Improved Land Cover Classification of VHR Optical Remote Sensing Imagery Based Upon Detail Injection Procedure

Development of very-high-resolution (VHR) remote sensing imaging platforms have resulted in a requirement for developing refined land cover classification maps for various applications. Therefore, aiming at exploring the accurate boundary and complex interior texture retrieval in VHR optical remote sensing images, a novel detail injection network (DI-Net) is proposed in this article, which is composed of three aspects. First, the decoupling refinement module embedded with a multiscale representation is designed to improve the feature extraction capabilities that precede the encoding-to-decoding process. Second, we pay attention to the hard examples of boundary and complex interior texture in land cover classification and design two detail injection attention modules to solve the feature inactivation phenomenon in gradually convolutional encoding-to-decoding process. Third, a specific stage grading loss is proposed to adaptively regulate the structural-level weights of the encoding and decoding stages, which facilitates the details retrieval and produce refined land cover classification results. Finally, various datasets [incl. International Society for Photogrammetry and Remote Sensing (ISPRS) and Gaofen Image Dataset (GID)] are employed to demonstrate that the proposed DI-Net achieves better performance than state-of-the-art methods. DI-Net provides more accurate boundaries and more consistent interior textures, and it achieves 86.86x0025; PA and 68.37x0025; mIoU on ISPRS dataset as well as 77.04x0025; PA and 64.38x0025; mIoU on GID dataset, respectively.

Automated summary

The proposed DI-Net in this article aims at accurate boundary and complex interior texture retrieval in VHR optical remote sensing images, achieving better performance than state-of-the-art methods on ISPRS and GID datasets. DI-Net provides more accurate boundaries and consistent interior textures, achieving 86.86% PA and 68.37% mIoU on ISPRS dataset as well as 77.04% PA and 64.38% mIoU on GID dataset.