Research on Mural Style Generation and Parameter Adjustment Based on Stable Diffusion

In this paper, we propose a method for the restoration and generation of mural art based on the Stable Diffusion Inpainting model, which aims to explore the influence of different parameters on the stylization and detail preservation of murals. The system architecture includes an input module, a processing module, and an output module, allowing users to upload concept sketches and masks, and generate murals with three adjustable parameters: style strength, modern blend scale, and historical detail accuracy. By building an interactive system, users can upload concept sketches and custom masks to automatically generate mural images that conform to historical textures and styles. The methods mainly include image and mask preprocessing, model parameter adjustment, intelligent image scaling, style fusion and online interactive interface design. The results show that the appropriate parameter setting can enhance the details and stylization effect while retaining the artistic characteristics of the mural, and provide technical support for the restoration and innovation of the mural. Finally, the advantages and limitations of the system are discussed, and the future optimization direction is discussed. Limitations and future improvement directions.