Publication

Abstract : Fine-grained classification is one among the major research hotspots in remote sensing (RS) image interpretation that has gained attention span for several applications. RS image classification at the finest level is a highly challenging task to perform as it should consider the presence of high intraclass similarity, subtle interclass differences, and spatial scale variations. To address this issue, we propose an end-to-end saliency-guided feature mining network (SGFM-Net) specifically crafted to extract essential features that enable differentiation of highly similar instances. The essential components in the proposed SGFM-Net framework include: 1) a modified ConvNeXt as the backbone architecture that integrates pyramid convolutions (PyConvs) to extract the multi-scale features; 2) a saliency-guided feature mining module; 3) a multiple hierarchies attention module (MHAM) to filter and enhance intrinsic features; 4) a bilinear polymerization pooling (BPP) to fuse intrinsic and attention maps into a discriminative vector; and 5) a feature mapping network (FMN) to decouple and normalize that vector before classification. Experimental results carried out on three benchmark archives: FGSC-23, Aircraft-16, and FGSCR-42 have shown remarkable improvements over existing state-of-the-art methods.