriminant analysis indicate that the level of past experience is the most affectable in classification of trail choice. Such variables as motivation for close-nature and preference for nature were also appeared as affecting factors on classification of trail choice. Two discriminant functions were available, and 90.5 percent of analysis sample were correctly classified. In the validity analysis, 89 percent of holdout sample were correctly classified. These hit ratios ensures an accuracy by Press Q test. The result of this study is to be useful knowledge of the choice of detailed use environments in the same recreation areas.

• International Journal of Fuzzy Logic and Intelligent Systems
• /
• 제12권3호
• /
• pp.250-255
• /
• 2012

Face recognition has wide applications in security and surveillance systems as well as in robot vision and machine interfaces. Conventional challenges in face recognition include pose, illumination, and expression, and face recognition at a distance involves additional challenges because long-distance images are often degraded due to poor focusing and motion blurring. This study investigates the effectiveness of applying photon-counting linear discriminant analysis (Pc-LDA) to face recognition in harsh environments. A related technique, Fisher linear discriminant analysis, has been found to be optimal, but it often suffers from the singularity problem because the number of available training images is generally much smaller than the number of pixels. Pc-LDA, on the other hand, realizes the Fisher criterion in high-dimensional space without any dimensionality reduction. Therefore, it provides more invariant solutions to image recognition under distortion and degradation. Two decision rules are employed: one is based on Euclidean distance; the other, on normalized correlation. In the experiments, the asymptotic equivalence of the photon-counting method to the Fisher method is verified with simulated data. Degraded facial images are employed to demonstrate the robustness of the photon-counting classifier in harsh environments. Four types of blurring point spread functions are applied to the test images in order to simulate long-distance acquisition. The results are compared with those of conventional Eigen face and Fisher face methods. The results indicate that Pc-LDA is better than conventional facial recognition techniques.

• Journal of the Korean Academy of Child and Adolescent Psychiatry
• /
• 제30권1호
• /
• pp.9-16
• /
• 2019

Objectives: The purpose of this study was to analyze the discriminant validity and the clinical cut off scores of the Child Behavior Checklist 1.5-5 (CBCL 1.5-5) in the diagnosis of autism spectrum disorder (ASD) and non-ASD. Methods: In total, 104 ASD and 441 non-ASD infants were included in the study. T-test, discriminant analysis, receiver operating characteristic (ROC) curve analysis, and odds ratio analysis were performed on the data. Results: The discriminant validity was confirmed by mean differences and discriminant analysis on the subscales of Emotionally reactive, Somatic complaints, Withdrawn, Sleep problems, Attention problems, Aggressive behavior, Internalizing problems, Externalizing problems, and Total problems, along with the Diagnostic and Statistical Manual of Mental Disorders (DSM)-oriented scales between the two groups. ROC analysis showed that the following subscales significantly separated ASD from normal infants: Emotionally reactive, Somatic complaints, Withdrawn, Sleep problems, Attention problems, Aggressive behavior, Internalizing problems, Externalizing problems, Total problems, and DSM pervasive developmental problems. Moreover, the clinical cut off score criteria adopted in the Korean-CBCL 1.5-5 were shown to be valid for the subscales Withdrawn, Internalizing problems, Externalizing problems, Total problems, and DSM pervasive developmental problems. Conclusion: The subscales of Withdrawn, Internalizing problems, Externalizing problems, Total problems, and DSM pervasive developmental problems significantly discriminated infants with ASD.