Research on bronze wine vessel classification using improved SSA-CBAM-GNNs.

Area of Science:

• Archaeometry
• Computer Science
• Materials Science

Background:

• Accurate classification of ancient bronze drinking utensils is crucial for understanding historical dynasties and cultural contexts.
• Traditional classification methods face challenges due to the complexity of cultural characteristics and variations across dynasties.
• Developing automated and precise identification techniques is essential for archaeological research.

Purpose of the Study:

• To propose an advanced classification algorithm for bronze drinking utensils.
• To enhance the accuracy and efficiency of identifying and classifying cultural features.
• To address the challenges in dynasty classification based on artifact analysis.

Main Methods:

• Integration of the Sparrow Search Algorithm (SSA) for network optimization and accelerated convergence.
• Utilization of Convolutional Block Attention Module (CBAM) for optimizing feature extraction weights in Graph Neural Networks (GNNs).
• Development of a novel SSA-CBAM-GNNs algorithm combining SSA, CBAM, and GNNs for artifact classification.

Main Results:

• The SSA-CBAM-GNNs algorithm demonstrated outstanding performance in accurately identifying and classifying cultural features of bronze drinking utensils.
• Experimental results validated through various performance indicators confirmed the algorithm's effectiveness.
• Comparative experiments showed the superiority of the proposed algorithm over existing methods.

Conclusions:

• The study successfully developed a highly efficient identification and classification algorithm for bronze drinking utensils.
• The SSA-CBAM-GNNs algorithm effectively extracts and identifies crucial cultural features, aiding in precise historical classification.
• The proposed method offers a significant advancement in the field of archaeometric artifact analysis.