An Improved Neural Network Model Based on DenseNet for Fabric Texture Recognition
Li Tan, Qiang Fu, Jing Li, Mattia Previtali, Loris Nanni · 2024
Read An Improved Neural Network Model Based on DenseNet for Fabric Texture Recognition by Li Tan, Qiang Fu, Jing Li, Mattia Previtali, Loris Nanni (2024) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/pmc-11644935.txt.
Description
Li Tan and co-authors present a 2024 neural network model for sorting knitted fabric by its surface pattern. In garment factories, naming fabric texture by eye is slow and error-prone, yet the step guides automation and quality checks. Older machine methods leaned on hand-picked features, which took time and capped accuracy.
Their model builds on a design in which each layer feeds all later layers. A learnable weight scheme, steered by channel attention, picks the most useful signal channels and trims the rest. Varied learning rates keep that picking sharp through the whole training run, which also prunes the model. Since no public data set existed for the task, the team built one, named KF9. Their model beat a tuned rival network by five points of accuracy.
Catalog record
| CREATOR | Li Tan, Qiang Fu, Jing Li, Mattia Previtali, Loris Nanni |
|---|---|
| YEAR | 2024 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 37 KB |
| VERSION | 1.0 |
| LICENSE | CC BY 4.0 |
| RIGHTS | CC BY |
| SOURCE | Europe PMC |