Multi-sensor remote sensing image matching method, device and system based on unmanned aerial vehicle

Abstract

The invention relates to the field of unmanned aerial vehicle technology, in particular to a multi-sensor remote sensing image matching method and system based on an unmanned aerial vehicle. The multi-sensor remote sensing image matching system based on the unmanned aerial vehicle comprises the unmanned aerial vehicle, a server and multiple image collection sensors carried on the unmanned aerial vehicle, wherein the server can be used for realizing a single-lens oblique photography measuring method based on the unmanned aerial vehicle. The unmanned aerial vehicle is used for carrying the image collection sensors of different types to perform image collection work, and synchronous images obtained after projection distortion processing are registered to obtain registered images generated after multiple kinds of image information of different types are superposed. In this way, image information details can be increased, the definition of the images output by the server can be raised, and therefore the quality of images acquired through a single image collection sensor is improved. The single-lens oblique photography measuring method based on the unmanned aerial vehicle can be implemented through a computer program stored in a computer readable storage medium by establishing functional modules and combining the functional modules into a functional module framework.

Description

technical field [0001] The invention relates to the technical field of unmanned aerial vehicles, in particular to a multi-sensing remote sensing image matching method and system based on unmanned aerial vehicles. For the UAV-based multi-sensing remote sensing image matching method, functional modules can be established and combined into a functional module framework, which can be implemented by a computer program stored in a computer-readable storage medium. Background technique [0002] With the development of image acquisition sensor technology, image acquisition sources are no longer limited to optical cameras, and high-performance image acquisition sensors such as visible light digital cameras, multispectral cameras, thermal infrared cameras, color infrared photography, and panoramic cameras are widely used in image understanding. , computer vision and remote sensing. [0003] Affected by factors such as the principle of sensing imaging, the working environment of the i...

AI Technical Summary

Problems solved by technology

[0003] Affected by factors such as the principle of sensing imaging, the working environment of the image acquisition sensor, and the state of the image target, different image acquisition sensors have different capabilities to acquire image information. Visible light digital cameras have high imaging resolution, and the acquired image information contains rich contrast, Color and shape information, but the visible light digital camera is easily affected by the ambient light level when working, and cannot work at night; the multi-spectral camera uses multiple spectral bands to measure the target at the same time, which is suitable for terrain mapping, detection and analysis, but the acquired image The imaging quality of the information is not high; the thermal infrared camera has a certain ability to penetrate obstacles such as smoke, fog, snow, etc., and the imaging resolution is high, but the image information obtained when the shooting distance is long is not stable enough

[0004] Considering the differences in the acquisition of image information by various image acquisition sensors, when using image acquisition platforms such as drones to acquire images of ground objects, some image acquisition personnel will carry multiple different types of image acquisition sensors on the drone at the same time. However, due to the different shooting positions and angles of different image acquisition sensors, the three-dimensional geographic coordinates of each image are also different, resulting in differences in the content and size of the images collected by multiple image acquisition sensors, and it is difficult to achieve accurate matching between the acquired images