• Journal of Aerospace System Engineering
• /
• v.6 no.1
• /
• pp.13-18
• /
• 2012

This paper presents a structural health monitoring method of aircraft reciprocating engine using Principal Component Analysis (PCA) which analyzes vibration expressed by Averaged Normalized Power Spectral Density (ANPSD). Because ANPSD of the rotating shaft is sensitive to the rotating speed, this paper proposes to use a post-processing method of ANPSD is used to reduce the sensitivity. The PCA extracts compressed information from the post-processed ANPSDs and the information means the difference between current and normal cases of the engine. The experimental results demonstrate the feasibility and effectiveness of the proposed method to detect abnormal cases of the engine.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2019.10a
• /
• pp.145-148
• /
• 2019

Principal Component Analysis (PCA) is a powerful technique in data analysis and widely used to detect anomalies in Wireless Sensor Networks. However, the performance of conventional PCA is not high on time-series data collected by sensors. In this paper, we propose a Joint Exponential Smoothing and Trend-based Principal Component Analysis (JES-TBPCA) for Anomaly Detection which is based on conventional PCA. Experimental results on a real dataset show a remarkably higher performance of JES-TBPCA comparing to conventional PCA model in detection of stuck-at and offset anomalies.

• Journal of Information Processing Systems
• /
• v.8 no.3
• /
• pp.483-494
• /
• 2012

Many recent events, such as terrorist attacks, exposed defects in most sophisticated security systems. Therefore, it is necessary to improve security data systems based on the body or behavioral characteristics, often called biometrics. Together with the growing interest in the development of human and computer interface and biometric identification, human face recognition has become an active research area. Face recognition appears to offer several advantages over other biometric methods. Nowadays, Principal Component Analysis (PCA) has been widely adopted for the face recognition algorithm. Yet still, PCA has limitations such as poor discriminatory power and large computational load. This paper proposes a novel algorithm for face recognition using a mid band frequency component of partial information which is used for PCA representation. Because the human face has even symmetry, half of a face is sufficient for face recognition. This partial information saves storage and computation time. In comparison with the traditional use of PCA, the proposed method gives better recognition accuracy and discriminatory power. Furthermore, the proposed method reduces the computational load and storage significantly.

• Journal of Korean Society of Water and Wastewater
• /
• v.33 no.2
• /
• pp.159-167
• /
• 2019

In this paper the potential of the principal component analysis(PCA) technique for the application of detecting leaks in water pipe networks was evaluated. For this purpose the PCA was conducted to evaluate the relevance of the calculated outliers of a PCA model utilizing the recorded pipe flows and the recorded pipe leak incidents of a case study water distribution system. The PCA technique was enhanced by applying the computational algorithms developed in this study which were designed to extract a partial set of flow data from the original 24 hour flow data so that the effective outlier detection rate was maximized. The relevance of the calculated outliers of a PCA model and the recorded pipe leak incidents was analyzed. The developed algorithm may be applied in determining further leak detection field work for water distribution blocks that have more than 70% of the effective outlier detection rate. However, the analysis suggested that further development on the algorithm is needed to enhance the applicability of the PCA in detecting leaks by considering series of leak reports happening in a relatively short period.

• Journal of The Korean Society of Agricultural Engineers
• /
• v.54 no.1
• /
• pp.73-81
• /
• 2012

For planning and developing rural regions, it is very important to understand and utilize regional characteristics including social, demographic, and economic aspects. The purpose of this study is to find effective analysis techniques and provide a procedure design for mining regional characteristics in South Korea through reviewing and analyzing 41 related studies. The engaged research methods can be classified into five categories (PCA+CA, PCA, CA, GIS, and PCA+GIS) with the combination of three methodologies: principal component analysis (PCA), cluster analysis (CA), and geographical information system (GIS). The combination of PCA and CA occupied about 40 % of research methods used in related studies. The analysis tool of Korean Rural Information Supporting System (KRISS) is designed based on the outcomes of this study and applied to classify the regional capacity of agriculture using agricultural census data (2000) for evaluating its applicability.

• Journal of information and communication convergence engineering
• /
• v.22 no.2
• /
• pp.139-144
• /
• 2024

In this study, four types of fall detection systems - designed with YOLOPose, principal component analysis (PCA), convolutional neural network (CNN), and long short-term memory (LSTM) architectures - were developed and compared in the detection of everyday falls. The experimental dataset encompassed seven types of activities: walking, lying, jumping, jumping in activities of daily living, falling backward, falling forward, and falling sideways. Keypoints extracted from YOLOPose were entered into the following architectures: RAW-LSTM, PCA-LSTM, RAW-PCA-LSTM, and PCA-CNN-LSTM. For the PCA architectures, the reduced input size stemming from a dimensionality reduction enhanced the operational efficiency in terms of computational time and memory at the cost of decreased accuracy. In contrast, the addition of a CNN resulted in higher complexity and lower accuracy. The RAW-LSTM architecture, which did not include either PCA or CNN, had the least number of parameters, which resulted in the best computational time and memory while also achieving the highest accuracy.

• The Journal of the Acoustical Society of Korea
• /
• v.22 no.6
• /
• pp.452-456
• /
• 2003

We discuss how to reduce the number of inverse matrix and its dimensions requested in MLLR framework for speaker adaptation. To find a smaller set of variables with less redundancy, we adapt PCA (principal component analysis) and ICA (independent component analysis) that would give as good a representation as possible. The amount of additional computation when PCA or ICA is applied is as small as it can be disregarded. 10 components for ICA and 12 components for PCA represent similar performance with 36 components for ordinary MLLR framework. If dimension of SI model parameter is n, the amount of computation of inverse matrix in MLLR is proportioned to O(n⁴). So, compared with ordinary MLLR, the amount of total computation requested in speaker adaptation is reduced by about 1/81 in MLLR with PCA and 1/167 in MLLR with ICA.

• Journal of The Korean Society of Agricultural Engineers
• /
• v.55 no.1
• /
• pp.31-38
• /
• 2013

The purpose of this study was to evaluate climate change vulnerability over the agricultural infrastructure in terms of flood and drought using principal component analysis. Vulnerability was assessed using vulnerability resilience index (VRI) which combines climate exposure, sensitivity, and adaptive capacity. Ten flood proxy variables and six drought proxy variables for the vulnerability assessment were selected by opinions of researchers and experts. The statistical data on 16 proxy variables for the local governments (Si, Do) were collected. To identify major variables and to explain the trend in whole data set, principal component analysis (PCA) was conducted. The result of PCA showed that the first 3 principal components explained approximately 83 % and 89 % of the total variance for the flood and drought, respectively. VRI assessment for the local governments based on the PCA results indicated that provinces where having the relatively large cultivation areas were categorized as vulnerable to climate change.

• Phonetics and Speech Sciences
• /
• v.1 no.1
• /
• pp.69-73
• /
• 2009

This paper proposes an efficient global covariance-based principal component analysis (GCPCA) for speaker identification. Principal component analysis (PCA) is a feature extraction method which reduces the dimension of the feature vectors and the correlation among the feature vectors by projecting the original feature space into a small subspace through a transformation. However, it requires a larger amount of training data when performing PCA to find the eigenvalue and eigenvector matrix using the full covariance matrix by each speaker. The proposed method first calculates the global covariance matrix using training data of all speakers. It then finds the eigenvalue matrix and the corresponding eigenvector matrix from the global covariance matrix. Compared to conventional PCA and Gaussian mixture model (GMM) methods, the proposed method shows better performance while requiring less storage space and complexity in speaker identification.

• International Journal of Contents
• /
• v.14 no.4
• /
• pp.24-29
• /
• 2018

EEG-based brain-computer interfaces has focused on explicitly expressed intentions to assist physically impaired patients. For EEG-based-computer interfaces to function effectively, it should be able to understand users' implicit information. Since it is hard to gather EEG signals of human brains, we do not have enough training data which are essential for proper classification performance of implicit intention. In this paper, we improve the subject independent classification of implicit intention through the generation of additional training data. In the first stage, we perform the PCA (principal component analysis) of training data in a bid to remove redundant components in the components within the input data. After the dimension reduction by PCA, we train ICA (independent component analysis) network whose outputs are statistically independent. We can get additional training data by adding Gaussian noises to ICA outputs and projecting them to input data domain. Through simulations with EEG data provided by CNSL, KAIST, we improve the classification performance from 65.05% to 66.69% with Gamma components. The proposed sample generation method can be applied to any machine learning problem with fewer samples.