Comparative Analysis of Machine Learning Algorithms in Automatic Identification and Extraction of Water Boundaries
| Autor: | Youcheng Xu, Aimin Li, Guangduo Qin, Hailong Wang, Meng Fan |
|---|---|
| Rok vydání: | 2021 |
| Předmět: |
Technology
QH301-705.5 Computer science QC1-999 Decision tree Context (language use) Machine learning computer.software_genre remote sensing Range (statistics) General Materials Science Biology (General) QD1-999 Instrumentation Fluid Flow and Transfer Processes Artificial neural network business.industry Physics Process Chemistry and Technology Decision tree learning General Engineering Engineering (General). Civil engineering (General) water extraction Computer Science Applications Random forest Support vector machine Chemistry machine learning algorithm Test set Artificial intelligence TA1-2040 business Algorithm computer modified normalized difference water index (MNDWI) |
| Zdroj: | Applied Sciences Volume 11 Issue 21 Applied Sciences, Vol 11, Iss 10062, p 10062 (2021) |
| ISSN: | 2076-3417 |
| DOI: | 10.3390/app112110062 |
| Popis: | Monitoring open water bodies accurately is important for assessing the role of ecosystem services in the context of human survival and climate change. There are many methods available for water body extraction based on remote sensing images, such as the normalized difference water index (NDWI), modified NDWI (MNDWI), and machine learning algorithms. Based on Landsat-8 remote sensing images, this study focuses on the effects of six machine learning algorithms and three threshold methods used to extract water bodies, evaluates the transfer performance of models applied to remote sensing images in different periods, and compares the differences among these models. The results are as follows. (1) Various algorithms require different numbers of samples to reach their optimal consequence. The logistic regression algorithm requires a minimum of 110 samples. As the number of samples increases, the order of the optimal model is support vector machine, neural network, random forest, decision tree, and XGBoost. (2) The accuracy evaluation performance of each machine learning on the test set cannot represent the local area performance. (3) When these models are directly applied to remote sensing images in different periods, the AUC indicators of each machine learning algorithm for three regions all show a significant decline, with a decrease range of 0.33–66.52%, and the differences among the different algorithm performances in the three areas are obvious. Generally, the decision tree algorithm has good transfer performance among the machine learning algorithms with area under curve (AUC) indexes of 0.790, 0.518, and 0.697 in the three areas, respectively, and the average value is 0.668. The Otsu threshold algorithm is the optimal among threshold methods, with AUC indexes of 0.970, 0.617, and 0.908 in the three regions respectively and an average AUC of 0.832. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|