Evaluating the Performance of Multi-Class and Single-Class Classification Approaches for Mountain Agriculture Extraction
- PDF / 5,115,893 Bytes
- 11 Pages / 595.276 x 790.866 pts Page_size
- 100 Downloads / 182 Views
RESEARCH ARTICLE
Evaluating the Performance of Multi-Class and Single-Class Classification Approaches for Mountain Agriculture Extraction Using Time-Series NDVI Saptarshi Mondal1
•
Chockalingam Jeganathan1
Received: 12 April 2017 / Accepted: 8 October 2018 Ó Indian Society of Remote Sensing 2018
Abstract Supervised multi-class classification (MCC) approach is widely being used for regional-level land use–land cover (LULC) mapping and monitoring. However, it becomes inefficient if the end user wants to map only one particular class. Therefore, an improved single-class classification (SCC) approach is required for quick and reliable map production purpose. In this regard, the current study attempts to evaluate the performance of MCC and SCC approaches for extracting mountain agriculture area using time-series normalized differential vegetation index (NDVI). At first, samples of eight LULC classes were acquired using Google Earth image, and corresponding temporal signatures (TS) were extracted from time-series NDVI to perform classification using minimum distance to mean (MDM) and spectral angle mapper (i.e., multi-class SAM—MCSAM) under MCC approach. Secondly, under SCC approach, the TS of three agriculture classes (i.e., agriculture, mixed agriculture and plantation) were utilized as a reference to extract agriculture extent using Euclidean distance (ED) and SAM (i.e., single-class SAM— SCSAM) algorithms. The area of all four maps (i.e., MDM—19.77% of total geographical area (TGA), MCSAM—21.07% of TGA, ED—15.23% of TGA, SCSAM—13.85% of TGA) was compared with reference agriculture area (14.54% of TGA) of global land cover product, and SCC-based maps were found to have close agreement. Also, the class-wise detection accuracy was evaluated using random sample point-based error matrix which reveals the better performance of ED-based map than rest three maps in terms of overall accuracy and kappa coefficient. Keywords Euclidean distance Spectral angle mapper MODIS NDVI Time series Mountain agriculture
Introduction Consistent, accurate and timely agriculture information at the regional level is crucial for the policy maker, government and non-government agencies and researchers (Wardlow et al. 2007; Husak et al. 2008; Wu et al. 2014). Irrespective of scale, information about the extent and location of agriculture area is mainly used as a baseline for time-to-time resource assessment for addressing food security issues (Justice and Becker-Reshef 2007). With the regular availability and
& Saptarshi Mondal [email protected] Chockalingam Jeganathan [email protected] 1
Department of Remote Sensing, Birla Institute of Technology (BIT), Mesra, Ranchi, India
significant improvement in terms of spatiotemporal resolution of remote sensing-based satellite data product over the last four decades, agriculture mapping became feasible, especially in those areas where reliable agriculture information is inconsistent mainly due to the inaccessible complex terrain (Wu et al. 2008; Delrue et al. 2013). Regiona
Data Loading...
