GDP Spatialization in Zhengzhou City Based on NPP Nighttime Lights and Socioeconomic Statistical Data Using Machine Learning

GIS DATASETS：
RESEARCH_AREA_NLD_DATA.zip: the vector maps data of the study area.
2012_NLD_of_Zhengzhou_City, 2013_NLD_of_Zhengzhou_City, 2014_NLD_of_Zhengzhou_City,
2015_NLD_of_Zhengzhou_City, 2016_NLD_of_Zhengzhou_City, 2017_NLD_of_Zhengzhou_City,
PROGRAMS DATASETS:
FCN8.py, Unet.py, predict.py, 训练数据.py
FIGUREs DATASETS:
Figure 1. Location map of the study area (Zhengzhou City, Henan Province),
Figure 2. Night Light Data (NPP) of Zhengzhou City, 2017,
Figure 3. Flow Diagram,
Figure 4. Night Light Data (NPP) of Zhengzhou city 2017,
Figure 5. Flow chart of classification process of SVM algorithm,
Figure 6. Flow chart of the semantic segmentation classification process,
Figure 7. Structure diagram of Full Convolution Neural Network (FCN) Model,
Figure 8. Structure diagram of U-Net Neural Network Model,
Figure 9. (a) NPP training results of SVM algorithm (2012-2017),
Figure 9. (b) NPP training results of FCN model (2012-2017),
Figure 9. (c) NPP training results of U-Net model (2012-2017),
Figure 10. Comparison of Calculated data,
Figure 11. Accuracy with Given Data by Municipal of bureau,
Figure 12. (a) GDP 2012 Model,
Figure 12. (b) GDP 2013 Model,
Figure 12. (c) GDP 2014 Model,
Figure 12. (d) GDP 2015 Model,
Figure 12. (e) GDP 2016 Model,
Figure 12. (f) GDP 2017 Model.