Extraction of Swamp Wetland Information Based on Multi-source Remote Sensing Data and Machine Learning Algorithm
This article is based on multi-source remote sensing data and machine learning algorithms,aiming to explore methods to improve the accuracy of swamp wetland information extraction.By using the Sentinel-1 synthetic aperture radar(SAR)data,Sentinel-2 multi-spectral data and terrain data obtained from the GEE(Google Earth Engine)platform,combined with a variety of machine learning algorithms,the remote sensing information of swamp wetlands was extracted and analyzed.Verification tests were carried out in the Nanwenghe nature reserve in the Daxingan mountains and the Momoge nature reserve in the Songnen plain area.The results show that the research method using multi-source remote sensing data combined with machine learning algorithms significantly improve the accuracy of swamp wetland information extraction.In the extraction of swamp wetland information in plain areas,radar data has a higher contribution than multi-spectral data;In the extraction of swamp wetland information in mountainous areas,the two have close contributions.The combination of Sentinel-1,Sentinel-2 data and terrain humidity index data is more conducive to the extraction of swamp wetland information in various terrains.The random forest is the most effective machine learning algorithm for extracting swamp wetland information,achieving the highest accuracy.
万方数据
维普
