A high-efficiency processing system for UAV images

Abstract

The invention provides an unmanned aerial vehicle (UAV) image efficient processing system. Key technologies, such as the image compression encoding technology and the network transmission technology, involved in UAV image acquisition and transmission, are improved. The UAV image stitching fusion method is improved. A UAV aerial image acquisition, transmission and efficient processing system is realized. The system achieves functions such as UAV aerial image acquisition, compression encoding and real-time transmission through programming based on Android on a mobile phone. At a server side, a transmitted video is decoded, played and stored. The degree of overlap between stored images can be judged, and the judgment result can guide image thinning. After thinning, part of the images are selected and stitched. The quality of stitching is ensured, and the speed of region stitching is increased. The system can meet the requirement of real-time acquisition and transmission, can realize UAV image stitching and processing without control points, and can achieve a good visual effect.

Description

technical field [0001] The invention relates to a UAV image processing system, in particular to an efficient acquisition, transmission and processing system of UAV images, and belongs to the technical field of UAV photogrammetry. Background technique [0002] UAV remote sensing system refers to a system that uses advanced unmanned aerial vehicle technology, remote sensing sensor technology, telemetry and remote control technology, communication technology, GPS differential positioning technology and remote sensing application technology to automatically, intelligently and quickly obtain ground remote sensing information. The UAV remote sensing system makes up for the shortcomings of the development of low-altitude and high-resolution remote sensing, and becomes an important means of real-time acquisition and processing of low-altitude and high-resolution remote sensing data. It has a certain practical value in the fields of remote sensing disaster monitoring, natural disaste...

AI Technical Summary

Problems solved by technology

Finally, the overlapping areas of the image stitching are processed. The direct averaging method is simple to calculate, but cannot achieve a smooth transition of the image. The wavelet transform algorithm itself has high complexity and takes a long time to calculate, which cannot meet the needs of high-efficiency applications.

The smooth transition of the two images cannot be achieved, and the ideal fusion effect cannot be obtained

[0008] The third is that there are many theoretical researches on existing technologies, and there is no design and implementation of a complete UAV aerial photography image acquisition, transmission and efficient processing system, and the use of specific hardware and software to carry out the functions of the UAV image efficient processing system Testing and Validation

The functions of acquisition, compression and encoding, and real-time transmission of drone aerial images were not realized in the programming environment, and the decoding, playback and storage of the transmitted video were not realized on the server side. At the same time, the judgment of the overlapping degree of the stored images was tested specifically. There is no basis Judgment results guide the thinning of images, and the image stitching effect is poor. The system cannot meet the requirements of real-time acquisition and transmission, and cannot realize the stitching and processing of UAV images without control points, and cannot obtain better visual effects.

Images