Virtual re-imaging image generation method

Abstract

The embodiment of the present invention relates to a method for generating a virtual re-imaged image, comprising: calculating the geometric positioning parameters of the virtual re-imaged image according to the relevant parameters of the original image; performing grayscale projection on each original pixel point in the original image and calculating the relevant parameters of the corresponding virtual pixel point; performing inverse distance weighted interpolation calculation on at least a portion of the virtual pixel points to obtain the target grayscale value of the corresponding virtual pixel point; and outputting the virtual re-imaged image according to the target grayscale value. The grayscale value of the virtual re-imaged image projection and the distance relationship between each pixel point in the virtual re-image and the nearest integer pixel point are weighted according to the original image, and the target grayscale value is calculated to correct the original grayscale value to at least partially solve the problem of missing pixels or image accuracy loss in the generated virtual re-imaged image due to rounding errors.

Description

Technical Field

[0001] The embodiments of the present invention relate to an optical satellite remote sensing image processing method, and more particularly to a virtual re-imaging image generation method suitable for optical satellite remote sensing image processing. Background Art

[0002] At present, the calibration of satellite remote sensing image sensors is generally completed by direct stitching of the image plane or virtual re-imaging technology. When using virtual re-imaging to calibrate satellite remote sensing image sensors, direct methods are often used for image generation due to computational efficiency considerations. In the process of using direct methods for image generation, the nearest neighbor pixel method is generally used for processing. However, the above processing methods will cause missing pixels or loss of image accuracy in the generated virtual re-imaging images due to rounding errors. Summary of the invention

[0003] In view of the existing technical problems, the present invention provides a virtual re-imaging image generation method for at least partially solving the above technical problems.

[0004] A virtual re-image generation method provided by the present invention includes: calculating the geometric positioning parameters of the virtual re-image according to the relevant parameters of the original image; performing grayscale projection on each original pixel point in the original image and calculating the relevant parameters of the corresponding virtual pixel point; performing inverse distance interpolation calculation on at least a part of the virtual pixel points to obtain the target grayscale value of the corresponding virtual pixel point; and outputting the virtual re-image according to the target grayscale value.

[0005] According to an embodiment of the present disclosure, the relevant parameters of the original image include: the satellite time, attitude, orbit file, precession, nutation, polar motion and leap second file, camera calibration file and digital elevation model file for generating the original image.

[0006] According to an embodiment of the present disclosure, the geometric positioning parameters of the virtual re-imaged image include rational polynomial coefficients.

[0007] According to an embodiment of the present disclosure, grayscale projection is performed on each original pixel point in the original image and relevant parameters of the corresponding virtual pixel point are calculated, including: selecting suitable calibration parameters based on the charge coupled device camera where the original pixel point is located and the corresponding information of the charge coupled camera; calculating the position information of the original pixel point based on the calibration parameters; calculating the position of the corresponding virtual pixel point on the virtual re-imaged image using the geometric positioning parameters of the virtual re-imaged image and the position information of the original pixel point; selecting a part of the virtual pixel points based on the position of the virtual pixel points; calculating the nearest integer pixel point of each virtual pixel point that meets the conditions; and recording the grayscale value of each virtual pixel point and the row and column pixel difference with the nearest integer pixel point as a data structure.

[0008] According to an embodiment of the present disclosure, the charge coupled device camera where the original pixel point is located and the corresponding information of the charge coupled device camera include the number of charge coupled devices in the charge coupled device camera and the number of pixels of each charge coupled device.

[0009] According to an embodiment of the present disclosure, the suitable calibration parameters represent the calibration parameters corresponding to the original pixel points.

[0010] According to an embodiment of the present disclosure, calculating the position information of the original pixel point includes calculating the longitude and latitude and the elevation value of the original pixel point through a mathematical elevation model.

[0011] According to an embodiment of the present disclosure, a portion of the virtual pixel points are selected based on the positions of the virtual pixel points, including presetting the output area of ​​the virtual re-imaging image, and selecting the virtual pixel points based on being located within the output area of ​​the virtual re-imaging image as a screening condition for the virtual pixel points.

[0012] According to an embodiment of the present disclosure, an inverse distance weighted interpolation calculation is performed on at least a portion of virtual pixel points to obtain a target grayscale value of the corresponding virtual pixel point, including: calculating the Euclidean distance between the virtual pixel point and the integer pixel point in the data structure of the integer pixel point position and four neighborhood points in the output area of ​​each virtual re-imaged image; calculating the weight value of each virtual pixel point based on the Euclidean distance; and calculating the target grayscale value based on the grayscale value and the weight value of each virtual pixel point; wherein the integer pixel point is characterized as a complete valid pixel encoded during the calculation process.

[0013] According to an embodiment of the present disclosure, outputting a virtual reimaged image according to the target grayscale value includes assigning the target grayscale value to each integer pixel point in the output area of ​​the virtual reimaged image and outputting the virtual reimaged image.

[0014] The present invention provides a method for generating a virtual reimage, which weights the grayscale value of the virtual reimage projection according to the original image and the distance relationship between each pixel point in the virtual reimage and the nearest integer pixel point, and calculates a target grayscale value for at least partially solving the problem of missing pixels or image accuracy loss in the generated virtual reimage due to rounding errors. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] Other objects and advantages of the present invention will become apparent from the following description of the present invention with reference to the accompanying drawings, which will help to provide a comprehensive understanding of the present invention.

[0016] Figure 1 The flowchart of the method for generating a virtual re-imaging image according to an embodiment of the present disclosure is schematically shown. DETAILED DESCRIPTION

[0017] In order to make the purpose, technical solution and advantages of the embodiment of the present invention clearer, the technical solution of the embodiment of the present invention will be clearly and completely described below in conjunction with the drawings of the embodiment of the present invention. Obviously, the described embodiment is a part of the embodiment of the present invention, not all of the embodiments. Based on the described embodiment of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0018] Unless otherwise defined, the technical terms or scientific terms used in the present invention shall have the usual meanings understood by those of ordinary skill in the art. The words "first", "second" and similar words used in the present invention do not indicate any order, quantity or importance, but are only used to distinguish different components. The words "include" or "comprise" and similar words mean that the elements or objects appearing before the word include the elements or objects listed after the word and their equivalents, without excluding other elements or objects.

[0019] Figure 1 The flowchart of the method for generating a virtual re-imaging image according to an embodiment of the present disclosure is schematically shown.

[0020] According to a virtual re-imaging image generation method provided by the present invention, Figure 1 As shown, it includes: calculating the geometric positioning parameters of the virtual re-imaged image according to the relevant parameters of the original image; performing grayscale projection on each original pixel point in the original image and calculating the relevant parameters of the corresponding virtual pixel point; performing inverse distance interpolation calculation on at least a part of the virtual pixel points to obtain the target grayscale value of the corresponding virtual pixel point; and outputting the virtual re-imaged image according to the target grayscale value.

[0021] According to an embodiment of the present disclosure, the virtual re-imaging image generation method further includes allocating memory space for storing the data structure of each virtual pixel.

[0022] Detailed, memory space that can be allocated based on empirical values.

[0023] Specifically, the data structure includes but is not limited to grayscale values ​​and distance differences between the virtual pixel and the nearest integer pixel row in the output area of ​​the virtual re-image (or other ways of reflecting the distance difference, such as coordinate difference).

[0024] According to an embodiment of the present disclosure, the relevant parameters of the original image include but are not limited to the satellite time, attitude, orbit file, precession nutation polar motion leap second file, camera calibration file and mathematical elevation model (DEM) file used to generate the original image.

[0025] According to an embodiment of the present disclosure, the geometric positioning parameters of the virtual re-imaged image include but are not limited to rational polynomial coefficients (RPC).

[0026] According to an embodiment of the present disclosure, grayscale projection is performed on each original pixel in the original image and relevant parameters of the corresponding virtual pixel are calculated, including: selecting suitable calibration parameters according to the charge coupled device (CCD) camera where the original pixel is located and the corresponding information of the charge coupled device (CCD) camera; calculating the position information of the original pixel according to the calibration parameters and the orbit, attitude, time, precession, nutation, polar shift, leap second and digital elevation model; calculating the position of the corresponding virtual pixel on the virtual re-imaged image using the geometric positioning parameters of the virtual re-imaged image and the position information of the original pixel; selecting a part of the virtual pixel according to the position of the virtual pixel; calculating the nearest integer pixel of each virtual pixel that meets the conditions; and recording the grayscale value of each virtual pixel and the row and column pixel difference with the nearest integer pixel as a data structure.

[0027] Specifically, the integer pixel is represented as a valid pixel encoded in the calculation process. Furthermore, the nearest integer pixel is represented as an integer pixel that is closest to the selected virtual pixel.

[0028] According to an embodiment of the present disclosure, the charge coupled device (CCD) camera where the original pixel is located and the corresponding information of the charge coupled device (CCD) camera include the number of charge coupled devices (CCDs) in the charge coupled device (CCD) camera and the number of pixels per charge coupled device.

[0029] According to an embodiment of the present disclosure, the suitable calibration parameters represent calibration parameters corresponding to the original pixel points.

[0030] According to an embodiment of the present disclosure, calculating the position information of the original pixel point includes calculating the longitude and latitude and elevation values ​​of the original pixel point based on calibration parameters and orbit, attitude, time, precession, nutation, polar motion, leap second and digital elevation model (DEM).

[0031] According to an embodiment of the present disclosure, a portion of virtual pixel points are selected based on their positions, including presetting an output area of ​​a virtual re-image, and selecting virtual pixel points based on the virtual pixel points being located within the output area of ​​the virtual re-image as a screening condition.

[0032] In detail, the output area of ​​the preset virtual reconstruction image can be based on empirical values ​​or can be customized.

[0033] According to an embodiment of the present disclosure, an inverse distance interpolation calculation is performed on at least a portion of virtual pixel points to obtain a target grayscale value of the corresponding virtual pixel point, including: calculating the Euclidean distance between the integer pixel point position in the output area of ​​each virtual re-imaged image and the virtual pixel point and the integer pixel point in the data structure recorded by the four neighborhood points; calculating the weight value of each virtual pixel point based on the Euclidean distance; and calculating the target grayscale value based on the grayscale value and weight value of each virtual pixel point.

[0034] In detail, integer pixels are represented as complete valid pixels that are encoded during the calculation process.

[0035] In detail, Euclidean distance is represented as Euclidean distance.

[0036] According to an embodiment of the present disclosure, the four neighborhood points are represented as virtual pixel points stored in a data structure that are adjacent to the selected integer pixel point above, below, left, and right. It should be understood that the protection scope of the present disclosure is not limited thereto.

[0037] For example, eight neighborhood points may be used for calculation, where the eight neighborhood points are represented as virtual pixel points stored in a data structure adjacent to the selected integer pixel point above, below, left, right, upper left, lower left, upper right, and lower right.

[0038] According to an embodiment of the present disclosure, the weight value of each virtual pixel is calculated according to the Euclidean distance, including the following formula (1):

[0039]

[0040] Among them, D i is the Euclidean distance from the pixel to the point; p is the exponent of the distance.

[0041] According to an embodiment of the present disclosure, the target grayscale value is calculated according to the Euclidean distance and weight value of each virtual pixel point, including the following formula (2):

[0042]

[0043] Among them, g target is the target gray value; g i is the original grayscale value of the virtual pixel.

[0044] According to an embodiment of the present disclosure, outputting a virtual reimaged image according to a target grayscale value includes assigning the target grayscale value to each integer pixel point in an output area of ​​the virtual reimaged image and outputting the virtual reimaged image.

[0045] In detail, the output virtual re-imaged image can be output according to the format required by the target image.

[0046] Although some embodiments according to the overall technical concept of the present invention have been shown and described, it will be appreciated by those skilled in the art that changes may be made to these embodiments without departing from the principles and spirit of the overall technical concept of the present invention, the scope of the present invention being defined by the claims and their equivalents.

Claims

1. A method for generating a virtual re-imaging image, comprising: Calculating geometric positioning parameters of the virtual re-imaged image based on relevant parameters of the original image; Perform grayscale projection on each original pixel in the original image and calculate the relevant parameters of the corresponding virtual pixel, including: Selecting appropriate calibration parameters according to the charge coupled device camera where the original pixel is located and corresponding information of the charge coupled device camera; Calculating the position information of the original pixel point according to the calibration parameters; Calculating the position of the corresponding virtual pixel point on the virtual re-imaged image using the geometric positioning parameters of the virtual re-imaged image and the position information of the original pixel point; Selecting a portion of the virtual pixel points according to the positions of the virtual pixel points; Calculate the nearest integer pixel point of each virtual pixel point that meets the conditions; Recording the grayscale value of each virtual pixel and the row and column pixel difference with the nearest integer pixel as a data structure; Performing inverse distance weighted interpolation calculation on at least a portion of virtual pixel points to obtain a target grayscale value of the corresponding virtual pixel point includes: Calculate the Euclidean distance between the virtual pixel point and the integer pixel point in the data structure of the four neighborhood point records in the output area of ​​each virtual re-imaged image; Calculating a weight value of each virtual pixel point according to the Euclidean distance; and Calculating the target grayscale value according to the grayscale value and the weight value of each virtual pixel; Among them, the integer pixel is characterized as a complete valid pixel encoded in the calculation process; The virtual re-imaged image is outputted according to the target grayscale value.

2. The image generation method according to claim 1, wherein: The relevant parameters of the original image include: the satellite time, attitude, orbit file, precession, nutation, polar motion and leap second file, camera calibration file and digital elevation model file used to generate the original image.

3. The image generation method according to claim 1, wherein: The geometric positioning parameters of the virtual re-imaged image include rational polynomial coefficients.

4. The image generation method according to claim 1, wherein: The charge coupled device camera where the original pixel point is located and the corresponding information of the charge coupled device camera include the number of charge coupled devices in the charge coupled device camera and the number of pixels of each charge coupled device.

5. The image generation method according to claim 1, wherein: The suitable calibration parameters represent the calibration parameters corresponding to the original pixel points.

6. The image generation method according to claim 1, wherein: Calculating the location information of the original pixel point includes calculating the latitude, longitude and elevation value of the original pixel point through a mathematical elevation model.

7. The image generation method according to claim 1, wherein: Selecting a portion of the virtual pixel points according to the positions of the virtual pixel points includes presetting an output area of ​​the virtual re-imaged image, and selecting the virtual pixel points based on being located within the output area of ​​the virtual re-imaged image as a screening condition for the virtual pixel points.

8. The image generation method according to claim 1, wherein: Outputting the virtual reimaged image according to the target grayscale value includes assigning the target grayscale value to each integer pixel in the output area of ​​the virtual reimaged image and outputting the virtual reimaged image.

Images