Combination of deep learning and vegetation index for coastal wetland mapping using GF-2 remote sensing images

The biomass and growth of coastal wetland vegetation vary greatly due to different water and salt conditions in the growing area, and the spectral features of certain vegetation at the peak biomass are highly similar, making it easy for coastal wetland vegetation to be misclassified. In response to this problem, this study proposes a new semantic segmentation network called MFVNet to be combined with vegetation index for the fine mapping of coastal wetlands. In the proposed MFVNet, an Enhanced Multiscale Feature Extraction (E-MFE) module was first constructed on the basis of atrous convolution and attention mechanism to capture features of different scales adaptively. Then, the E-MFE module was used to replace the double convolution operations in traditional encoder-decoder network architecture, such as UNet. It was also used to merge the semantic features and detailed features of different resolutions to enhance feature representation. Finally, some typical vegetation indices were selected and input into the proposed MFVNet to improve the ability of coastal wetland fine mapping. The experiments of this study were conducted using GF-2 remote sensing images to study the coastal wetlands of the Yellow River Estuary. Experimental results indicated that the proposed MFVNet achieved good performance with an overall accuracy of 93.89% and a Kappa coefficient of 0.9072. On typical vegetation, such as reeds, spartina alterniflora, tamarix mixed area, and seagrass beds in the Yellow River Estuary, the F1 scores of MFVNet were 0.91, 0.87, 0.82, and 0.76, respectively, which were better than that of other methods. Moreover, ablation experiments showed that the combination of the E-MFE module and the vegetation index can increase the overall accuracy from 91.46% to 93.89%. (1) Compared with deep semantic segmentation networks, such as HRNet, MFVNet can more effectively extract vegetation information of coastal wetlands. (2) The proposed EMFE module can adaptively capture features of different scales and improve the overall accuracy, which justified its effectiveness in coastal wetland mapping. (3) The inclusion of vegetation index can enhance the spectral features of coastal wetland vegetation and improve the accuracy of vegetation information extraction, indicating the importance of vegetation index in coastal wetland mapping. (4) Simultaneously splicing modified soil adjusted vegetation index, difference vegetation index, and ratio vegetation index in remote sensing images contributed the most to the extraction of coastal wetland information. © 2023 National Remote Sensing Bulletin. All rights reserved.