Dynamic Cloud Scene Image Simulation Based on a Multi-Scale Superposition Method with Temporal Smoothing

Addressing the challenge of real-time simulation for large-scale scenarios, which is often hindered by the high computational complexity and memory consumption of traditional dynamic cloud simulation methods, this paper introduces an efficient dynamic cloud simulation approach that integrates Multi-Scale Superposition (MSS) and Inter-Frame Interpolation (IFI). Initially, the method employs the MSS algorithm to generate discrete interval frame cloud images along the time axis, thereby significantly reducing the computational load compared to conventional methods. Subsequently, IFI technology is utilized to complete continuous frames from the interval frames, producing a dynamic cloud sequence. This resolves the issue of low efficiency in modeling continuous processes within traditional simulations. Experimental results demonstrate that this method achieves a more than tenfold increase in computational efficiency compared to conventional methods across large-scale scenarios ranging from 10 k × 10 k to 100 k × 100 k pixels. The advantage of this method becomes more pronounced with increasing image dimensions, providing an effective solution for large-scale dynamic cloud simulation.