Fine-grained image target recognition, also known as sub-category fine-grained recognition, is a research hotspot in the field of computer vision in recent years. It aims to classify targets in more detail and with high accuracy on the basis of base-class recognition. For example, ships are classified not only into civilian and military ships, but also military ships are further classified into aircraft carriers, destroyers, frigates, cruisers, etc. Similarly, civilian ships are further classified into subclasses such as cargo ships, fishing vessels, cruises, and barges. Compared with traditional target recognition, fine-grained target recognition is characterized by inter-class similarity and intra-class difference, which requires recognition algorithms to effectively extract fine-grained features and identify the minor feature differences between different targets in order to realize the fine-grained classification, thus increasing the recognition difficulty and making it more challenging.