Research on Photographic Image Classification Based on Multi-model Fusion and Data Augmentation

The study adopted a multi model fusion approach to improve image classification accuracy, using two classic convolutional neural network models, AlexNet and VGGNet, for training. A data augmentation method based on optimized classification was also proposed. This method improves the classification performance of the network model by adding input samples for categories with poor classification performance. The experimental results show that the accuracy of the fusion model on a single dataset reaches 0.95. Compared with other models, the fusion model exhibits higher classification accuracy, stronger classification ability, and higher AUC value. In summary, data augmentation methods and multi model fusion can effectively improve the accuracy of image classification. The research results indicate that the application of fusion models and appropriate data augmentation strategies in power terminal security protection models can improve classification accuracy and effectiveness.