Jelly effect repairing method and system in aerial photography scene based on artificial intelligence

Abstract

The invention relates to the technical field of artificial intelligence, in particular to a jelly effect repairing method and system in an aerial photography scene based on artificial intelligence. The method comprises the following steps: acquiring a first image and a first depth image of the unmanned aerial vehicle in a hovering state; analyzing the correlation of the first image and the first depth image line by line, and collecting a second depth image when a pixel line with abnormal correlation appears; obtaining an abnormal region and an abnormal region boundary line according to the first depth image and the second depth image; removing or filling pixels in the first image according to the abnormal type of the abnormal region to generate a second image; respectively cutting the second image and the first depth image according to the abnormal pixel lines and the abnormal region boundary lines to obtain a first cut image and a second cut image; and processing the first cut image and the second cut image through a spatial transformation network to obtain a repair feature matrix, and eliminating the jelly effect of the first image. The jelly effect is repaired for different abnormal types, and the repairing precision and efficiency are improved.

Description

technical field [0001] The invention relates to the technical field of artificial intelligence, in particular to an artificial intelligence-based method and system for repairing the jelly effect in an aerial photography scene. Background technique [0002] When using UAVs for aerial photography, in order to obtain clear images, high-definition cameras with rolling shutters are often deployed on UAVs. However, since the UAV is a device with high-speed motion capability, and there will be vibrations of different degrees and frequencies during aerial photography, these factors make an image taken by the rolling shutter have areas with different degrees of jelly effect, so that The image appears to be pulled, twisted, wobbly, etc. Repairing the jelly effect image can improve the efficiency of aerial photography, but the restoration of the entire image often does not take into account the situation of different areas of the jelly effect, and cannot achieve accurate and effective...

AI Technical Summary

Problems solved by technology

However, since the UAV is a device with high-speed motion capability, and there will be vibrations of different degrees and frequencies during aerial photography, these factors make an image taken by the rolling shutter have areas with different degrees of jelly effect, so that The image appears to be pulled, twisted, wobbly, etc.

Repairing the jelly effect image can improve the efficiency of aerial photography, but the repair of the entire image often does not take into account the situation of different areas of the jelly effect, and cannot achieve accurate and effective jelly effect repair

Images