An MLC and U-Net Integrated Method for Land Use/Land Cover Change Detection Based on Time Series NDVI-Composed Image from PlanetScope Satellite

Land use/land cover change (LUCC) detection based on optical remote-sensing images is an important research direction in the field of remote sensing. The key to it select appropriate data source and method. In recent years, continuous expansion construction land urban areas has become main reason for increase LUCC demand. However, due complexity diversity land-cover types, difficult obtain high-precision classification results. this article, a 12-month time series NDVI (Normalized Difference Vegetation Index) image study area was generated high spatial temporal resolution PlanetScope satellite images. According image, representative samples were selected, changed selected at same time. This method could directly results through classification. First, Maximum Likelihood Classification (MLC), classical machine-learning method, used supervised classification, needed deep learning according Then, U-Net model, which can fully identify explore semantic information Finally, article made comparative analysis two demonstrate that overall accuracy significantly higher than single-scene mean NDVI. proposed effectively extract areas. MLC model 79.38% 85.26%, respectively. Therefore, deep-learning improve detection.