ResNet-50-NTS digital painting image style classification based on Three-Branch convolutional attention

Abstract

Addressing the difficulties and challenges faced by current traditional digital painting image style classification methods, the study enhances the residual neural network model by incorporating a three-branch convolutional attention mechanism. Furthermore, it integrates the improved residual neural network model with a fine-grained image classification model, ultimately presenting a novel approach for digital painting image style classification. The experimental results show that the final model can reach 100%, 98.61%, and 99.31% for the image classification precision, recall, and F1 value of ancient Greek pottery style, respectively. The improved residual neural network model proposed in this study has significant advantages in the task of digital painting image style classification, and can provide an efficient and reliable solution for classifying and recognizing digital painting image styles.