A new isotropic locality improved kernel for pattern classifications in remote sensing imagery
(2016) In Spatial Statistics 17. p.71-82- Abstract
Kernel based learning algorithms are sensitive to the choice of appropriate kernel function and parameter setting. Classification accuracies yielded by the kernel based classifiers may show variation depending on the choice of the kernel and its associated parameters. Suggesting an efficient kernel function and effective setting of kernel parameters are thus important problems for kernel based classifiers. In this article, we have investigated the performance of the existing kernel functions with our proposed kernel using support vector machines (SVMs). Linear, polynomial, sigmoid, radial basis function (RBF) and the proposed kernel are applied for the classification of the ten real life data sets having features and classes ranging... (More)
Kernel based learning algorithms are sensitive to the choice of appropriate kernel function and parameter setting. Classification accuracies yielded by the kernel based classifiers may show variation depending on the choice of the kernel and its associated parameters. Suggesting an efficient kernel function and effective setting of kernel parameters are thus important problems for kernel based classifiers. In this article, we have investigated the performance of the existing kernel functions with our proposed kernel using support vector machines (SVMs). Linear, polynomial, sigmoid, radial basis function (RBF) and the proposed kernel are applied for the classification of the ten real life data sets having features and classes ranging from 4 to 19 and 2 to 5 respectively. The performance of different kernels is also demonstrated on two multispectral and two hyperspectral images. Experimental results on these data sets show the effectiveness of the proposed kernel on several data distributions.
(Less)
- author
- Chakraborty, Debasis ; Sarkar, Anasua LU and Maulik, Ujjwal
- publishing date
- 2016-08-01
- type
- Contribution to journal
- publication status
- published
- keywords
- Kernel function, Pattern classification, Remote sensing image, Structural risk minimization, Support vector machines
- in
- Spatial Statistics
- volume
- 17
- pages
- 71 - 82
- publisher
- Elsevier
- external identifiers
-
- scopus:84969822570
- ISSN
- 2211-6753
- DOI
- 10.1016/j.spasta.2016.04.003
- language
- English
- LU publication?
- no
- id
- 93d26d99-4443-41b4-84a2-7012c98eaed2
- date added to LUP
- 2018-10-09 09:45:24
- date last changed
- 2022-02-15 05:13:09
@article{93d26d99-4443-41b4-84a2-7012c98eaed2, abstract = {{<p>Kernel based learning algorithms are sensitive to the choice of appropriate kernel function and parameter setting. Classification accuracies yielded by the kernel based classifiers may show variation depending on the choice of the kernel and its associated parameters. Suggesting an efficient kernel function and effective setting of kernel parameters are thus important problems for kernel based classifiers. In this article, we have investigated the performance of the existing kernel functions with our proposed kernel using support vector machines (SVMs). Linear, polynomial, sigmoid, radial basis function (RBF) and the proposed kernel are applied for the classification of the ten real life data sets having features and classes ranging from 4 to 19 and 2 to 5 respectively. The performance of different kernels is also demonstrated on two multispectral and two hyperspectral images. Experimental results on these data sets show the effectiveness of the proposed kernel on several data distributions.</p>}}, author = {{Chakraborty, Debasis and Sarkar, Anasua and Maulik, Ujjwal}}, issn = {{2211-6753}}, keywords = {{Kernel function; Pattern classification; Remote sensing image; Structural risk minimization; Support vector machines}}, language = {{eng}}, month = {{08}}, pages = {{71--82}}, publisher = {{Elsevier}}, series = {{Spatial Statistics}}, title = {{A new isotropic locality improved kernel for pattern classifications in remote sensing imagery}}, url = {{http://dx.doi.org/10.1016/j.spasta.2016.04.003}}, doi = {{10.1016/j.spasta.2016.04.003}}, volume = {{17}}, year = {{2016}}, }