• Journal of the Korean Institute of Intelligent Systems
• /
• v.19 no.5
• /
• pp.648-654
• /
• 2009

In this paper, a method called bilateral diagonal 2DLDA is proposed for face recognition. Two methods called Dia2DPCA and Dia2DLDA were suggested to reserve the correlations between the variations in the rows and columns of diagonal images. However, these methods work in the row direction of these images. A row-directional projection matrix can be obtained by calculating the between-class and within-class covariance matrices making an allowance for the column variation of alternative diagonal face images. In addition, column-directional projection matrix can be obtained by calculating the between-class and within-class covariance matrices making an allowance for the row variation in diagonal images. A bilateral projection scheme was applied using left and right multiplying projection matrices. As a result, the dimension of the feature matrix and computation time can be reduced. Experiments carried out on an ORL face database show that the proposed method with three different distance measures, namely, Frobenius, Yang and AMD, is more accurate than some methods, such as 2DPCA, B2DPCA, 2DLDA, etc.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.15 no.5
• /
• pp.649-654
• /
• 2005

In this paper, we describe an implementation of real-time face recognition system under ubiquitous computing environments. First, face image is captured by PDA with CMOS camera and then this image with user n and name is transmitted via WLAN(Wireless LAN) to the server and finally PDA receives verification result from the server The proposed system consists of server and client parts. Server uses PCA and LDA algorithm which calculates eigenvector and eigenvalue matrices using the face images from the PDA at enrollment process. And then, it sends recognition result using Euclidean distance at verification process. Here, captured image is first compressed by the wave- let transform and sent as JPG format for real-time processing. Implemented system makes an improvement of the speed and performance by comparing Euclidean distance with previously calculated eigenvector and eignevalue matrices in the learning process.

• Food Science of Animal Resources
• /
• v.41 no.6
• /
• pp.936-949
• /
• 2021

This study discriminated fatty acid profile and flavor characteristics of Beijing You Chicken (BYC) as a precious local breed and Dwarf Beijing You Chicken (DBYC) eggs. Fatty acid profile and flavor characteristics were analyzed to identify differences between BYC and DBYC eggs. Four classification algorithms were used to build classification models. Arachidic acid, oleic acid (OA), eicosatrienoic acid, docosapentaenoic acid (DPA), hexadecenoic acid, monounsaturated fatty acids (MUFA), polyunsaturated fatty acids (PUFA), unsaturated fatty acids (UFA) and 35 volatile compounds had significant differences in fatty acids and volatile compounds by gas chromatography-mass spectrometry (GC-MS) (p<0.05). For fatty acid data, k-nearest neighbor (KNN) and support vector machine (SVM) got 91.7% classification accuracy. SPME-GC-MS data failed in classification models. For electronic nose data, classification accuracy of KNN, linear discriminant analysis (LDA), SVM and decision tree was all 100%. The overall results indicated that BYC and DBYC eggs could be discriminated based on electronic nose with suitable classification algorithms. This research compared the differentiation of the fatty acid profile and volatile compounds of various egg yolks. The results could be applied to evaluate egg nutrition and distinguish avian eggs.

• Journal of Positioning, Navigation, and Timing
• /
• v.12 no.2
• /
• pp.91-100
• /
• 2023

The existing indoor localization method using Wi-Fi fingerprinting has a high collection cost and relatively low accuracy, thus requiring integrated correction of convergence with other technologies. This paper proposes a new method that significantly reduces collection costs compared to existing methods using Wi-Fi fingerprinting. Furthermore, it does not require labeling of data at collection and can estimate pedestrian travel paths even in large indoor spaces. The proposed pedestrian movement path estimation process is as follows. Data collection is accomplished by setting up a feature area near an indoor space intersection, moving through the set feature areas, and then collecting data without labels. The collected data are processed using Kernel Linear Discriminant Analysis (KLDA) and the valley point of the Euclidean distance value between two data is obtained within the feature space of the data. We build learning data by labeling data corresponding to valley points and some nearby data by feature area numbers, and labeling data between valley points and other valley points as path data between each corresponding feature area. Finally, for testing, data are collected randomly through indoor space, KLDA is applied as previous data to build test data, the K-Nearest Neighbor (K-NN) algorithm is applied, and the path of movement of test data is estimated by applying a correction algorithm to estimate only routes that can be reached from the most recently estimated location. The estimation results verified the accuracy by comparing the true paths in indoor space with those estimated by the proposed method and achieved approximately 90.8% and 81.4% accuracy in two experimental spaces, respectively.

• Journal of the Korea Institute of Information Security & Cryptology
• /
• v.29 no.3
• /
• pp.491-501
• /
• 2019

In side-channel analysis, which exploit physical leakage from a cryptographic device, deep learning based attack has been significantly interested in recent years. However, most of the state-of-the-art methods have been focused on classifying side-channel information in a profiled scenario where attackers can obtain label of training data. In this paper, we propose a new method based on deep learning to improve non-profiling side-channel attack such as Differential Power Analysis and Correlation Power Analysis. The proposed method is a signal preprocessing technique that reduces the noise in a trace by modifying Auto-Encoder framework to the context of side-channel analysis. Previous work on Denoising Auto-Encoder was trained through randomly added noise by an attacker. In this paper, the proposed model trains Auto-Encoder through the noise from real data using the noise-reduced-label. Also, the proposed method permits to perform non-profiled attack by training only a single neural network. We validate the performance of the noise reduction of the proposed method on real traces collected from ChipWhisperer board. We demonstrate that the proposed method outperforms classic preprocessing methods such as Principal Component Analysis and Linear Discriminant Analysis.

• Journal of Wetlands Research
• /
• v.25 no.4
• /
• pp.306-314
• /
• 2023

Stream sediments are an important component of water quality management because they are receptors of various pollutants such as heavy metals and organic matters emitted from upland sources and can be secondary pollution sources, adversely affecting water environment. To effectively manage the stream sediments, identification of primary sources of sediment contamination and source-associated control strategies will be required. We evaluated the performance of machine learning models in identifying primary sources of sediment contamination based on the physico-chemical properties of stream sediments. A total of 356 stream sediment data sets of 18 quality parameters including 10 heavy metal species(Cd, Cu, Pb, Ni, As, Zn, Cr, Hg, Li, and Al), 3 soil parameters(clay, silt, and sand fractions), and 5 water quality parameters(water content, loss on ignition, total organic carbon, total nitrogen, and total phosphorous) were collected near abandoned metal mines and industrial complexes across the four major river basins in Korea. Two machine learning algorithms, linear discriminant analysis (LDA) and support vector machine (SVM) classifiers were used to classify the sediments into four cases of different combinations of the sampling period and locations (i.e., mine in dry season, mine in wet season, industrial complex in dry season, and　industrial complex in wet season). Both models showed good performance in the classification, with SVM outperformed LDA; the accuracy values of LDA and SVM were 79.5% and 88.1%, respectively. An SVM ensemble model was used for multi-label classification of the multiple contamination sources inlcuding landuses in the upland areas within 1 km radius from the sampling sites. The results showed that the multi-label classifier was comparable performance with sinlgle-label SVM in classifying mines and industrial complexes, but was less accurate in classifying dominant land uses (50~60%). The poor performance of the multi-label SVM is likely due to the overfitting caused by small data sets compared to the complexity of the model. A larger data set might increase the performance of the machine learning models in identifying contamination sources.

• Journal of Life Science
• /
• v.32 no.10
• /
• pp.803-811
• /
• 2022

Despite the importance of traditional Korean fermented foods, little is known about the microbial communities and diversity of fermented soy products. To gain insight into the unexplored microbial communities of both Doenjang (DJ) and Ganjang (GJ) that may contribute to the fermentation in Korean traditional foods, we carried out next-generation sequencing (NGS) based on the V3-V4 region of 16S rDNA gene analysis. The alpha diversity analysis results revealed that both the Shannon and Simpson diversity indices were significantly different between the two groups, whereas the richness indices, including ACE, CHAO, and Jackknife, were not significant. Firmicutes were the most dominant phylum in both groups, but several taxa were found to be more abundant in DJ than in GJ. The proportions of Bacillus, Kroppenstedtia, Clostridium, and Pseudomonas and most halophiles and halotolerant bacteria, such as Tetragenococcus, Chromohalobacter, Lentibacillus, and Psychrobacter, were lower in DJ than in GJ. Linear discriminant effect size (LEfSe) analysis was carried out to discover discriminative functional biomarkers. Biomarker discovery results showed that Bacillus and Tetragenococcus were identified as the most important features for the classification of subjects to DJ and GJ. Paired-permutational multivariate analysis of variance (PERMANOVA) further revealed that the bacterial community structure between the two groups was statistically different (p=0.001).

• Journal of Life Science
• /
• v.33 no.1
• /
• pp.25-33
• /
• 2023

The fermentation process of kimchi, which is a traditional Korean food, influences the resulting compo- sition of microorganisms, such as the genera Leuconostoc, Weissella, and Lactobacillus. In addition, several factors, including the type of kimchi, fermentation conditions, materials, and ingredients, can influence the distribution of the kimchi microbial community. In this study, next-generation sequencing (NGS) of kimchi samples obtained from central (Gangwon-do and Gyeonggi-do) and southern (Jeolla-do and Gyeongsang-do) regions in Korea was performed, and the microbial communities in samples from the two regions were compared. Good's coverage prediction for all samples was higher than 99%, indicating that there was sufficient reliability for comparative analysis. However, in a α -diversity analysis, there was no significant difference in species richness and diversity between samples. The Firmicutes phylum was common in both regions. At the species level, Weissella kandleri dominated in central (46.5%) and southern (30.8%) regions. Linear discriminant analysis effect size (LEfSe) analysis was performed to identify biomarkers representing the microbial community in each region. The LEfSe results pointed to statistically significant differences between the two regions in community composition, with Leuconostocaceae (71.4%) dominating in the central region and Lactobacillaceae (61.0%) dominating in the southern region. Based on these results, it can be concluded that the microbial communities of kimchi are significantly influenced by regional properties and that it can provide more useful scientific data to study the relationship between regional characteristics of kimchi and their microbial distribution.

• Journal of Korean Medicine for Obesity Research
• /
• v.19 no.2
• /
• pp.97-105
• /
• 2019

Objectives: We investigated differences between the tracheostomized and the non-tracheostomized stroke patients through microbiological analysis for the purpose of preliminary explorations of full-scale clinical research in the future. Methods: We collected tracheal aspirates samples from 5 stroke patients with tracheostomy and expectorated sputum samples from 5 stroke patients without tracheostomy. Genomic DNA from sputum samples was isolated using QIAamp DNA mini kit. The sequences were processed using Quantitative Insights into Microbial Ecology 1.9.0. Alpha-diversity was calculated using the Chao1 estimator. Beta-diversity was analyzed by UniFrac-based principal coordinates analysis (PCoA). To confirm taxa with different abundance among the groups, linear discriminant analysis effect size analysis was performed. Results: Although alpha-diversity value of the tracheostomized group was higher than that of the non-tracheostomized group, there was no statistically significant difference. In PCoA, clear separation was seen between clusters of the tracheostomized group and that of the non-tracheostomized group. In both groups, Bacteroidetes, Proteobacteria, Fusobacteria, Firmicutes, Actinobacteria were identified as dominant in phylum level. In particular, relative richness of Proteobacteria was found to be 31% more in the tracheotomized group (36.6%) than the non-tracheostomized group (5.6%)(P<0.05). In genus level, Neisseria (24%), Prevotella (17%), Streptococcus (13%), Fusobacteria (11%), Porphyromonas (7%) were identified as dominant in the tracheostomized group. In the non-tracheostomized group, Prevotella (38%), Veillonella (20%), Neisseria (9%) were genera that found to be dominant. Conclusions: It is meaningful in that the tracheostomized group has been identified a higher rate of microbiotas known as pathogenic in respiratory diseases compared to the non-tracheostomized group, confirming the possibility that the risk of opportunity infection may be higher.

• Journal of Periodontal and Implant Science
• /
• v.53 no.3
• /
• pp.233-244
• /
• 2023

Purpose: An implant-supported prosthesis consists of an implant fixture, an abutment, an internal screw that connects the abutment to the implant fixture, and the upper prosthesis. Numerous studies have investigated the microorganisms present on the implant surface, surrounding tissues, and the subgingival microflora associated with peri-implantitis. However, there is limited information regarding the microbiome within the internal screw space. In this study, microbial samples were collected from the supragingival surfaces of natural teeth, the peri-implant sulcus, and the implant-abutment screw hole, in order to characterize the microbiome of the internal screw space in healthy subjects. Methods: Samples were obtained from the supragingival region of natural teeth, the peri-implant sulcus, and the implant screw hole in 20 healthy subjects. DNA was extracted, and the V3-V4 region of the 16S ribosomal RNA was sequenced for microbiome analysis. Alpha diversity, beta diversity, linear discriminant analysis effect size (LEfSe), and network analysis were employed to compare the characteristics of the microbiomes. Results: We observed significant differences in beta diversity among the samples. Upon analyzing the significant taxa using LEfSe, the microbial composition of the implant-abutment screw hole's microbiome was found to be similar to that of the other sampling sites' microbiomes. Moreover, the microbiome network analysis revealed a unique network complexity in samples obtained from the implant screw hole compared to those from the other sampling sites. Conclusions: The bacterial composition of the biofilm collected from the implant-abutment screw hole exhibited significant differences compared to the supra-structure of the implant. Therefore, long-term monitoring and management of not only the peri-implant tissue but also the implant screw are necessary.