Scheduled System Maintenance:
On May 6th, single article purchases and IEEE account management will be unavailable from 8:00 AM - 5:00 PM ET (12:00 - 21:00 UTC). We apologize for the inconvenience.
By Topic

A Hybrid Model for Classification of Remote Sensing Images With Linear SVM and Support Vector Selection and Adaptation

Sign In

Cookies must be enabled to login.After enabling cookies , please use refresh or reload or ctrl+f5 on the browser for the login options.

Formats Non-Member Member
$31 $13
Learn how you can qualify for the best price for this item!
Become an IEEE Member or Subscribe to
IEEE Xplore for exclusive pricing!
close button

puzzle piece

IEEE membership options for an individual and IEEE Xplore subscriptions for an organization offer the most affordable access to essential journal articles, conference papers, standards, eBooks, and eLearning courses.

Learn more about:

IEEE membership

IEEE Xplore subscriptions

1 Author(s)
Kaya, G.T. ; Inst. of Earthquake Eng. & Disaster Manage., Istanbul Tech. Univ., Istanbul, Turkey

A Linear Support Vector Machine (LSVM) is based on determining an optimum hyperplane that separates the data into two classes with a maximum margin. LSVM typically has higher classification accuracy for linearly separable data than nonlinearly separable data. For this type of data, Support Vector Selection and Adaptation (SVSA) method was developed that uses the support vectors obtained by LSVM and adapts them with respect to training dataset. However, the SVSA's classification performance may not be very satisfactory for linearly separable data in comparison to LSVM depending on the dataset used. In this paper, a hybrid model was presented that combines the results of LSVM and the SVSA efficiently. The main idea of the hybrid model is to utilize the performance of LSVM with the SVSA as LSVM model is already available during implementation of the SVSA. The method which uses the proposed hybrid model is called the Hybrid Support Vector Selection and Adaptation method (HSVSA). In order to show the effectiveness of the proposed model, one real multispectal and two hyperspectral dataset were experimented with the SVSA, LSVM, nonlinear SVM and the HSVSA in the classification. The results showed that when LSVM performs better than the SVSA, the HSVSA achieves LSVM's performance with the hybrid model, and vice versa.

Published in:

Selected Topics in Applied Earth Observations and Remote Sensing, IEEE Journal of  (Volume:6 ,  Issue: 4 )