And realizing building instance mask extraction method of remote sensing image based on U-shaped CNN model

Abstract

The invention provides a building mask extraction method for realizing a remote sensing image based on a multi-scale and multi-task U-shaped CNN network model. Accurate building mask extraction of a high-resolution remote sensing image under different scales is realized in an instance segmentation mode. According to the building mask extraction method provided by the invention, a multi-scale multi-task network integrating a U-shaped network, a region suggestion attention network and edge constraint loss is applied, and the whole network is quickly converged by optimizing multi-task mixed lossof a joint ECL and the like, so that excessive compensation in limited available training data is inhibited. According to the method, excellent performance and high robustness are achieved on different scales, and the precision of various data sets is remarkably improved.

Description

technical field [0001] The present invention relates to the technical field of building instance extraction from high-resolution remote sensing images, and more specifically, to a method for extracting building instance masks from remote sensing images based on a multi-scale and multi-task U-shaped CNN model. Background technique [0002] Abstract buildings extracted from high-resolution remote sensing images have been widely used for automatic surveying and mapping through deep learning. It remains an important and challenging task due to the following issues, on the one hand, due to the inability of current semantic segmentation methods to efficiently extract individual building masks from cadastral databases with variable scales; on the other hand, Existing fully convolutional networks struggle to obtain accurate masks using limited available training sets when the convolutional layers become deeper. Contents of the invention [0003] The technical problem to be solved...

AI Technical Summary

Problems solved by technology

It remains an important and challenging task due to the following issues, on the one hand, due to the inability of current semantic segmentation methods to efficiently extract individual building masks from cadastral databases with variable scales; on the other hand, Existing fully convolutional networks struggle to obtain accurate masks using the limited available training set as the convolutional layers become deeper

Images