A method for automatic recognition of new buildings in high-resolution remote sensing images

Abstract

The invention relates to a method for automatic recognition of newly added buildings in high-resolution remote sensing images. The main steps include: 1) performing precise geometric correction and relative radiation correction on high-resolution images of different time phases; 2) calculating and acquiring two time phases before and after MBI (Morphological Building Index), Harris, NDVI (Normalized Difference Vegetation Index) features of the image; 3) remove the vegetation area from the architectural features; 4) calculate the feature heat map; 5) take the threshold of the feature heat map; The valued feature map takes the connected domain to get the object, and uses the object voting process to get the new building patch; 7) the post-processing of the new building, including: the current state map of the construction land in the early stage and the minimum mapping unit filtering. The invention can automatically extract new buildings based on remote sensing images, and provides effective technical means for land monitoring and management.

Description

technical field [0001] The invention belongs to the technical field of remote sensing image processing, and mainly relates to a new building change detection method in land use change detection, in particular to an automatic recognition method for new buildings in high-resolution remote sensing images based on MBI (morphological building index) and Harris . Background technique [0002] Since the expansion of construction land needs to occupy other non-construction areas (such as: cultivated land, forest land, bare land, etc.), land use changes frequently, and the original land survey method in my country, which mainly relies on a large amount of manpower to visually interpret images, can no longer meet the new situation. Under the requirements of land management. The rapid development of remote sensing technology and computer technology has become an important means to monitor land use and find illegal buildings. With the deepening of land and resources management, it is r...

AI Technical Summary

Problems solved by technology

However, the features used in the existing methods are relatively single, and cannot fully express the characteristics of buildings. The integrity and accuracy of identifying buildings are low, and there are many noises. Secondly, the existing supervised classification methods still need to select training Samples, more manual intervention, low degree of automation

Images