• Journal of the Institute of Electronics Engineers of Korea SP
• /
• v.42 no.6
• /
• pp.55-62
• /
• 2005

This paper proposes a method to recognize a face using DCT/LDA where LDA is applied to DCT coefficients of an input face image. In the proposed method, SSS problem of LDA due to less number of training data than the size of feature space can be avoided by expressing an input image in low dimensional space using DCT coefficients. In terms of the recognition rate, both the proposed method and the PCA/LDA method have shown almost equal performance while the training time of the proposed method is much shorter than the other. This is because DCT has the fixed number of basis vectors while the property of energy compaction rate is similar to that of PCA. Although depending on the number of coefficients employed for the recognition, the experimental results show that the performance of the proposed method in terms of recognition rate is very comparable to PCA/LDA method and other DCT/LDA methods, and it can be trained 13,000 times faster than PCA/LDA method.

• The Journal of the Korea Contents Association
• /
• v.4 no.3
• /
• pp.107-113
• /
• 2004

This paper improved the performance of the face recognition system using the PCA/LDA hybrid method based on the facial geometric feature and the Wavelet transform. Because the previous PCA/LDA methods have measured the similarity according to the formal dispersion, they could not reflect facial boundaries exactly In order to recover this defect, this paper proposed the method using the distance between eyes and mouth. If the difference of the measured distances on the query and the training images is over the given threshold, then the method reorders the candidate images according to energy feature vectors of eyes, a nose, and a chin. To evaluate the performance of the proposed method the computer simulations have been performed with four hundred facial images in the ORL database. The results showed that our method improves about 4％ recognition rate over the previous PCA/LDA method.

• Journal of the Institute of Electronics Engineers of Korea SP
• /
• v.41 no.4
• /
• pp.45-50
• /
• 2004

In this paper, we propose a new PCA-based LDA Mixture Algorithm(PLMA) for real-time face recognition system. This system greatly consists of the two parts: 1) face extraction part; 2) face recognition part. In the face extraction part we applied subtraction image, color filtering, eyes and mouth region detection, and normalization method, and in the face recognition part we used the method mixing PCA and LDA in extracted face candidate region images. The existing recognition system using only PCA showed low recognition rates, and it is hard in the recognition system using only LDA to apply LDA to the input images as it is when the number of image pixels ire small as compared with the training set. To overcome these shortcomings, we reduced dimension as we apply PCA to the normalized images, and apply LDA to the compressed images, therefore it is possible for us to do real-time recognition, and we are also capable of improving recognition rates. We have experimented using self-organized DAUface database to evaluate the performance of the proposed system. The experimental results show that the proposed method outperform PCA, LDA and ICA method within the framework of recognition accuracy.

• Journal of Korea Multimedia Society
• /
• v.17 no.12
• /
• pp.1446-1452
• /
• 2014

Existing LDA uses the transform matrix that maximizes distance between classes. So we have to convert from an image to one-dimensional vector as training vector. However, in 2D-LDA, we can directly use two-dimensional image itself as training matrix, so that the classification performance can be enhanced about 20% comparing LDA, since the training matrix preserves the spatial information of two-dimensional image. However 2D-LDA uses same calculation schema for transformation matrix and therefore both LDA and 2D-LDA has the heteroscedastic problem which means that the class classification cannot obtain beneficial information of spatial distances of class clusters since LDA uses only data correlation-based covariance matrix of the training data without any reference to distances between classes. In this paper, we propose a new method to apply training matrix of 2D-LDA by using WPS-LDA idea that calculates the reciprocal of distance between classes and apply this weight to between class scatter matrix. The experimental result shows that the discriminating power of proposed 2D-LDA with weighted between class scatter has been improved up to 2% than original 2D-LDA. This method has good performance, especially when the distance between two classes is very close and the dimension of projection axis is low.

• Journal of Korea Multimedia Society
• /
• v.13 no.6
• /
• pp.817-824
• /
• 2010

Linear Discriminant Analysis (LDA) has been successfully applied for dimension reduction in face recognition. However, LDA requires the transformation of a face image to a one-dimensional vector and this process can cause the correlation information among neighboring pixels to be disregarded. On the other hand, 2D-LDA uses 2D images directly without a transformation process and it has been shown to be superior to the traditional LDA. Nevertheless, there are some problems in 2D-LDA. First, it is difficult to determine the optimal number of feature vectors in a reduced dimensional space. Second, the size of rectangular windows used in 2D-LDA makes strong impacts on classification accuracies but there is no reliable way to determine an optimal window size. In this paper, we propose a new algorithm to overcome those problems in 2D-LDA. We adopt an ensemble approach which combines several classifiers obtained by utilizing various window sizes. And a practical method to determine the number of feature vectors is also presented. Experimental results demonstrate that the proposed method can overcome the difficulties with choosing an optimal window size and the number of feature vectors.

• ETRI Journal
• /
• v.31 no.4
• /
• 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%.

• Journal of KIISE:Software and Applications
• /
• v.32 no.8
• /
• pp.789-794
• /
• 2005

LDA (Linear Discriminant Analysis) provides the projection that discriminates the data well, and shows a very good performance for face recognition. However, since LDA provides only one transformation matrix over whole data, it is not sufficient to discriminate the complex data consisting of many classes like honan faces. To overcome this weakness, we propose a new face recognition method, called LDA mixture model, that the set of alf classes are partitioned into several clusters and we get a transformation matrix for each cluster. This detailed representation will improve the classification performance greatly. In the simulation of face recognition, LDA mixture model outperforms PCA, LDA, and PCA mixture model in terms of classification performance.

• The Journal of the Acoustical Society of Korea
• /
• v.18 no.7
• /
• pp.17-22
• /
• 1999

In this paper we propose a new mixed method of LDA and tri-tone model to predict Korean prosodic break indices(PBI) for a given utterance. PBI can be used as an important cue of syntactic discontinuity in continuous speech recognition(CSR). The model consists of three steps. At the first step, PBI was predicted with the information of syllable and pause duration through the linear discriminant analysis (LDA) method. At the second step, syllable tone information was used to estimate PBI. In this step we used vector quantization (VQ) for coding the syllable tones and PBI is estimated by tri-tone model. In the last step, two PBI predictors were integrated by a weight factor. The proposed method was tested on 200 literal style spoken sentences. The experimental results showed 72% accuracy.

• Journal of the Institute of Electronics Engineers of Korea CI
• /
• v.44 no.3
• /
• pp.98-107
• /
• 2007

For high-dimensional pattern recognition, such as face classification, the small number of training samples leads to the Small Sample Size problem when the number of pattern samples is smaller than the number of dimensionality. Recently, various LDA-extensions have been developed, including LDA, PCA+LDA, and Direct-LDA, to address the problem. This paper proposes a method of improving the classification efficiency by increasing the number of (sub)-classes through pre-clustering a training set prior to the execution of Direct-LDA. In LDA (or Direct-LDA), since the number of classes of the training set puts a limit to the dimensionality to be reduced, it is increased to the number of sub-classes that is obtained through clustering so that the classification performance of LDA-extensions can be improved. In other words, the eigen space of the training set consists of the range space and the null space, and the dimensionality of the range space increases as the number of classes increases. Therefore, when constructing the transformation matrix, through minimizing the null space, the loss of discriminatve information resulted from this space can be minimized. Experimental results for the artificial data of X-OR samples as well as the bench mark face databases of AT&T and Yale demonstrate that the classification efficiency of the proposed method could be improved.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.10 no.6
• /
• pp.2527-2545
• /
• 2016

Measuring the similarity of given samples is a key problem of recognition, clustering, retrieval and related applications. A number of works, e.g. kernel method and metric learning, have been contributed to this problem. The challenge of similarity learning is to find a similarity robust to intra-class variance and simultaneously selective to inter-class characteristic. We observed that, the similarity measure can be improved if the data distribution and hidden semantic information are exploited in a more sophisticated way. In this paper, we propose a similarity learning approach for retrieval and recognition. The approach, termed as LDA-FEK, derives free energy kernel (FEK) from Latent Dirichlet Allocation (LDA). First, it trains LDA and constructs kernel using the parameters and variables of the trained model. Then, the unknown kernel parameters are learned by a discriminative learning approach. The main contributions of the proposed method are twofold: (1) the method is computationally efficient and scalable since the parameters in kernel are determined in a staged way; (2) the method exploits data distribution and semantic level hidden information by means of LDA. To evaluate the performance of LDA-FEK, we apply it for image retrieval over two data sets and for text categorization on four popular data sets. The results show the competitive performance of our method.