IIoT ‐Driven Digitalization of Intangible Cultural Heritage: Mongolian Embroidery Generation Technology Based on Stable Diffusible LoRA Multi‐Level Tag Optimization

This study, under the background of the Industrial Internet of Things (IIoT) promoting the digitalization of intangible cultural heritage, proposes an improved solution based on Stable Diffusion and LoRA fine‐tuning. By constructing a three‐level label system of “pattern type‐color combination‐cultural connotation,” the generation effect of Mongolian embroidery is optimized. The experiment adopts a three‐group control design and verifies it with indicators such as PSNR and cultural compliance. The results show that the three‐level label model is significantly superior to other groups in terms of pattern restoration and cultural authenticity. This study provides a lightweight solution of “cloud‐edge‐device” collaboration for the digital inheritance of ethnic culture in the IIoT environment and points out the direction for the generation of automated labels and the optimization of distributed computing.