CGMFN: Conditional Generative Model Fusion Network for Land Surface Temperature Generation

Land surface temperature (LST) is an important parameter representing surface energy, which is of great significance for monitoring urban heat islands, agricultural drought, and global climate. The high-resolution LST observations will address new applications in hydrology. The spatiotemporal fusion method can generate LST with high temporal and spatial resolution. However, the missing data due to cloud cover becomes a main limitation to improve the accuracy of spatiotemporal fusion models. The purpose of the fusion of LST is image prediction and generation, and deep learning generative models provide an effective idea to solve this problem. Therefore, we proposed a conditional generative model fusion network (CGMFN) for LST generation in this article. First, based on the generated model, we construct an unsupervised generation network (GAN) that simultaneously learns and iterates, which can generate fine-spatiotemporal-resolution LST data from reference images with missing values. Then, the spectral normalization (SN) was applied to generators and discriminators to stabilize the training process. The pretraining mechanism was adopted to improve the iteration efficiency of the model. We tested and evaluated the model in the Heihe River Basin (HRB) using FY-4A LST and Moderate Resolution Imaging Spectroradiometer (MODIS) LST datasets. Compared with different methods, CGMFN produces a lower RMSE (average < 0.4 K), LPIPS (average < 0.11), and higher SSIM (average > 0.97). In practical applications, CGMFN can reduce the influence of reference image missing values on fusion results and generate LST products with reliable accuracy.