# FDTransUnet: An aluminum surface defect segmentation model based on feature differentiation

## TL;DR

FDTransUnet is a new model for detecting defects on aluminum surfaces that improves accuracy and generalization using data augmentation and a Transformer-enhanced U-net architecture.

## Contribution

Proposes FDTransUnet, a novel segmentation model combining feature differentiation and Transformer architecture for industrial defect detection.

## Key findings

- FDTransUnet achieved 94.5% MPA and 89.7% Dice coefficient on the aluminum surface defect dataset.
- The model demonstrated good generalization performance on the steel surface defect dataset.
- The composite loss function improved segmentation accuracy by addressing foreground-background imbalance.

## Abstract

Aiming at the current problems in the field of industrial defect segmentation, such as difficulty of obtaining a large number of defect samples, low recognition accuracy and lack of segmentation accuracy, a surface defect segmentation model for aluminum based on feature differentiation is proposed: FDTransUnet. First, the limited defective samples are effectively expanded by the feature differentiation data augmentation strategy to alleviate the overfitting problem caused by the insufficient sample. Second, the Transformer architecture is added by improving the U-net network, and the improved network combines the global self-attention mechanism of the Transformer and the hierarchical structure of the U-net, which can effectively extract the local and global information in the defect sample. Finally, a composite loss function is constructed to address the problem of unbalanced foreground and background sizes of defective samples and to improve segmentation accuracy. The experimental results show that FDTransUnet achieves 94.5% MPA and 89.7% Dice coefficient on the aluminum surface defect dataset. In the final generalization experiment, FDTransUnet is validated with other mainstream segmentation models on the steel surface defect dataset, and the experiment proves that the segmentation model has good generalization performance and robustness, and can be applied to different scenarios of industrial inspection.