Deep Learning-Based Image Recognition and Brand Identity Design for Intangible Cultural Heritage in Southern Sichuan

The increasing emphasis on digital protection and creative transformation of intangible cultural heritage highlights the urgent need to apply artificial intelligence to accurately identify and brand rich cultural heritage resources in southern Sichuan. This study puts forward an image recognition and brand design model suitable for intangible cultural heritage in southern Sichuan. This model has obvious advantages in extracting key visual features such as texture, color, and shape, effectively coping with the challenges brought by complex texture and category imbalance in intangible cultural heritage images, and achieving stable and reliable recognition performance. The proposed model improves the overall accuracy, F1 score, and recognition rate of complex texture images, and the overall accuracy reaches 94.1%. Among them, the recognition accuracy of texture, color, and shape reached 92.8%, 95.2%, and 90.1%, respectively, which were better than the traditional models, and accurately captured the detailed features and visual style of intangible cultural heritage items. In the application of brand logo design, the evaluation results of user satisfaction show that the score is high in aesthetics, commercial value, and narrative quality, among which the average score of aesthetics is 3.92, which highlights the advantages of this model in visual language and artistic expression. This study has made a breakthrough in the high-precision and multi-dimensional feature extraction of image recognition of intangible cultural heritage in southern Sichuan Province, and provided an intelligent tool for the brand logo design of intangible cultural heritage. At the same time, the research promotes the integration of digital protection and creative transformation, which is of great value to cultural heritage inheritance, brand promotion, and creative industry application. The research results provide practical guidance for future model optimization, cross-regional intangible cultural heritage identification and multimodal design exploration.

@article{Chen2026,
  title     = {Deep Learning-Based Image Recognition and Brand Identity Design for Intangible Cultural Heritage in Southern Sichuan},
  journal   = {International Journal of Advanced Computer Science and Applications},
  volume    = {17},
  number    = {7},
  year      = {2026},
  publisher = {The Science and Information Organization},
  author    = {Zhihong Chen},
  doi       = {10.14569/IJACSA.2026.0170791},
  url       = {https://doi.org/10.14569/IJACSA.2026.0170791}
}