Remote-Sensing Scene Classification via Multistage Self-Guided Separation Network

In recent years, remote-sensing scene classification is one of the research hotspots and has played an important role in the field of intelligent interpretation of remote-sensing data. However, various complex objects and backgrounds form a variety of remote-sensing scenes through spatial combination and correlation, which brings great challenges to accurately classifying different scenes. Among them, the insufficient feature difference brought about the unbalanced change of background and target between interclass samples and the feature representation inconsistency caused by the difference of representation among the intraclass samples have become obstacles to effectively distinguishing different scene images. To address these issues, a multistage self-guided separation network (MGSNet) is proposed for remote-sensing scene classification. First of all, different from the previous work, it attempts to utilize the background information outside the effective target in the image as a decision aid through a target–background separation strategy to improve the distinguishability between target similarity-background difference samples. In addition, the diversity of feature concerns among different network branches is expanded through contrastive regularization (CR) to improve the separation of target–background information. Additionally, a self-guided network is proposed to find common features between intraclass samples and improve the consistency of feature representation. It combines the texture and morphological features of images to guide feature learning, effectively reducing the impact of intraclass differences. Extensive experimental results on three benchmarks demonstrate that MGSNet can achieve better classification performance compared to the state-of-the-art approaches.