CN114252059A - Method for manufacturing large scene of aerospace and aviation integrated satellite - Google Patents

Abstract

The invention discloses a method for manufacturing a large scene of an integrated aerospace satellite. The existing aviation large-scene stereo imaging method is specific to a single aviation platform, space platform data is not considered, and a more general remote sensing large-scene stereo generation and measurement method is not provided. According to the method, remote sensing image data are obtained according to the basic situation of a measuring area, air-to-three calculation is carried out, and a standardized aerospace and aviation integrated basic data engineering file is formed; carrying out color and light evening processing on the image data to generate DOM data; and finally, generating a large-scene three-dimensional model, and acquiring the coordinates of the three-dimensional ground object points through three-dimensional measurement. The invention unifies the difference caused by the difference of the aerospace images, so that the data source and platform data do not need to be concerned in the large scene making and mapping, and the application field of the large scene is enlarged.

Description

Method for manufacturing large scene of aerospace and aviation integrated satellite

Technical Field

The invention belongs to the technical field of railway survey, and particularly relates to a method for manufacturing a large scene of an integrated aerospace satellite.

Background

The three-dimensional information is an important information basis in the surveying field and provides underlying data for an intelligent application scene and a high-level application platform, so that the technologies of acquiring, displaying, measuring and the like of the three-dimensional information are important space geographic information technologies.

The existing three-dimensional acquisition technology is divided into an active type and a passive type, wherein the active type comprises manual measurement and laser point cloud scanning, and the passive type mainly takes a remote sensing image as a main mode. The three-dimensional acquisition mode of the active mode has the disadvantages of high equipment cost, operation requiring the addition of prior knowledge and high engineering cost; the passive mode is that the camera carries out reconstruction operation after passive sensing, a large amount of work is compressed to the automatic reconstruction calculation of the interior industry, the field collection efficiency is high, the precision is controllable, the technical threshold is low, and the cost is low. Therefore, the photogrammetry method is adopted to carry out three-dimensional data space acquisition, which becomes the current main three-dimensional data acquisition means, however, the three-dimensional reconstruction has high calculation cost and long time, and the three-dimensional result is easily influenced by weak texture, matching error and environmental conditions, so that the three-dimensional model has wrinkles and distortion and low plane flatness. Therefore, the aviation large-scene stereo imaging technology is proposed.

The existing aviation large-scene stereo imaging method is specific to a single aviation platform, space platform data is not considered, and a more general remote sensing large-scene stereo generation and measurement method is not provided.

Disclosure of Invention

In order to make up the defects of the prior art, the invention provides a method for manufacturing a large scene of an aerospace integrated satellite,

in order to achieve the purpose, the technical scheme adopted by the invention is as follows:

a method for manufacturing a large scene of an integrated aerospace satellite comprises the following steps:

the method comprises the following steps: surveying the measured area according to the requirement of the project to obtain the basic condition of the measured area;

step two: according to the requirements of the measuring area items, combining the specific conditions of the measuring area to obtain remote sensing image data with proper resolution;

step three: performing space-three solution on the remote sensing image data;

step four: forming a standardized aerospace and aviation integrated basic data engineering file by data standardization;

step five: carrying out color and light homogenizing treatment on the standardized remote sensing image data;

step six: generating DOM data with specified resolution;

step seven: generating a stereo splicing line of an overlapping area between stereo pairs;

step eight: generating a large-scene three-dimensional model;

step nine: and obtaining the coordinates of the three-dimensional ground object points through three-dimensional measurement.

Specifically, the step eight includes:

8.1, according to the splicing area in the seventh step, combining DSM/DEM, carrying out digital differential correction to generate DOM data which is used as large-scene left stereoscopic image data;

8.2, shifting pixel points through a parallax generator, repeating the digital differential correction process, and generating the large-scene right stereoscopic image data.

The invention has the beneficial effects that:

1) the method unifies theoretical differences brought by aerospace image differences, so that data sources and platform data do not need to be concerned in large scene manufacturing and mapping, the large scene theory is generalized compared with the existing method, the application field of large scenes is expanded, and the requirements of large, medium and small-scale mapping are completely covered;

2) the invention provides an optional three-dimensional surveying means for surveying engineering, and selects the most suitable three-dimensional data sensing mode in a surveying area according to factors of requirements, resolution, range and cost.

Drawings

FIG. 1 is a flow chart of aerospace large scene generation;

FIG. 2 is a mapping relation diagram of an object space and an image space of a remote sensing image;

fig. 3 is a flow chart of a large-scene stereo measurement technique.

Detailed Description

The present invention will be described in detail with reference to specific embodiments.

The invention provides a space flight and aviation integrated three-dimensional scene generation method, which meets the requirements on heterogeneous data and completes the theoretical sublimation of a large scene, and specifically comprises the following steps:

the method comprises the following steps: surveying the survey area according to the requirements of the project, and acquiring basic conditions of the survey area, such as survey area range, survey area trend, survey area fluctuation and other data;

step two: according to the requirements of the measuring area items, combining the specific conditions of the measuring area to obtain remote sensing image data with proper resolution;

if the data is aviation data, airspace application and air route planning are required to be carried out in advance, and two control point laying modes in advance and in the rear can be selected; if the data is space data, clear data with small time span and few clouds needs to be picked out, and the mode of laying control points afterwards is generally selected for the future data. The control point layout should meet the relevant aerial survey specification requirements.

Step three: performing space-three solution on the remote sensing image data;

according to the images and control points acquired in the previous step, the remote sensing data is required to be subjected to space attitude calculation and geometric correction, and the mapping relation from the ground point to the image point is restored, as shown in formula 1:

wherein x is_c(j, i) represents the pixel coordinates of projection point i in image j, subscript c represents the image coordinate system, subscript o represents the object coordinate system, X_o(i) The point of the object space is represented,

the mapping relation from the object coordinate system to the image point is obtained.

The method can comprise mathematical mapping relations in an imaging process, including collinear equations, rational function models, camera distortion models and geographical projection transformation, as shown in figure 2. The space-three solution of the remote sensing image is known as x_c(j,i)，X_o(i) Solving the corresponding mapping relation for concrete expression.

Step four: forming a standardized aerospace and aviation integrated basic data engineering file by data standardization;

the standardization of data mainly solves the difference of basic data generated by different data sources and different platforms, and specifically comprises the gray process of solving a high-bit image through remapping of image information; extracting standardized interfaces through abstracting different mapping relations; dynamically accessing a geographical projection mapping transformation relation; and forming a standardized aerospace and aviation integrated basic data engineering file.

Step five: carrying out color and light homogenizing treatment on the standardized remote sensing image data;

the acquisition of the remote sensing image is usually accompanied by the influence of different sources of space and time, so that the remote sensing image has obvious color and tone difference, and therefore, the consistency expression of a large scene in a measuring area needs to be comprehensively considered, and the influence of the consistency expression on the three-dimensional expression of the large scene is eliminated.

Step six: generating DOM data with specified resolution;

generating a DOM with a specified resolution according to the mapping relation between the DSM and the images, wherein DOM data of each image is mainly used for generating a subsequent splicing line, so that the situation of ground object tearing is avoided;

step seven: generating a stereo splicing line of an overlapping area between stereo pairs;

step eight: generating a large-scene three-dimensional model;

8.1, according to the splicing area in the seventh step, combining DSM/DEM, carrying out digital differential correction to generate DOM data which is used as large-scene left stereoscopic image data;

8.2, shifting pixel points through a parallax generator, repeating the digital differential correction process, and generating the large-scene right stereoscopic image data.

The parallax generator may employ a reference altitude H, a ground point height Z, a base height Z of the stereopair₀The base length B is generated as shown in equation 2:

step nine: and obtaining the coordinates of the three-dimensional ground object points through three-dimensional measurement.

Stereo measurement is the process from the homonymous point of the left and right stereo images of a large scene to the homonymous point of the object, as shown in fig. 3. Firstly, reversely calculating the same-name points of the left image of the large scene to obtain original image points; then, inversely calculating original image points of the large-scene right stereoscopic image through an inverse parallax generator and a mapping relation; then, an overdetermined system of equations, such as a common equation, for the original image points of the same name is listed according to the mapping relation of formula 1

And (4) obtaining the coordinates of the three-dimensional ground object points as shown in the formula 3.

Application example: the method comprises the steps of using a world view-3 image of a certain area, generating a large-scene stereo model of the aerospace high-resolution satellite image through the method, and then pricking same-name points in left and right large-scene images in a stereo measurement mode, wherein the obtained same-name points are geographical projection coordinates. Then the left homonym point is mapped into the original left image, and the right image is projected into the original right image by combining the inverse parallax generator and the mapping function. Finally, by comparing the reprojection error, the equivalent medium error can reach within 1 pixel.

The invention is not limited to the examples, and any equivalent changes to the technical solution of the invention by a person skilled in the art after reading the description of the invention are covered by the claims of the invention.

Claims (2)

1. A method for manufacturing a large scene of an integrated aerospace satellite is characterized by comprising the following steps: the method comprises the following steps:

the method comprises the following steps: surveying the measured area according to the requirement of the project to obtain the basic condition of the measured area;

step two: according to the requirements of the measuring area items, combining the specific conditions of the measuring area to obtain remote sensing image data with proper resolution;

step three: performing space-three solution on the remote sensing image data;

step four: forming a standardized aerospace and aviation integrated basic data engineering file by data standardization;

step five: carrying out color and light homogenizing treatment on the standardized remote sensing image data;

step six: generating DOM data with specified resolution;

step seven: generating a stereo splicing line of an overlapping area between stereo pairs;

step eight: generating a large-scene three-dimensional model;

step nine: and obtaining the coordinates of the three-dimensional ground object points through three-dimensional measurement.

2. The aerospace integrated satellite large scene manufacturing method according to claim 1, wherein the method comprises the following steps: the eighth step comprises:

8.1, according to the splicing area in the seventh step, combining DSM/DEM, carrying out digital differential correction to generate DOM data which is used as large-scene left stereoscopic image data;

8.2, shifting pixel points through a parallax generator, repeating the digital differential correction process, and generating the large-scene right stereoscopic image data.

Images