Improving the Accuracy of Land Use and Land Cover Classification of Landsat Data Using Post-Classification Enhancement

Abstract

Classifying remote sensing imageries to obtain reliable and accurate land use and land cover (LULC) information still remains a challenge that depends on many factors such as complexity of landscape, the remote sensing data selected, image processing and classification methods, etc. The aim of this paper is to extract reliable LULC information from Landsat imageries of the Lower Hunter region of New South Wales, Australia. The classical maximum likelihood classifier (MLC) was first applied to classify Landsat-MSS of 1985 and Landsat-TM of 1995 and 2005. The major LULC identified were Woodland, Pasture/scrubland, Vineyard, Built-up and Water-body. By applying post-classification correction (PCC) using ancillary data and knowledge-based logic rules the overall classification accuracy was improved from about 72% to 91% for 1985 map, 76% to 90% for 1995 map and 79% to 87% for 2005 map. The improved overall Kappa statistics due to PCC were 0.88 for the 1985 map, 0.86 for 1995 and 0.83 for 2005. The PCC maps, assessed by McNemar's test, were found to have much higher accuracy in comparison to their counterpart MLC maps. The overall improvement in classification accuracy of the LULC maps is significant in terms of their potential use for land change modelling of the region.

A comparison of supervised, unsupervised and synthetic land use classification methods in the north of Iran

“…Landsat data are particularly applicable for land use classification on a regional scale because of their lower cost, longer history and higher frequency of archives than other remote sensing data sources (Rozenstein and Karnieli 2010). As a result, Landsat images have been successfully used to classify land use in a large variety of landscapes from homogeneous tropical landscapes to heterogeneous Mediterranean landscapes (Alrababah and Alhamad 2006;Koutsias and Karteris 2003;Manandhar et al 2009;Schulz et al 2010;Brandt et al 2013;Sexton et al 2013;Zegre et al 2013). Landsat's spatial resolution is 30 m. The Landsat sensors include the Landsat 5 Thematic Mapper (TM), the Landsat 7 enhanced thematic mapper plus (ETM?)…”

“…Synthetic methods use additional maps next to satellite images to separate land use types with the same reflectance. This integration of remotely sensed data with other data sources such as previously existing land use data, geology, transportation network or digital elevation models (DEMs, with possibly derivatives such as slope and curvature) can result in higher classification accuracy (Manandhar et al 2009;Stefanov et al 2001;Shalaby and Tateishi 2007). Synthetic methods can take the form of decision rules that use data sources in combination (Lillesand and Kiefer 2000).…”

Exploring the Use of Google Earth Imagery and Object-Based Methods in Land Use/Cover Mapping

“…Features usually define the upper and lower limits of a range of measurable characteristics of objects. Image objects within the defined limits are assigned to a specific class, while those outside of the ranges are assigned to other classes [36]. The feature selection is thus to search out the most relevant features for each class so as to help efficiently perform an image classification with a high accuracy [32].…”

Spatial air pollution modelling for a West-African town

“…The final classified image was again examined for the presence of flagrant misclassifications and corrected with a post classification process. In order to provide a measure for the accuracy of the LULC classification, well recognizable regions of interests for each class were defined in the town of Kaédi and its closer surroundings and a confusion matrix was calculated (Manandhar et al, 2009). Based on the final LULC data set, geographical variables for each land class were calculated for buffer sizes of 100, 300, 500 and 1000 m. Buffers (zones of influence) for all of the predictor variables were made to take into account the environmental processes of the variables and the geographic extent of our study area.…”