Article

Particulate matter characterization by gray level co-occurrence matrix based support vector machines

EECS Department, University of Toledo, Toledo, OH 43606, USA.
Journal of hazardous materials (Impact Factor: 4.33). 04/2012; 223-224:94-103. DOI: 10.1016/j.jhazmat.2012.04.056
Source: PubMed

ABSTRACT An efficient and highly reliable automatic selection of optimal segmentation algorithm for characterizing particulate matter is presented in this paper. Support vector machines (SVMs) are used as a new self-regulating classifier trained by gray level co-occurrence matrix (GLCM) of the image. This matrix is calculated at various angles and the texture features are evaluated for classifying the images. Results show that the performance of GLCM-based SVMs is drastically improved over the previous histogram-based SVMs. Our proposed GLCM-based approach of training SVM predicts a robust and more accurate segmentation algorithm than the standard histogram technique, as additional information based on the spatial relationship between pixels is incorporated for image classification. Further, the GLCM-based SVM classifiers were more accurate and required less training data when compared to the artificial neural network (ANN) classifiers.

0 Bookmarks
 · 
128 Views
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: In this study, we address the regression problem on set-valued samples that appear in applications. To solve this problem, we propose a support vector regression approach for set-valued samples that generalizes the classicale-support vector regression. First, an initial representative point (or an element) for every set-valued sample is selected, and a weighted distance between the initial representative point and other points is determined. Second, based on the classification consistency principle, a search algorithm to determine the best representative point for every set-valued datum is designed. Thus, the set-valued samples are converted into numeric samples. Finally, a support vector regression that is based on set-valued data is constructed, and the regression results of the set-valued samples can be approximated using the method used for the numeric samples. Furthermore, the feasibility and efficiency of the proposed method is demonstrated using experiments with real-world examples concerning wind speed prediction and the prediction of peak particle velocity.
    Expert Systems with Applications 02/2015; 42:2502–2509. · 1.97 Impact Factor
  • [Show abstract] [Hide abstract]
    ABSTRACT: Grey-level co-occurrence matrices (GLCM) have been on the scene for almost forty years and continue to be widely used today. In this paper we present a method to improve accuracy and robustness against rotation of GLCM features for image classification. In our approach co-occurrences are computed through digital circles as an alternative to the standard four directions. We use discrete Fourier transform normalisation to convert rotation dependent features into rotation invariant ones. We tested our method on four different datasets of natural and synthetic images. Experimental results show that our approach is more accurate and robust against rotation than the standard GLCM features.
    Pattern Recognition Letters 10/2014; 48:34–41. DOI:10.1016/j.patrec.2014.04.006 · 1.06 Impact Factor
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: In this study, we address the regression problem on set-valued samples that appear in applications. To solve this problem, we propose a support vector regression approach for set-valued samples that generalizes the classicale-support vector regression. First, an initial representative point (or an element) for every set-valued sample is selected, and a weighted distance between the initial representative point and other points is determined. Second, based on the classification consistency principle, a search algorithm to determine the best representative point for every set-valued datum is designed. Thus, the set-valued samples are converted into numeric samples. Finally, a support vector regression that is based on set-valued data is constructed, and the regression results of the set-valued samples can be approximated using the method used for the numeric samples. Furthermore, the feasibility and efficiency of the proposed method is demonstrated using experiments with real-world examples concerning wind speed prediction and the prediction of peak particle velocity.
    Expert Systems with Applications 02/2015; 42:2502–2509. · 1.97 Impact Factor