Image mosaicking method and system for surface features of remote sensing images

Abstract

The invention relates to an image mosaicking method and system for surface features of remote sensing images. The method comprises the following steps of: S1, identifying and segmenting all the surface features in a remote sensing image through a full-convolutional network, so as to obtain segmented images of all the surface features in the remote sensing image; S2, judging whether images of high buildings exist in mosaicking areas of the segmented images; and S3, when the fact that high buildings exist is determined, moving mosaicking lines in the mosaicking areas to ensure that the mosaicking lines are not superposed with the images of the high buildings, or selecting the identified image areas except the buildings to replace the images of the high buildings. The method and system have the beneficial effects that identification and segmentation are carried out on the surface features in the remote sensing image, the high buildings are kept away in the mosaicking process, so that the condition that one building swings to a plurality of buildings in different directions can be avoided.

Description

technical field [0001] The invention relates to the technical field of remote sensing image processing, in particular to an image mosaic method and system for ground objects of remote sensing images. Background technique [0002] In the field of remote sensing technology, the image stitching technology can be used to stitch the captured ground images into a more accurate and complete image as the basis for further processing. very important role. The remote sensing images are shot at high altitude, and there are many tall buildings in the images. The high-rise buildings taken from different shooting angles have different shapes. When splicing, it is very likely that one building will become multiple buildings falling in different directions. Obviously, this way The result does not meet the needs of users. Contents of the invention [0003] The technical problem to be solved by the present invention is: the high-rise buildings in the remote sensing images have different s...

AI Technical Summary

Problems solved by technology

The remote sensing images are shot at high altitude, and there are many tall buildings in the images. The high-rise buildings are different in shape from different shooting angles. When splicing, it is very likely that one building will become multiple buildings falling in different directions. Obviously, this The result does not meet the needs of users

Images