Automatic extraction of impervious surfaces from high resolution remote sensing images based on deep learning | Semantic Scholar

A lightweight semantic segmentation network model based on CNN, and it is named LWIBNet, achieves a bit higher segmentation accuracy with less computation cost, and its computation speed is faster.

A self-trained deep-forest (STDF)-based ISA extraction method is proposed which exploits the complementary information contained in multispectral and polarimetric synthetic aperture radar (PolSAR) images using limited numbers of samples.

The experimental results show that the deep learning technique has been implemented to investigate the extraction of impervious surfaces in Marrakesh city, based on Landsat images and it is shown that this method is efficient and precise for mapping the impervious surface of Marrakeesh city.

A semantic network model-guided extraction method for HSRRSI impervious surfaces in rural–urban fringes is proposed and results show that the highest impervious surface extraction accuracy of the SVM classifiers is obtained when the segmentation scale is at 210 and 215.

The water body recognition method is applied to remote sensing images by combining the deep convolution generation network and the combined features, which has the advantage of high recognition accuracy.

Based on the VGG-16 network, this paper proposes a target recognition network with deep fully convolutional network, and uses the extracted feature map as the input of thetarget recognition network to realize the target recognition of the remote sensing image.

A block-based processing strategy is adopted and a majority voting method based on a grid offset is adopted to achieve a refined extraction of built-up areas in HR images and achieves a higher detection accuracy and preserves better shape integrity and boundary smoothness in the extracted results.

This paper examines current practices, problems, and trends in satellite image processing in automated land cover classification to provide comprehensive guidance for subsequent research direction.