Papers by Keyword: Principal Component Analysis (PCA)

Abstract: Denoising is an important issue for laser active image. This paper attempted to process laser active image in the low-dimensional sub-space. We adopted the principal component analysis with local pixel grouping (LPG-PCA) denoising method proposed by Zhang [1], and compared it with the conventional denoising method for laser active image, such as wavelet filtering, wavelet soft threshold filtering and median filtering. Experimental results show that the image denoised by LPG-PCA has higher BIQI value than other images, most of the speckle noise can be reduced and the detail structure information is well preserved. The low-dimensional sub-space idea is a new direction for laser active image denoising.

Abstract: To address the problem that the dimension of the feature vector extracted by Local Binary Pattern (LBP) for face recognition is too high and Principal Component Analysis (PCA) extract features are not the best classification features, an efficient feature extraction method using LBP, PCA and Maximum scatter difference (MSD) has been introduced in this paper. The original face image is firstly divided into sub-images, then the LBP operator is applied to extract the histogram feature. and the feature dimensions are further reduced by using PCA. Finally,MSD is performed on the reduced PCA-based feature.The experimental results on ORL and Yale database demonstrate that the proposed method can classify more effectively and can get higher recognition rate than the traditional recognition methods.

Abstract: Problem of merging multi-attribute, small-batch orders is studied in this paper. A merge algorithm based on clustering is proposed to solve the problem. We mainly use the method of principal component analysis and K-means clustering technology in the algorithm and then merge the orders with similar attributes into production bathes. Finally we verify the algorithm through actual order data and the results show that the algorithm can merge the customer orders efficiently. The algorithm has a great guiding significance.

Abstract: As a zero-emission mode of transportation, an increasing number of Electric Vehicles (EV) have come into use in our daily lives. The EV charging station is an important component of the Smart Grid which is now facing the challenges of big data. This paper presents a data compression and reconstruction method based on the technique of Principal Component Analysis (PCA). The data reconstruction error Normalized Absolute Percent Error (NAPE) is taken into consideration to balance the compression ratio and data reconstruction quality. By using the simulated data, the effectiveness of data compression and reconstruction for EV charging stations are verified.

Abstract: In order to identify the key factors influencing the development of China's enterprise management innovation and study the interaction between them. 23 factors concerning management innovation are analyzed with the method of principal component analysis on the basis of the former research. The analytical results show that enterprise supply chain management method, enterprise field management method, enterprise performance management method, enterprise quality management, enterprise production management, enterprise process management are the key factors in management innovation. Paying more attention on the key factors helps to capture the principal contradiction, and to improve the efficiency of the analysis on enterprise management innovation.

Abstract: The county economy in Liaoning Province has made great progress in recent years. But the economic development capacity is different among different regions. It’s necessary to analysis the comprehensive strength to support decision. This paper analysis the data of 44 counties in 2010 and a method to evaluate the comprehensive strength is given. This method is based on principal component analysis of 16 statistical indicators. And the rank of comprehensive strength evaluation is given.

Abstract: There is no consensus on the impact of free cash flow upon corporate performance. Based on the data from 2006-2012 of all listed real estate companies in China, authors studied the relationship between the free cash flow and performance of these firms. Using principal component analysis and regression analysis, key financial performance indicators were calculated out of 18 financial performance indicators, and these key indicators of sample companies were correlated to their free cash flow. The results showed that the free cash flow of a company is negatively linear-correlated to its performance, i.e., too much free cash flow will lead the corporate performance to decline. Therefore, the investors and the managers should avoid business inefficient because of too much free cash flow, which triggers the investment risk and loss.

Abstract: Principal Component Analysis (PCA) is a method of multivariate statistical analysis and has been widely used in statistical and mathematical analysis. We use this method in the evaluation of competitiveness of small firms. Using the data of 30 small firms, we build the index system to evaluate competitiveness. Our results show that Principal Component Analysis (PCA) is useful in dimension reducing and we find that profitability, growth,size and human resource are important influencing factors in the competitiveness of small firms.

Abstract: This article analyzed the defects of SVM-RFE feature selection algorithm, put forward new feature selection method combined SVM-RFE and PCA. Firstly, get the best feature subset through the method of cross validation of k based on SVM-RFE. Then, the PCA decreased the dimension of the feature subset and got the independent feature subset. The independent feature subset was the training and testing subset of SVM. Make experiments on five subsets of UCI, the results indicated that the training and testing time was shortened and the recognition accuracy rate of the SVM was higher.

Abstract: In order to better predict the aviation material unsafe events, a BP neural network model based on PCA feature extraction is proposed. Firstly, the training samples of aviation material unsafe events are used to carry out the PCA feature extraction, and then using the extracted basic features, BP neural network model is established. The numerical example shows that, the hybrid model proposed is better than that of alone BP neural network model, and it is effective and feasible to establish the unsafe events model for aviation material.