A Comprehensive Evaluation of Approaches for Built-Up Area Extraction from Landsat OLI Images Using Massive Samples

Abstract

Detailed information about built-up areas is valuable for mapping complex urban environments. Although a large number of classification algorithms for such areas have been developed, they are rarely tested from the perspective of feature engineering and feature learning. Therefore, we launched a unique investigation to provide a full test of the Operational Land Imager (OLI) imagery for 15-m resolution built-up area classification in 2015, in Beijing, China. Training a classifier requires many sample points, and we proposed a method based on the European Space Agency’s (ESA) 38-m global built-up area data of 2014, OpenStreetMap, and MOD13Q1-NDVI to achieve the rapid and automatic generation of a large number of sample points. Our aim was to examine the influence of a single pixel and image patch under traditional feature engineering and modern feature learning strategies. In feature engineering, we consider spectra, shape, and texture as the input features, and support vector machine (SVM), random forest (RF), and AdaBoost as the classification algorithms. In feature learning, the convolutional neural network (CNN) is used as the classification algorithm. In total, 26 built-up land cover maps were produced. The experimental results show the following: (1) The approaches based on feature learning are generally better than those based on feature engineering in terms of classification accuracy, and the performance of ensemble classifiers (e.g., RF) are comparable to that of CNN. Two-dimensional CNN and the 7-neighborhood RF have the highest classification accuracies at nearly 91%; (2) Overall, the classification effect and accuracy based on image patches are better than those based on single pixels. The features that can highlight the information of the target category (e.g., PanTex (texture-derived built-up presence index) and enhanced morphological building index (EMBI)) can help improve classification accuracy. The code and experimental results are available at https://github.com/zhangtao151820/CompareMethod.