A Comparison of Model-Assisted Estimators to Infer Land Cover/Use Class Area Using Satellite Imagery

Yizhan Li,Jianyu Gu,Yaozhong Pan,Xiufang Zhu,Xianfeng Liu,Anzhou Zhao

doi:10.3390/rs6098904

Abstract

Remote sensing provides timely, economic, and objective data over a large area and has become the main data source for land cover/use area estimation. However, the classification results cannot be directly used to calculate the area of a given land cover/use type because of classification errors. The main purpose of this study is to explore the performance and stability of several model-assisted estimators in various overall accuracies of classification and sampling fractions. In this study, the confusion matrix calibration direct estimator, confusion matrix calibration inverse estimator, ratio estimator, and simple regression estimator were implemented to infer the areas of several land cover classes using simple random sampling without replacement in two experiments: a case study using simulation data based on RapidEye images and that using actual RapidEye and Huan Jing (HJ)-1A images. In addition, the simple estimator using a simple random sample without replacement was regarded as a basic estimator. The comparison results suggested that the confusion matrix calibration estimators, ratio estimator, and simple regression estimator could provide more accurate and stable estimates than the simple random sampling estimator. In addition, high-quality classification data played a positive role in the estimation, and the confusion matrix inverse estimators were more sensitive to the classification accuracy. In the simulated experiment, the average deviation of a confusion matrix calibration inverse estimator decreased by about 0.195 with the increasing overall accuracy of classification; otherwise, the variation of the other three model-assisted estimators was 0.102. Moreover, the simple regression estimator was slightly superior to the confusion matrix calibration estimators and required fewer sample units under acceptable classification accuracy levels of 70%–90%.

Highlights

The area of each land cover class is an essential parameter of resources investigation, resources management, and earth observation [1,2,3,4,5]
When the auxiliary data has low accuracy (62%), the coefficient of variance (CV), average deviation (AD), and relative root mean square error (RRMSE) derived from inverse estimator (INE) is worse than that derived from simple random sampling estimator (SRSE)
The precision of estimation derived from confusion matrix calibration depends on how well the confusion matrix describes the error in each class

Summary

Introduction

The area of each land cover class is an essential parameter of resources investigation, resources management, and earth observation [1,2,3,4,5]. Model-based inference uses models to predict the response variable for individual population units, and the sample could be non-probability and purposive [11] It focuses on the parameters of the superpopulation, and the validity of estimates depends on the fitness of model [9]. In this type of inference, satellite imageries are often used as auxiliary data or variables to establish the model. These estimators take advantage of the correlation of the auxiliary variables and variables of interest from the probability sample to establish a model for improving the precision of the estimation, acting as a middle ground between the simple random sampling estimator and model-based estimator [8] In these estimations, satellite imageries are usually viewed as key auxiliary data from which the auxiliary variables are derived

Objectives

Methods

Results

Conclusion