Scholarlink

<p>Porcelain fragment classification is crucial for cultural relic restoration. Traditional manual methods relying on macroscopic features struggle to balance accuracy and efficiency. This study proposes a multi-channel graph convolutional network (GCN) architecture integrated with multi-directional Gaussian filtering. First, images are converted into graph structures and processed with multi-directional Gaussian filtering to reduce noise while preserving texture details. The proposed multi-channel GCN extracts rich interconnected features from multiple perspectives. Experimental results achieved 93.33% accuracy, outperforming ResNet50 by 3.33% and DenseNet121 by 2.80%. This approach effectively addresses noise interference and uneven feature distribution in microscopic images.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Multiple channel GCN with multiple directional Gaussian for porcelain microscopic image classification

  • 期刊论文

  • 出版模式：

    开放获取

  • 发表日期：

    2025年6月10日

查看全文
  • Xinda Liu,
  • Jinkai Zhen,
  • Yangyang Liu,
  • Guohua Geng,
  • Wuyang Shui

摘要

Porcelain fragment classification is crucial for cultural relic restoration. Traditional manual methods relying on macroscopic features struggle to balance accuracy and efficiency. This study proposes a multi-channel graph convolutional network (GCN) architecture integrated with multi-directional Gaussian filtering. First, images are converted into graph structures and processed with multi-directional Gaussian filtering to reduce noise while preserving texture details. The proposed multi-channel GCN extracts rich interconnected features from multiple perspectives. Experimental results achieved 93.33% accuracy, outperforming ResNet50 by 3.33% and DenseNet121 by 2.80%. This approach effectively addresses noise interference and uneven feature distribution in microscopic images.