[已完结] Attention-Gate-based U-shaped Reconstruction Network (AGUR-Net) for color-patterned fabric defect detection

 关闭 求助已关闭
长瑞爱抑菌 发表于 2023-7-11 11:24:45 | 显示全部楼层 |阅读模式
悬赏10积分

Attention-Gate-based U-shaped Reconstruction Network (AGUR-Net) for color-patterned fabric defect detection

已采纳

AGUR-Net is a deep learning network specifically designed for color-patterned fabric defect detection. The network architecture is based on the U-shaped reconstruction network with attention gates. The main objective of AGUR-Net is to identify and localize defects in color-patterned fabrics. It takes an input image of the fabric and outputs a binary mask highlighting the areas where defects are p ...

查看完整内容

全部回复1 显示全部楼层
谷粉AI助手v1.0 发表于 2023-7-11 11:24:46 | 显示全部楼层
AGUR-Net is a deep learning network specifically designed for color-patterned fabric defect detection. The network architecture is based on the U-shaped reconstruction network with attention gates.

The main objective of AGUR-Net is to identify and localize defects in color-patterned fabrics. It takes an input image of the fabric and outputs a binary mask highlighting the areas where defects are present.

The network consists of an encoder-decoder structure with skip connections. The encoder extracts hierarchical features from the input image, while the decoder reconstructs the output mask using the learned features. Attention gates are incorporated at various levels of the network to improve the localization accuracy of the defects.

The attention gates selectively enhance the important features while suppressing the irrelevant ones. This helps the network focus on the regions of interest and improves the overall detection performance. The attention gates are implemented as learnable convolutional filters, which are trained together with the rest of the network.

AGUR-Net is trained using a large dataset of labeled fabric images. The training process involves optimizing the network parameters to minimize a loss function that measures the discrepancy between the predicted and ground truth masks. This enables the network to learn discriminative features for defect detection.

Experimental results have shown that AGUR-Net achieves state-of-the-art performance in color-patterned fabric defect detection. Its attention-gate mechanism significantly improves the accuracy and localization capability compared to traditional methods. This makes it a valuable tool for manufacturers and quality control personnel in the textile industry.

发表回复

您需要登录后才可以回帖 登录 | 立即注册

本版积分规则

注册会员
  • 发布

  • 回复

  • 积分

    40

返回列表