Building regular contour extraction method and system based on deep learning

Abstract

The invention discloses a building regular contour extraction method and system based on deep learning, and the method comprises the steps: obtaining a corresponding initial building contour according to a preset obtaining rule and a first remote sensing image, and obtaining a corresponding minimum enclosing rectangle according to the initial building contour; respectively calculating a first area of the initial building contour and a second area of the minimum enclosing rectangle to obtain a first ratio between the first area and the second area; when the first ratio is greater than or equal to a first preset value, determining the minimum enclosing rectangle as a final extraction result; and when the first ratio is smaller than a first preset value, dividing a plurality of to-be-detected areas according to the intersection point of the initial building contour and the minimum enclosing rectangle, and determining whether to reject the maximum inscribed rectangle corresponding to each to-be-detected area on the basis of the minimum enclosing rectangle. According to the method, the initial building contour is optimized based on the algorithm of the minimum enclosing rectangle and the maximum inscribed rectangle, so that the building contour regularization extraction precision is improved.

Description

technical field [0001] The invention relates to the technical field of image processing, and in particular, to a method and system for extracting regular outlines of buildings based on deep learning. Background technique [0002] With the rapid progress of image sensor technology, the difficulty of obtaining remote sensing images has gradually decreased, and the resolution of images has been continuously improved. Satellite remote sensing images can easily reach sub-meter ground resolution. In recent years, with the rapid development of my country's economy, the application scope of remote sensing images has also expanded, including land change detection, urban data update, disaster prevention and emergency response, urban planning, population estimation, topographic map production and update and other aspects. Therefore, the use of remote sensing images to identify and extract buildings has become a hot spot in remote sensing scientific work in recent years. [0003] The t...

AI Technical Summary

Problems solved by technology

However, traditional building extraction algorithms are easily affected by various factors such as seasonal weather changes, sensor quality, and building styles, which lead to insufficient extraction accuracy and are difficult to apply in real life

Therefore, in real life, most of the building extraction methods still use manual labeling, which consumes a lot of manpower and financial resources, and the time cost is high and the work efficiency is low.

Images