An Integrated Land Cover Mapping Method Suitable for Low-Accuracy Areas in Global Land Cover Maps

Tian Zeng,Qingke Wen,Le Yu,Xiao Wang,Lei Wang,Zengxiang Zhang

doi:10.3390/rs11151777

Abstract

In land cover mapping, an area with complex topography or heterogeneous land covers is usually poorly classified and therefore defined as a low-accuracy area. The low-accuracy areas are important because they restrict the overall accuracy (OA) of global land cover classification (LCC) data generated. In this paper, low-accuracy areas in China (extracted from the MODIS global LCC maps) were taken as examples, identified as the regions having lower accuracy than the average OA of China. An integrated land cover mapping method targeting low-accuracy regions was developed and tested in eight representative low-accuracy regions of China. The method optimized procedures of image choosing and sample selection based on an existent visually-interpreted regional LCC dataset with high accuracies. Five algorithms and 16 groups of classification features were compared to achieve the highest OA. The support vector machine (SVM) achieved the highest mean OA (81.5%) when only spectral bands were classified. Aspect tended to attenuate OA as a classification feature. The optimal classification features for different regions largely depends on the topographic feature of vegetation. The mean OA for eight low-accuracy regions was 84.4% by the proposed method in this study, which exceeded the mean OA of most precedent global land cover datasets. The new method can be applied worldwide to improve land cover mapping of low-accuracy areas in global land cover maps.

Highlights

Land cover data are indispensable for many studies and for practical applications such as global change assessment, sustainable development, hydrological modeling and land resource management [1,2,3,4,5,6]
This study provides a way for locating low-accuracy areas and an integrated method for the land cover mapping of low-accuracy areas
The integrated method has improved the accuracy of low-accuracy regions so that it has exceeded a lot over the global land cover classification (LCC) datasets derived from the same image source

Summary

Introduction

Land cover data are indispensable for many studies and for practical applications such as global change assessment, sustainable development, hydrological modeling and land resource management [1,2,3,4,5,6]. Multiple global land cover datasets have been produced in the last two decades. Most of these datasets were assessed with accuracies of less than 80% by users or producers [7,8,9,10,11,12]. According to an accuracy assessment for global land cover datasets based on 38,664 test samples, some regions located in highly heterogeneous areas have lower accuracies than others [13]. Areas with complex topography or varied land cover types tend to be characterized by high heterogeneity and low accuracy, so they can be included among the low-accuracy areas. How to improve the accuracy of low-accuracy areas is essential for improving the overall accuracy of land cover mapping at large scale

Objectives

Methods

Findings

Conclusion