Abstract:
Land cover is one of basic data layers in geographic information system for physical planning and environmental
monitoring. Digital image classification is generally performed to produce land cover maps from remote sensing data,
particularly for large areas. In the present study the multispectral image from IRS LISS-III image along with ancillary data
such as vegetation indices, principal component analysis and digital elevation layers, have been used to perform image
classification using maximum likelihood classifier and decision tree method. The selected study area that is located in
north-east Iran represents a wide range of physiographical and environmental phenomena. In this study, based on Land
Cover Classification System (LCCS), seven land cover classes were defined. Comparison of the results using statistical
techniques showed that while supervised classification (i.e. MLC) produces an overall accuracy of about 72%; the
decision tree method, which improves the classification accuracy, can increase the results by about 7 percent to 79%. The
results illustrated that ancillary data, especially vegetation indices and DEM, are able to improve significantly
classification accuracy in arid and semi arid regions, and also the mountainous or hilly areas.
Machine summary:
Digital image classification is generally performed to produce land cover maps from remote sensing data, particularly for large areas.
In the present study the multispectral image from IRS LISS-III image along with ancillary data such as vegetation indices, principal component analysis and digital elevation layers, have been used to perform image classification using maximum likelihood classifier and decision tree method.
The results illustrated that ancillary data, especially vegetation indices and DEM, are able to improve significantly classification accuracy in arid and semi arid regions, and also the mountainous or hilly areas.
The limitation to achieve higher classification accuracies discussed by Defries et al (1998), Loveland et al (1999) and Hansen et al (2000), emphasize data quality of the input and the number and nature of the land cover classes of interest.
e. , MLC and DT) to separation various land cover classes; secondly, to evaluate performance of ancillary data for improve image classification.
Generation of ancillary data Principal components analysis (PCA), Digital Elevation Model (DEM) and Vegetation Indices (VI) data layers were used as additional bands (referred as ancillary data) to perform and improve DT classification.
For selection and use of vegetation indices in DT method, the correlation analysis between field data and percentage cover of plant species were performed in statistical software.
The case study presented in this paper showed a remarkable increase in accuracy of land cover classification on incorporation of ancillary data layers with IRS LISS-III image.
Decision tree classification of land cover from remotely sensed data.