Spatial information aided unmanned aerial vehicle video big data multi-temporal association analysis method

Abstract

The invention discloses a spatial information aided unmanned aerial vehicle big data multi-temporal association analysis method. The method comprises the following steps: acquiring and encoding a video in real time, acquiring unmanned aerial vehicle spatial information and camera parameter and other auxiliary information, and recording the auxiliary information in a compressed bit-stream at the same time; dividing the method into two conditions that the acquisition points are the same and different, establishing an associated judgement rule between any two multi-temporal images, and measuring an association coefficient; and organizing the multi-temporal videos under the same scene according to the order of the original acquisition time through the scene classification of the time sequence maintenance. By use of the analysis method disclosed by the invention, the multi-temporal video associated with the same scene can be fast and reliably positioned from mass unmanned aerial vehicle video big data of different scenes shot at different time, thereby providing evidence for the change detection of the disaster scene.

Description

technical field [0001] The invention belongs to the technical field of video analysis, and relates to a video analysis method of an unmanned aerial vehicle, in particular to a multi-time correlation analysis method for big data of unmanned aerial vehicle video assisted by spatial information. [0002] technical background [0003] In recent years, natural disasters have occurred frequently, posing serious threats to social economy and security. Low-altitude drones have quickly become an effective platform for quickly obtaining geographic data due to their advantages of flexibility, real-time performance, and low cost. The UAV disaster emergency measurement system quickly detects the disaster situation, grasps the spatial information of the damaged ground objects, can complete the tasks of geological disaster monitoring, emergency rescue and disaster assessment, and provides accurate basis for the rapid formulation of geological disaster prevention and rescue plans. [0004] ...

AI Technical Summary

Problems solved by technology

However, image changes before and after disasters are not only reflected in geometric changes and illumination changes, but more importantly, fundamental changes in foreground and background content information. There are still great limitations in the application of such algorithms to image registration in disaster scenes.

[0005] In short, since natural disasters have completely changed the geomorphology and content of the scene, the traditional method based on video content analysis is not suitable for the task of correlation analysis of multi-temporal videos captured at the same geographical location at different times

Images