• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.37 no.8A
• /
• pp.639-647
• /
• 2012

Vision-based on-road vehicle detection is one of the key techniques in automotive driver assistance systems. However, due to the huge within-class variability in vehicle appearance and environmental changes, it remains a challenging task to develop an accurate and reliable detection system. In general, a vehicle detection system consists of two steps. The candidate locations of vehicles are found in the Hypothesis Generation (HG) step, and the detected locations in the HG step are verified in the Hypothesis Verification (HV) step. Since the final decision is made in the HV step, the HV step is crucial for accurate detection. In this paper, we propose using a reduced multivariate polynomial pattern classifier (RM) for the HV step. Our experimental results show that the RM classifier outperforms the well-known Support Vector Machine (SVM) classifier, particularly in terms of the fast decision speed, which is suitable for real-time implementation.

• Korean Journal of Optics and Photonics
• /
• v.31 no.3
• /
• pp.125-133
• /
• 2020

An innovative method for separating overlapping latent fingerprints, using laser-induced plasma spectroscopy (LIPS) combined with multivariate analysis, is reported in the current study. LIPS provides the capabilities of real-time analysis and high-speed scanning, as well as data regarding the chemical components of overlapping fingerprints. These spectra provide valuable chemical information for the forensic classification and reconstruction of overlapping latent fingerprints, by applying appropriate multivariate analysis. This study utilizes principal-component analysis (PCA) and partial-least-squares (PLS) techniques for the basis classification of four types of fingerprints from the LIPS spectra. The proposed method is successfully demonstrated through a classification example of four distinct latent fingerprints, using discrimination such as soft independent modeling of class analogy (SIMCA) and partial-least-squares discriminant analysis (PLS-DA). This demonstration develops an accuracy of more than 85% and is proven to be sufficiently robust. In addition, by laser-scanning analysis at a spatial interval of 125 ㎛, the overlapping fingerprints were separated as two-dimensional forms.

• Current Optics and Photonics
• /
• v.1 no.4
• /
• pp.412-420
• /
• 2017

Multinucleated bone resorptive osteoclasts differentiate from bone marrow-derived monocyte/macrophage precursor cells. During osteoclast differentiation, mononuclear pre-osteoclasts change their morphology and biochemical characteristics. In this study, Raman spectroscopy with multivariate techniques such as Principal Component Analysis (PCA) and Linear Discriminant Analysis (LDA) were used to extract biochemical information related to various cellular events during osteoclastogenesis. This technique allowed for label-free and noninvasive monitoring of differentiating cells, and clearly discriminated four different time points during osteoclast differentiation. The Raman band intensity showed significant time-dependent changes that increased up to day 4. The results of Raman spectroscopy agreed with results from atomic force microscopy (AFM) and tartrate-resistant acid phosphatase (TRAP) staining, a conventional biological assay. Under AFM, normal spindle-like mononuclear pre-osteoclasts became round and smaller at day 2 after treatment with a receptor activator of nuclear $factor-{\kappa}B$ ligand and they formed multinucleated giant cells at day 4. Thus, Raman spectroscopy, in combination with PCA-LDA, may be useful for noninvasive label-free quality assessment of cell status during osteoclast differentiation, enabling more efficient optimization of the bioprocesses.

• Restorative Dentistry and Endodontics
• /
• v.38 no.3
• /
• pp.154-159
• /
• 2013

Objectives: The aim of this study was to compare the cleaning efficacy of different final irrigation regimens in canal and isthmus of mandibular molars, and to evaluate the influence of related variables on cleaning efficacy of the irrigation systems. Materials and Methods: Mesial root canals from 60 mandibular molars were prepared and divided into 4 experimental groups according to the final irrigation technique: Group C, syringe irrigation; Group U, ultrasonics activation; Group SC, VPro StreamClean irrigation; Group EV, EndoVac irrigation. Cross-sections at 1, 3 and 5 mm levels from the apex were examined to calculate remaining debris area in the canal and isthmus spaces. Statistical analysis was completed by using Kruskal-Wallis test and Mann-Whitney U test for comparison among groups, and multivariate linear analysis to identify the significant variables (regular replenishment of irrigant, vapor lock management, and ultrasonic activation of irrigant) affecting the cleaning efficacy of the experimental groups. Results: Group SC and EV showed significantly higher canal cleanliness values than group C and U at 1 mm level (p < 0.05), and higher isthmus cleanliness values than group U at 3 mm and all levels of group C (p < 0.05). Multivariate linear regression analysis demonstrated that all variables had independent positive correlation at 1 mm level of canal and at all levels of isthmus with statistical significances. Conclusions: Both VPro StreamClean and EndoVac system showed favorable result as final irrigation regimens for cleaning debris in the complicated root canal system having curved canal and/or isthmus. The debridement of the isthmi significantly depends on the variables rather than the canals.

• Journal of Korean Society on Water Environment
• /
• v.35 no.1
• /
• pp.19-27
• /
• 2019

The purpose of this study is to evaluate the water quality and phytoplankton community in Bukhan River which account for 44.4 % of the total inflow into Lake Paldang, using multivariate statistical techniques (i.e., correlation analysis, principal component analysis (PCA)/factor analysis (FA)). Water samples were collected from March to November 2015 and the following parameters measured; water temperature, pH, DO, EC, SS, BOD, Chl-a, COD, TN, $NO_3-N$, $NH_3-N$, TP, DTP, $PO_4-P$, and phytoplankton community. The water quality of the main stream and the tributaries were not significantly different apart from the relatively high concentration of BOD, COD and nutrients recorded in MH. The highest cell density of Stephanodiscus hantzschii and Merismopedia glauca dominated phytoplankton was observed in PD. Based on the correlation analysis, total phytoplankton and cyanophyceae were highly correlated with BOD, COD and nutrients. PCA/FA resulted in four main factors accounting for 82.240 % of the total variance in the water quality dataset. The group of component 1 (TN, DTN, DO, $NO_3-N$, water temperature) and component 2 ($PO_4-P$, T-P, DTP, SS) were classified as nutrient element factor whereas component 3 (Chl-a, COD, BOD, $NH_3-N$, pH) was related to organic substances. Hence, the identification of the main potential environmental pollution factors in Bukhan River will help policy makers make better and more informed decisions on how to improve the water quality.

• Journal of Korean Society for Quality Management
• /
• v.50 no.4
• /
• pp.701-716
• /
• 2022

Purpose: This study aims to predict the dry cargo transportation market economy. The subject of this study is the BDI (Baltic Dry Index) time-series, an index representing the dry cargo transport market. Methods: In order to increase the accuracy of the BDI time-series, we have pre-processed the original time-series via time-series decomposition and data augmentation techniques and have used them for ANN learning. The ANN algorithms used are Multi-Layer Perceptron (MLP), Recurrent Neural Network (RNN), and Long Short-Term Memory (LSTM) to compare and analyze the case of learning and predicting by applying time-series decomposition and data augmentation techniques. The forecast period aims to make short-term predictions at the time of t+1. The period to be studied is from '22. 01. 07 to '22. 08. 26. Results: Only for the case of the MAPE (Mean Absolute Percentage Error) indicator, all ANN models used in the research has resulted in higher accuracy (1.422% on average) in multivariate prediction. Although it is not a remarkable improvement in prediction accuracy compared to uni-variate prediction results, it can be said that the improvement in ANN prediction performance has been achieved by utilizing time-series decomposition and data augmentation techniques that were significant and targeted throughout this study. Conclusion: Nevertheless, due to the nature of ANN, additional performance improvements can be expected according to the adjustment of the hyper-parameter. Therefore, it is necessary to try various applications of multiple learning algorithms and ANN optimization techniques. Such an approach would help solve problems with a small number of available data, such as the rapidly changing business environment or the current shipping market.

• Journal of Information Technology Application
• /
• v.1
• /
• pp.173-211
• /
• 1999

The task of classification permeates all walks of life, from business and economics to science and public policy. In this context, nonlinear techniques from artificial intelligence have often proven to be more effective than the methods of classical statistics. The objective of knowledge discovery and data mining is to support decision making through the effective use of information. The automated approach to knowledge discovery is especially useful when dealing with large data sets or complex relationships. For many applications, automated software may find subtle patterns which escape the notice of manual analysis, or whose complexity exceeds the cognitive capabilities of humans. This paper explores the utility of a collaborative learning approach involving integrated models in the preprocessing and postprocessing stages. For instance, a genetic algorithm effects feature-weight optimization in a preprocessing module. Moreover, an inductive tree, artificial neural network (ANN), and k-nearest neighbor (kNN) techniques serve as postprocessing modules. More specifically, the postprocessors act as second0order classifiers which determine the best first-order classifier on a case-by-case basis. In addition to the second-order models, a voting scheme is investigated as a simple, but efficient, postprocessing model. The first-order models consist of statistical and machine learning models such as logistic regression (logit), multivariate discriminant analysis (MDA), ANN, and kNN. The genetic algorithm, inductive decision tree, and voting scheme act as kernel modules for collaborative learning. These ideas are explored against the background of a practical application relating to financial fraud management which exemplifies a binary classification problem.

• Geomechanics and Engineering
• /
• v.14 no.3
• /
• pp.257-269
• /
• 2018

Analysis of slope stability failures, as one of the complex natural hazards, is one of the important research issues in the field of civil engineering. Present paper adopts and investigates four soft computing-based techniques for this problem: Patient Rule-Induction Method (PRIM), M5' algorithm, Group Method of data Handling (GMDH) and Multivariate Adaptive Regression Splines (MARS). A comprehensive database consisting of 168 case histories is used to calibrate and test the developed models. Six predictive variables including slope height, slope angle, bulk density, cohesion, angle of internal friction, and pore water pressure ratio were considered to generate new models. The results of test studies are used for feasibility, effectiveness and practicality comparison of techniques with each other, and with the other available well-known methods in the literature. Results show that all methods not only are feasible but also result in better performance than previously developed soft computing based predictive models and tools. It is shown that M5' and PRIM algorithms are the most effective and practical prediction models.

• The Korean Journal of Applied Statistics
• /
• v.21 no.6
• /
• pp.981-996
• /
• 2008

Statistical matching techniques whose aim is to achieve a complete data file from different sources. Since the statistical matching method proposed by Rubin (1986) assumes the multivariate normality for data, using this method to data which violates the assumption would involve some problems. This research proposed the statistical matching method using robust regression as an alternative to the linear regression. Furthermore, we carried out a simulation study to compare the performance of the robust regression model and the linear regression model for the statistical matching.

• Membrane and Water Treatment
• /
• v.13 no.5
• /
• pp.245-257
• /
• 2022

Water quality characteristics and their spatial variations in the Nakdong River were statistically analyzed by multivariate techniques including correlation analysis, CA, and FA/PCA based on water quality parameters for 17 sites over 2017-2019, yielding PI values for primary factors. Site 10 indicated the highest parameter concentrations, and results of pearson's correlation analysis suggest that non-biodegradable organic matter had been distributed on the site. Five clusters were identified in order of descending pollution levels: I (Ib > Ia) > II (IIa > IIb) > III. Spatial variations started from sub-cluster Ib in which Daegu city and Geumho-river are joined. T-P, PO4-P, SS, COD, and TOC corresponded to VF 1 and 2, which were found to be principal components with strong influence on water quality. Sub-cluster Ib was strongly influenced by NO3-N and T-N compared to other clusters. According to the PIs, water quality pollution deteriorated due to non-biodegradable organic matter, nitrogen- and phosphorus-based nutrient salts in the middle and lower reaches, illustrating worsening water pollution due to inflows of anthropogenic sources on the Geumho-river, i.e., sewage and wastewater, discharged from Site 10, at which there is a concentration of urban, agricultural, and industrial areas.