• Journal of Biosystems Engineering
• /
• v.29 no.4
• /
• pp.317-322
• /
• 2004

This study was conducted to evaluate the feasibility of identifying freshness of beef using a surface acoustic wave (SAW) sensor based electronic nose. The beef was stored at 5$^{\circ}C$ and aroma was measured with the passage of time. Chromatographic analysis of the odor showed that number of volatile components and their amounts were rapidly increased after 19 days of storage. Classifying beefs according to their storage days was possible using principle component analysis (PCA). Classifying beefs processed from four different origins was also possible with PCA analysis of odor. This study shows that electronic nose can be applied to beef freshness evaluation and classification of its origin.

• Journal of the Institute of Electronics Engineers of Korea SP
• /
• v.41 no.6
• /
• pp.43-49
• /
• 2004

The classification time required by conventional multi-class SVMs(Support Vector Machines) greatly increases as the number of pattern classes increases. This is due to the fact that the needed set of binary class SVMs gets quite large. In this paper, we propose a method to reduce the number of classes by using nearest neighbor rule (NNR) in the principle component analysis and linear discriminant analysis (PCA+LDA) feature subspace. The proposed method reduces the number of face classes by selecting a few classes closest to the test data projected in the PCA+LDA feature subspace. Results of experiment show that our proposed method has a lower error rate than nearest neighbor classification (NNC) method. Though our error rate is comparable to the conventional multi-class SVMs, the classification process of our method is much faster.

• Korean Journal of Plant Taxonomy
• /
• v.41 no.3
• /
• pp.223-229
• /
• 2011

The taxonomy of the Isodon excisus complex has been ambiguous and problematic because the morphological characters, especially characters related to the leaf distinguishing subgroups of the complex in the original descriptions, are variable. To elucidate the taxonomic structure of the I. excisus complex in Korea, 34 characters were measured from 70 OTUs representing different locations and analyzed by principal component analysis (PCA). The analysis showed that principle component axis 1, 2, 3 (PC1, PC2, PC3) represents 52.0% of the total variance and characters showing high loading values for PC1 were leaf shape, density of non-glandular hairs on the lower surface of the leaf, and characters related to the teeth of the leaf. The length of apical tooth and the angle between two widest points of the leaf were highly correlated to PC2 and PC3, respectively. Three-dimensional scatter plotting of OTUs for PC1, PC2, and PC3 axis showed that the areas of previously recognized three subgroups of I. excisus completely overlapped. Our result supported that just one taxon, I. excisus var. excisus, should be recognized in the complex at the variety level.

• Food Science of Animal Resources
• /
• v.30 no.2
• /
• pp.185-189
• /
• 2010

An electronic nose was used to identify Salmonella contamination on beef based on odors. To detect pathogen contamination of beef, $100{\mu}L$ of $10^5CFU/g$ Salmonella Enteritidis or Salmonella Typhimurium cell suspensions were spiked onto 5 g beef sirloin samples in individual vials. Odor changes over time were then measured and analyzed using an electronic nose system to identify pathogen contamination. In principle, the electronic nose system based on a surface acoustic wave (SAW) detector produced different frequency responses depending on the time and amount of each chemical. Multivariate analysis of the odor data was conducted to detect Salmonella contamination of beef. Salmonella odors were successfully distinguished from uncontaminated beef odors by principal component analysis (PCA). The PCA results showed that Salmonella contamination of beef could be detected after 4 h of incubation. The numbers of cells enumerated by standard plate count after 4 h of inoculation were $2{\times}10^6CFU/g$ for both Salmonella Enteritidis and Salmonella Typhimurium.

• Proceedings of the KSPS conference
• /
• 2007.05a
• /
• pp.319-322
• /
• 2007

The goal of our research is to build a textindependent speaker identification system that can be used in mobile devices without any additional adaptation process. In this paper, we show that exploiting the advantages of both PCA(Principle Component Analysis) and LDA(Linear Discriminant Analysis) can increase the performance in the situation. The proposed method reduced the relative recognition error by 13.5%

• Proceedings of the KIEE Conference
• /
• 2005.07e
• /
• pp.94-96
• /
• 2005

Power Quality(PQ) problems are various owing to a wide variety of causes so detection and classification of many kinds of PQ problems are awkward. Almost all studies about it were about getting good results by Neural Networks(NN) which get input features from as random variables, FFT and wavelet transform. However they are discontented with results because it is very difficult to classify all PQ items. A study about feature extraction becomes needed. Thus, this paper suggests effective way of using principle Component Analysis (PCA) for PQ Problem classification. PCA found more effective features among all features so it will help us to get more good result of classification.

• Proceedings of the Korean Institute of Electrical and Electronic Material Engineers Conference
• /
• 2005.07a
• /
• pp.566-567
• /
• 2005

Until now variable pattern classification methods have been introduced. So, variable methods in PD source classification were applied. NN(neural network) the most used scheme as a PD(partial discharge) source classification. But in recent year another method were developed. These methods is present superior to NN in the field of image and signal process function of classification. In this paper, it is show classification result in PD source using three methods; that is, BP(back-propagation), ANFIS(adaptive neuro-fuzzy inference system), PCA-LDA(principle component analysis-linear discriminant analysis).

• Journal of the Korea Society of Computer and Information
• /
• v.21 no.5
• /
• pp.167-175
• /
• 2016

The volatile component analysis of 82 commercialized liquors(44 samples of single malt whisky, 20 samples of blended whisky and 18 samples of brandy) was carried out by gas chromatography after liquid-liquid extraction with dichloromethane. Pattern recognition techniques such as principle component analysis(PCA), cluster analysis(CA), linear discriminant analysis(LDA) and partial least square discriminant analysis(PLSDA) were applied for the discrimination of different liquor categories. Classification rules were validated by considering sensitivity and specificity of each class. Both techniques, LDA and PLSDA, gave 100% sensitivity and specificity for all of the categories. These results suggested that the common characteristics and identities as typification of whiskies and brandys was founded by using multivariate data analysis method.

• Journal of the Korean Society of Food Culture
• /
• v.37 no.3
• /
• pp.239-247
• /
• 2022

This study evaluated the sensory characteristics of sauerkraut prepared by adding 0.5, 1.0, 1.5, 2.0, and 2.5% (w/w) sea salt to cabbage. The quantitative descriptive analysis (QDA) and acceptance test of sauerkraut were determined for each salt concentration, and the principal component analysis (PCA) and partial least square regression (PLSR) analysis were performed to confirm the correlation between each factor. Results of the QDA determined 14 descriptive terms; furthermore, brightness and yellowness of appearance and the sour, salty, and bitter flavors differed significantly according to the salt concentration. Results from the PCA explained 22.56% PC1 and 65.34% PC2 of the total variation obtained. Sauerkraut prepared using 0.5, 1.0, and 1.5% sea salt had high brightness, moistness, sour odor, green odor, sour flavor, carbonation, hardness, chewiness, and crispness, whereas sauerkraut prepared with 2.0 and 2.5% sea salt had high yellowness, glossiness, salty flavor, sweet flavor, and bitter flavor. Hierarchical cluster analysis classified the products into two clusters: sauerkraut of 0.5, 1.0, and 1.5%, and sauerkraut of 2.0 and 2.5%. Results of PLSR determined that sauerkraut of 1.0 and 1.5% were the closest to texture, taste, and overall acceptance. We, therefore, conclude that sauerkrauts prepared using 1.0 and 1.5% sea salt have excellent characteristics in appearance, taste, and texture.

• Journal of the Korean Society for Aeronautical & Space Sciences
• /
• v.40 no.12
• /
• pp.1040-1047
• /
• 2012

This paper presents a fault diagnosis algorithm of control surfaces of small fixed-wing aircraft to reduce maintenance cost or to improve repair efficiency by estimation of fault occurrence or part replacement periods. The proposed fault diagnosis algorithm consists of ANPSD (Averaged Normalized Power Spectral Density), PCA (Principle Component Analysis), and GC (Geometric Classifier). ANPSD is used for frequency-domain vibration testing. PCA has advantage to extract compressed information from ANPSD. GC has good properties to minimize errors of the fault detection and isolation. The algorithm was verified by the accelerometer measurements of the scaled normal and faulty ailerons and the test results show that the algorithm is suitable for the detection and isolation of the control surface faults. This paper also proposes solutions for some kind of implementation problems.