A building contour automatic extraction algorithm based on a convolutional neural network and polygon regularization

Abstract

The invention discloses an automatic building contour extraction algorithm based on a convolutional neural network and polygon regularization. The automatic building contour extraction algorithm comprises the following steps: constructing a sample library according to an existing image and a building coverage vector file; Constructing a multi-scale fusion full convolutional neural network, training the multi-scale fusion full convolutional neural network through a sample library, and predicting the remote sensing image by using the trained network model to obtain a segmentation result coveredby the surface building of the remote sensing image; Performing building edge initialization based on the building semantic segmentation result, and obtaining an initial vector polygon; Removing wrongpolygons and wrong edges and nodes of the polygons by using a coarse adjustment algorithm; conducting regularization on the vector polygons through a regularization algorithm, and obtaining regular building vector edges. According to the method, the multi-scale fusion full convolutional neural network is high in scale robustness, the regularization algorithm can adapt to vector edges under various conditions, and the workload of manually drawing building edges is greatly reduced.

Description

technical field [0001] The invention relates to a deep learning method for extracting buildings from remote sensing images and a regularization algorithm for polygonal contours of buildings, which can be used for building extraction from remote sensing images, building vector edge generation, building change detection, and the like. Background technique [0002] Automatic extraction of buildings from remote sensing images is of great significance in applications such as urban planning, population estimation, map making and updating. Traditionally, the main work of extracting buildings from aerial / aerospace images focuses on: empirically designing an appropriate feature to express "what is a building", and creating corresponding features for automatic recognition and extraction of buildings. Commonly used metrics include pixels, spectra, lengths, edges, shapes, textures, shadows, heights, semantics, and more. However, these indicators will change significantly with seasons, ...

AI Technical Summary

Problems solved by technology

Therefore, this method of empirically designing features can often only deal with specific data, and cannot be truly automated

Images