• ETRI Journal
• /
• v.32 no.5
• /
• pp.784-794
• /
• 2010

Automatic facial expression recognition has many potential applications in different areas of human computer interaction. However, they are not yet fully realized due to the lack of an effective facial feature descriptor. In this paper, we present a new appearance-based feature descriptor, the local directional pattern (LDP), to represent facial geometry and analyze its performance in expression recognition. An LDP feature is obtained by computing the edge response values in 8 directions at each pixel and encoding them into an 8 bit binary number using the relative strength of these edge responses. The LDP descriptor, a distribution of LDP codes within an image or image patch, is used to describe each expression image. The effectiveness of dimensionality reduction techniques, such as principal component analysis and AdaBoost, is also analyzed in terms of computational cost saving and classification accuracy. Two well-known machine learning methods, template matching and support vector machine, are used for classification using the Cohn-Kanade and Japanese female facial expression databases. Better classification accuracy shows the superiority of LDP descriptor against other appearance-based feature descriptors.

• Journal of the Korean Society of Clothing and Textiles
• /
• v.41 no.1
• /
• pp.98-115
• /
• 2017

This study categorizes the formative aspects of dress and their implications according to the extent of revealing or concealing corporeality based on body perceptions. By considering the notion of dress as bodily practice to be a theoretical and methodological framework, this study combines a literature survey and case analysis to analyze and classify the forms of women's dress since the 1920s when contemporary fashion took hold. As examined in this study, the typology of dress was categorized as body-consciousness, deformation, transformation, and formlessness. Body-consciousness that is achieved through tailoring, bias cutting, and stretchy fabric displays corporeality focusing on the structure and function of the body as an internalized corset. Deformations in dress are categorized into two different subcategories. One is the expansion or reduction of bodily features based on the vertical or horizontal grids of the body, which visualizes the anachronistic restraint of the body through an innerwear as outerwear strategy. The other is exaggerations of the bodily features irrelevant to the grid, which break from the limitations and constraints of the body as well as traditional notions of the body. Transformations of the body refer to as follows. First, the deconstruction and restructuring of the body that deconstruct the stereotypes in garment construction. Second, the abstraction of the body that emphasizes the geometrical and architectural shapes. Third, transformable designs which pursue the expansion and multiplicity of function. Formlessness in dress denies the perception of three-dimensionality of the body through the planarization of the body.

• Computers and Concrete
• /
• v.10 no.6
• /
• pp.557-573
• /
• 2012

This paper discusses the combined application of two different techniques, Neural Networks (NN) and Principal Component Analysis (PCA), for improved prediction of concrete properties. The combination of these approaches allowed the development of six neural networks models for predicting slump and compressive strength of concrete with mineral additives such as blast furnace slag, fly ash and silica fume. The Back-Propagation Multi-Layer Perceptron (BPMLP) with Bayesian regularization was used in all these models. They are produced to implement the complex nonlinear relationship between the inputs and the output of the network. They are also established through the incorporation of a huge experimental database on concrete organized in the form Mix-Property. Thus, the data comprising the concrete mixtures are much correlated to each others. The PCA is proposed for the compression and the elimination of the correlation between these data. After applying the PCA, the uncorrelated data were used to train the six models. The predictive results of these models were compared with the actual experimental trials. The results showed that the elimination of the correlation between the input parameters using PCA improved the predictive generalisation performance models with smaller architectures and dimensionality reduction. This study showed also that using the developed models for numerical investigations on the parameters affecting the properties of concrete is promising.

• Environmental Engineering Research
• /
• v.22 no.3
• /
• pp.302-311
• /
• 2017

Nowadays, with the burgeoning development of economy, $CO_2$ emissions increase rapidly in China. It has become a common concern to seek effective methods to forecast $CO_2$ emissions and put forward the targeted reduction measures. This paper proposes a novel hybrid model combined principal component analysis (PCA) with regularized extreme learning machine (RELM) to make $CO_2$ emissions prediction based on the data from 1978 to 2014 in China. First eleven variables are selected on the basis of Pearson coefficient test. Partial autocorrelation function (PACF) is utilized to determine the lag phases of historical $CO_2$ emissions so as to improve the rationality of input selection. Then PCA is employed to reduce the dimensionality of the influential factors. Finally RELM is applied to forecast $CO_2$ emissions. According to the modeling results, the proposed model outperforms a single RELM model, extreme learning machine (ELM), back propagation neural network (BPNN), GM(1,1) and Logistic model in terms of errors. Moreover, it can be clearly seen that ELM-based approaches save more computing time than BPNN. Therefore the developed model is a promising technique in terms of forecasting accuracy and computing efficiency for $CO_2$ emission prediction.

• Journal of Korea Multimedia Society
• /
• v.17 no.1
• /
• pp.34-42
• /
• 2014

In this paper an embedding algorithm based on commute time is implemented by organizing patches according to the graph-based metric, and its performance is analyzed by comparing with the results of principal component analysis embedding. It is usual that the dimensionality reduction be done within some acceptable approximation error. However this paper shows the proposed manifold embedding method generates the intrinsic geometry corresponding to the signal despite severe approximation error, so that it can be applied to the areas such as pattern classification or machine learning.

• Journal of the Korea Computer Graphics Society
• /
• v.24 no.2
• /
• pp.29-40
• /
• 2018

Direct volume rendering (DVR) is a commonly used method to visualize inner structures in 3D volumetric datasets. However, conventional volume rendering on a 2D display lacks depth perception due to dimensionality reduction caused by ray casting. In this work, we investigate how emerging Virtual Reality (VR) can improve the usability of direct volume rendering. We developed real-time high-resolution DVR system in virtual reality, and measures the usefulness of volume rendering with improved depth perception via a user study conducted by 38 participants. The result indicates that virtual reality significantly improves the usability of DVR by allowing better depth perception.

• Proceedings of the Korean Society of Broadcast Engineers Conference
• /
• 2008.02a
• /
• pp.179-182
• /
• 2008

본 논문은 오디오 부호화 및 복호화 과정에서, 주파수 영역에서 표현된 오디오 신호를 차원 축소 방법으로 압축하여 표현함으로서 오디오 부호화 효율을 증대시키고자 하는 방식에 관한 것이다. 차원 축소는 행렬을 특정한 조건을 바탕으로 두 개의 행렬의 곱으로 표현하는 방식으로, 특정 행렬로 표현된 데이터를 좀 더 작은 데이터량으로 표현하는 것뿐만 아니라 이 과정에서 데이터에 내재되어 있는 추상적인 정보까지도 함축적으로 얻어낼 수 있기 때문에, 일반적으로 데이터의 압축에 좋은 성능을 보인다. 주파수 영역으로 변환된 신호는 일반적으로 (주파수 밴드의 개수) $\times$ (전체 프레임의 개수)인 행렬로 볼 수 있으며, 이 전체 행렬을 입력으로 간주하고, 차원 축소를 수행하여 신호의 압축 효과를 얻을 수 있다. 그러나 이 경우, 행렬 전체를 입력 신호로 보아야 하기 때문에 실시간 부호화가 불가능하며, 신호 전체 길이만큼의 부호화 지연이 발생한다. 이를 해소하기 위해, 본 논문에서는 특정 개수만큼의 프레임을 묶어서 여러 번의 차원 축소를 순차적으로 수행함으로써 부호화 지연을 최소화하는 방식을 제안한다.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.11 no.7
• /
• pp.3594-3607
• /
• 2017

The traditional feature extraction methods such as principal component analysis (PCA) cannot obtain the local structure of the samples, and locally linear embedding (LLE) cannot obtain the global structure of the samples. However, a common drawback of existing PCA and LLE algorithm is that they cannot deal well with the sparse problem of the samples. Therefore, by integrating the globality of PCA and the locality of LLE with a sparse constraint, we developed an improved and unsupervised difference algorithm called Sparse Difference Embedding (SDE), for dimensionality reduction of high-dimensional data in small sample size problems. Significantly differing from the existing PCA and LLE algorithms, SDE seeks to find a set of perfect projections that can not only impact the locality of intraclass and maximize the globality of interclass, but can also simultaneously use the Lasso regression to obtain a sparse transformation matrix. This characteristic makes SDE more intuitive and more powerful than PCA and LLE. At last, the proposed algorithm was estimated through experiments using the Yale and AR face image databases and the USPS handwriting digital databases. The experimental results show that SDE outperforms PCA LLE and UDP attributed to its sparse discriminating characteristics, which also indicates that the SDE is an effective method for face recognition.

• Communications for Statistical Applications and Methods
• /
• v.16 no.4
• /
• pp.579-586
• /
• 2009

We have used multifactor dimensionality reduction(MDR) method to study interaction effect of statistical model in general. But MDR method cannot be applied in all cases. It can be applied to the only case-control data. So, two methods are suggested E-MDR and D-MDR method using regression tree algorithm and dummy variables. We applied the methods on the identify interaction effects of single nucleotide polymorphisms(SNPs) responsible for longissimus mulcle dorsi area(LMA), carcass cold weight(CWT) and average daily gain(ADG) in a Hanwoo beef cattle population. Finally, we compare the results using permutation test.

• Journal of the Korean Institute of Telematics and Electronics C
• /
• v.35C no.6
• /
• pp.88-98
• /
• 1998

Subspace classifier is a popular pattern recognition method based on Karhunen-Loeve transform. This classifier describes a high dimensional pattern by using a reduced dimensional subspace. Because of the loss of information induced by dimensionality reduction, however, a subspace classifier sometimes shows unsatisfactory recognition performance to the patterns having quite similar principal components each other. In this paper, we propose the use of multiple novelty neural network classifiers constructed on novelty vectors to adopt minor components usually ignored and present a method of improving recognition performance through combining those with the subspace classifier. We develop the proposed classifier on handwritten numeral database and analyze its properties. Our proposed classifier shows better recognition performance compared with other classifiers, though it requires more weight links.