Model Building to Investigate the Role of Spatial Location in Classifying Satellite Image Using SVM, CART and mBACT: A C
- PDF / 3,231,501 Bytes
- 10 Pages / 595.276 x 790.866 pts Page_size
- 17 Downloads / 131 Views
RESEARCH ARTICLE
Model Building to Investigate the Role of Spatial Location in Classifying Satellite Image Using SVM, CART and mBACT: A Case Study Reshu Agarwal1
Received: 13 January 2014 / Accepted: 11 November 2014 Ó Indian Society of Remote Sensing 2016
Abstract This paper investigates the importance of spatial location of pixels in terms of row-column as an additional explanatory variable in classification along with available spectral bands of remotely sensed data. In view of this, a forward step-wise variable selection algorithm is used to select significant bands/variables and build an optimal model to extract the maximum accuracy. Author performed a case study on the area of town of Wolfville acquired by LANDSAT 5 TM data containing six 30 m resolution spectral bands and pixel location as an additional variable. Data are classified into seven classes using three advanced classifiers i.e. classification and regression trees (CART), support vector machines (SVM) and multi-class Bayesian additive classification tree (mBACT). Traditionally, it is assumed that addition of more explanatory variables always increase the accuracy of classified satellite images. However, results of this study show that adding more variables may sometimes confuse the classifier, that is, if selected carefully, fewer variables can provide the more accurate classification. Importance of row-column information turns out to be more beneficial for mBACT followed by SVM. Interestingly, spatial locations did not turn out to be useful for CART. Based on the findings of this study, mBACT appears to be a slightly better classifier than SVM and a substantially better than CART. Keywords Classification and regression tree Support vector machine LANDSAT data mBACT
& Reshu Agarwal [email protected] 1
Department of Mathematics and Statistics, Acadia University, Wolfville, Canada
Introduction Classified satellite images play vibrant role in numerous remote sensing data applications like land cover change monitoring, forest degradation assessment, hydrological modeling, sustainable development, wildlife habitat modeling, bio-diversity conservation, environmental management and so on. One of the most important products of a raw image is land use/land cover map. In view of this, it becomes essential to classify satellite images with high accuracy. In remote sensing literature, numerous classifiers have already been developed and implemented worldwide (Stuckens et al. 2000; Franklin et al. 2002; Pal and Mather 2003; Gallego 2004). However, accurate labeling of the pixels still remains a challenge due to the complexity of study area terrain, sensor characteristics and training sample size estimation (Blinn 2005; Song et al. 2012). Among various classification methods, Maximum Likelihood (ML) classifier is the most widely used classifier due to its simplicity and availability in remote sensing image processing softwares (Peddle 1993). The ML classifier is based on a parametric model that assumes normally distributed data which is often viol
Data Loading...
