A method, medium and system for correcting full fundus laser scanning images

Abstract

The present invention provides a full fundus laser scanning image correction method, medium and system, belonging to the field of image processing technology, the full fundus laser scanning image correction method comprises: establishing a target image and obtaining multiple scale full fundus images, the scale full fundus image includes a scale with a standard scale and the shooting angles of the scales in the multiple scale full fundus images are different; and respectively obtaining the scale coordinates of the standard scales with the same intervals on the scale in each scale full fundus image in a preset image coordinate system; performing nonlinear fitting according to the scale coordinates of all scale full fundus images and the values ​​of the standard scale to obtain a correction equation; determining the coordinate deviation of the pixel point on the target image according to the correction equation; and correcting the distance between the coordinates on the target image according to the obtained coordinate deviation. The present invention can solve the technical problems of needing to use a specific template for focusing, being only applicable to specific equipment, and having a complex implementation process.

Description

Technical Field

[0001] The present invention belongs to the field of image processing technology, and in particular, relates to a method, medium and system for correcting a full fundus laser scanning image. Background Art

[0002] In fundus image capture, full fundus laser scanning can obtain fundus tissue information in a wider range, and the captured full fundus image can provide more basis for the doctor's diagnosis; however, due to the large angle of the image capture, although the center of the full fundus image is flat, there is distortion around it, and the distortion is complex, and generally the distortion in the horizontal and vertical directions is inconsistent. For example, for patients with high myopia, the posterior pole image is deformed because the eyes are elongated and enlarged, so correction is required.

[0003] The Chinese invention patent with the authorization number CN105096329B (application number CN201510514238.0) discloses a method for accurately correcting the distortion of an ultra-wide-angle camera image. First, the correction error is calculated based on the distortion model, and then it is iteratively refined according to the descent algorithm. Finally, the distorted image is corrected according to the internal parameters of the camera lens and the distortion parameters obtained by the refinement. The present invention uses the parameters obtained by the lens calibration method to correct the distorted image. It is not only suitable for the distortion correction of ultra-wide-angle lenses, but also for ordinary lenses and wide-angle lenses with smaller distortion. The overall distortion correction of the image has a large angle and high accuracy.

[0004] The Chinese invention patent with the authorization number CN106600546B (application number CN201611008743.9) discloses a method and system for distortion correction of an ultra-wide-angle camera, including: according to the imaging model of the ultra-wide-angle camera, the original image captured by the ultra-wide-angle camera is projected onto a panoramic spherical surface to obtain a spherical panoramic image; with the center of the panoramic spherical surface as the starting point, the spherical panoramic image is projected onto a correction plane to obtain a distortion-corrected output image. The present invention projects the original image captured by the ultra-wide-angle camera onto a panoramic spherical surface so that an image without distortion can be seen from the center of the panoramic spherical surface, and then with the center of the panoramic spherical surface as the starting point, the spherical panoramic image is projected onto a correction plane to obtain a distortion-corrected output image.

[0005] The above two patents require the use of specific templates for focusing, but the specific templates are customized according to specific devices and are only applicable to specific devices, and the implementation process is complicated. Summary of the invention

[0006] In view of this, the present invention provides a method, medium and system for correcting the image of the whole fundus laser scanning, which can solve the technical problem that a specific template needs to be used for focusing, but the specific template is customized according to a specific device and is only suitable for the specific device, and the implementation process is complicated.

[0007] The present invention is achieved in that:

[0008] A first aspect of the present invention provides a method for correcting a full fundus laser scanning image, comprising the following steps:

[0009] S10: establishing a target image and acquiring a plurality of scale full fundus images, wherein the scale full fundus images include a scale with a standard scale and the shooting angles of the scales in the plurality of scale full fundus images are different from each other;

[0010] S20: respectively obtaining the scale coordinates of the standard scales with the same intervals on the ruler in each of the ruler full fundus images in a preset image coordinate system;

[0011] S30: performing nonlinear fitting according to the scale coordinates of all the scale full fundus images and the values ​​of the standard scale to obtain a correction equation;

[0012] S40: determining the coordinate deviation of the pixel point on the target image according to the correction equation;

[0013] S50: Correcting the distance between the coordinates on the target image according to the obtained coordinate deviation.

[0014] On the basis of the above technical solution, a method for correcting the whole fundus laser scanning image of the present invention can also be improved as follows:

[0015] The specific steps of step S20 are:

[0016] The first step: normalizing the scale coordinates of the plurality of scale full fundus images and the values ​​of the standard scale;

[0017] Step 2: Put the normalized scale coordinates and the standard scale values ​​into the preset cubic equation for fitting to obtain the cubic correction equation.

[0018] The preset cubic equation is expressed as follows:

[0019] y=ax 3 +bx 2 +cx

[0020] The cubic correction equation is expressed as follows:

[0021] vx=-2.48x j 3 -0.11x j 2 +4.25x j

[0022] vy=-1.78yj 3 -0.25y j 2 +3.62y j

[0023] Among them, vx and vy are the coordinates after correction, and xj and yj are the coordinates on the image;

[0024] Furthermore, before normalizing the scale coordinates of the plurality of scale full fundus images and the values ​​of the standard scale in step S20, the following steps are also included:

[0025] Step 201: Establishing a functional relationship between world coordinates and scale coordinates;

[0026] Step 202: Correcting the scale coordinates;

[0027] Step 203: establishing a functional correspondence between the scale coordinates and the image coordinate system;

[0028] Step 204: Use bilinear interpolation to convert the scale coordinates into integers.

[0029] 7. Furthermore, the functional relationship between the world coordinates and the scale coordinates in step S201 is:

[0030]

[0031] where t 3×1 is the translation vector, R 3×3 is the rotation matrix, [X c Y c Z c ] T is the camera coordinate, [X w Y w Z w ] T is the world coordinate.

[0032] The function formula used to correct the scale coordinates in step 202 is:

[0033]

[0034]

[0035] In the formula, and is the ideal value of the image point coordinates calculated by the pinhole model, (x, y) is the coordinates of the actual image point; m 2 =x 2 +y 2 , k1 and k2 are radial distortion coefficients, and p1 and p2 are tangential distortion coefficients.

[0036] The function corresponding formula of the scale coordinate and the image coordinate system in step 203 is:

[0037]

[0038] Among them, (u, v) represents the coordinates in the image coordinate system, the origin is located at the pixel point (u0, v0), k×l represents the size of a pixel, f is the focal length of the camera, and θ is the skewness of the camera coordinate system.

[0039] The formula for converting the scale coordinates into integers using bilinear interpolation in step 204 is:

[0040] f(x,y)=a1x+b1y+c1xy+d1.

[0041] Wherein, the normalization method is the cosine normalization method.

[0042] A second aspect of the present invention provides a computer-readable storage medium, wherein the computer-readable storage medium stores program codes for executing the above-mentioned full fundus laser scanning image correction method.

[0043] A third aspect of the present invention provides a full fundus laser scanning image correction system, which includes the above-mentioned computer-readable storage medium.

[0044] Compared with the prior art, the present invention provides a method, medium and system for correcting a full fundus laser scanning image, which has the following beneficial effects: by setting a ruler with a standard scale, obtaining multiple full fundus images of the ruler taken at different angles, marking the coordinates of the scales with the same intervals on the ruler on the obtained full fundus image, obtaining the corresponding relationship between the value of the scale in the image and the coordinates of the scale in the image, and obtaining the correction equation through nonlinear fitting; shooting the ruler image from various angles, so that the final correction equation integrates the degree of distortion in various directions on the image, and the determined target image coordinate deviation is more accurate, and the corresponding correction equation can be obtained by only shooting the ruler with a full fundus camera, and the calculation method is simple and efficient, and the coordinates of the pixels in the target image whose deviation values ​​need to be known are directly brought into the correction equation to obtain the estimated coordinate deviation value, and the deviation can be obtained intuitively, which is applied to assist doctors in judging the condition of fundus tissue, such as determining the actual cup-disc distance according to the cup-disc distance in the full fundus image. The present invention has a good effect in improving the accuracy of doctors' judgment. BRIEF DESCRIPTION OF THE DRAWINGS

[0045] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings required for use in the description of the embodiments of the present invention will be briefly introduced below. Obviously, the accompanying drawings in the following description are only some embodiments of the present invention. For ordinary technicians in this field, other accompanying drawings can be obtained based on these accompanying drawings without paying creative labor.

[0046] Figure 1 A flow chart of a method for correcting a full fundus laser scanning image provided by the present invention;

[0047] Figure 2 A flowchart of the steps before normalizing the scale coordinates of the plurality of scale full fundus images and the values ​​of the standard scale in step S20;

[0048] Figure 3 A schematic diagram of scale coordinates in a scaled full fundus image according to an embodiment of the present invention;

[0049] Figure 4 Schematic diagram of a scaled full fundus image at different shooting angles according to an embodiment of the present invention. DETAILED DESCRIPTION

[0050] In order to make the purpose, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments 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.

[0051] Therefore, the following detailed description of the embodiments of the present invention provided in the accompanying drawings is not intended to limit the scope of the invention claimed for protection, but merely represents selected embodiments of the present invention. Based on the embodiments 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.

[0052] It should be noted that similar reference numerals and letters denote similar items in the following drawings, and therefore, once an item is defined in one drawing, further definition and explanation thereof is not required in subsequent drawings.

[0053] In the description of the present invention, it should be understood that the terms "center", "longitudinal", "lateral", "length", "width", "thickness", "up", "down", "front", "back", "left", "right", "vertical", "horizontal", "top", "bottom", "inside", "outside", "clockwise", "counterclockwise" and the like indicate orientations or positional relationships based on the orientations or positional relationships shown in the accompanying drawings, and are only for the convenience of describing the present invention and simplifying the description, and do not indicate or imply that the referred device or element must have a specific orientation, be constructed and operated in a specific orientation, and therefore should not be understood as a limitation on the present invention.

[0054] In addition, the terms "first" and "second" are used for descriptive purposes only and should not be understood as indicating or implying relative importance or implicitly indicating the number of the indicated technical features. Therefore, the features defined as "first" and "second" may explicitly or implicitly include one or more of the features. In the description of the present invention, the meaning of "plurality" is two or more, unless otherwise clearly and specifically defined.

[0055] like Figure 1 FIG. 1 is a flowchart of a method for correcting a full fundus laser scanning image according to a first aspect of the present invention, wherein the method comprises the following steps:

[0056] S10: establishing a target image and acquiring a plurality of scaled full-eye fundus images, wherein the scaled full-eye fundus images include a scale with a standard scale and the shooting angles of the scales in the plurality of scaled full-eye+fundus images are different from each other;

[0057] S20: respectively obtaining the scale coordinates of the standard scales with the same intervals on the scale in each scale full fundus image in a preset image coordinate system;

[0058] S30: performing nonlinear fitting according to the scale coordinates of all scale full fundus images and the values ​​of the standard scale to obtain a correction equation;

[0059] S40: determining the coordinate deviation of the pixel point on the target image according to the correction equation;

[0060] S50: Correcting the distance between the coordinates on the target image according to the obtained coordinate deviation.

[0061] Among them, in the above technical solution, the specific steps of step S20 are:

[0062] Step 1: Normalize the scale coordinates of multiple scale full fundus images and the values ​​of the standard scale;

[0063] Step 2: Put the normalized scale coordinates and the standard scale values ​​into the preset cubic equation for fitting, and get the cubic correction equation, where the preset cubic equation is expressed as follows:

[0064] The cubic correction equation is expressed as follows:

[0065] vx=-2.48x j 3 -0.11x j 2 +4.25x j

[0066] vy=-1.78y j 3 -0.25y j 2 +3.62y j

[0067] Among them, vx and vy are the corrected coordinates, x j and j are the coordinates on the image;

[0068] Furthermore, in the above technical solution, before normalizing the scale coordinates of the plurality of scale full fundus images and the values ​​of the standard scale in step S20, the following steps are also included:

[0069] Step 201: Establishing a functional relationship between world coordinates and scale coordinates;

[0070] Step 202: Correcting the scale coordinates;

[0071] Step 203: establishing a functional correspondence between the scale coordinates and the image coordinate system;

[0072] Step 204: Use bilinear interpolation to convert the scale coordinates into integers.

[0073] 8. Further, in the above technical solution, the functional relationship between the world coordinates and the scale coordinates in step S201 is:

[0074]

[0075] where t 3×1 is the translation vector, R 3×3 is the rotation matrix, [X c Y c Z c ] T is the camera coordinate, [X w Y w Z w ] T is the world coordinate.

[0076] The function formula used to correct the scale coordinates in step 202 is:

[0077]

[0078]

[0079] In the formula, and is the ideal value of the image point coordinates calculated by the pinhole model, (x, y) is the coordinates of the actual image point; m 2 =x 2 +y 2 , k1 and k2 are radial distortion coefficients, and p1 and p2 are tangential distortion coefficients.

[0080] The function corresponding formula of the scale coordinate and the image coordinate system in step 203 is:

[0081]

[0082] Among them, (u, v) represents the coordinates in the image coordinate system, the origin is located at the pixel point (u0, v0), k×l represents the size of a pixel, f is the focal length of the camera, and θ is the skewness of the camera coordinate system.

[0083] The formula for converting the scale coordinates into integers using bilinear interpolation in step 204 is:

[0084] f(x,y)=a1x+b1y+c1xy+d1.

[0085] Among them, in the above technical solution, the normalization method is the cosine normalization method.

[0086] A second aspect of the present invention provides a computer-readable storage medium, wherein the computer-readable storage medium stores program codes for executing the above-mentioned full fundus laser scanning image correction method.

[0087] A third aspect of the present invention provides a full fundus laser scanning image correction system, which includes the above-mentioned computer-readable storage medium.

[0088] The above is only a specific embodiment of the present invention, but the protection scope of the present invention is not limited thereto. Any person skilled in the art can easily think of changes or substitutions within the technical scope disclosed by the present invention, which should be included in the protection scope of the present invention. Therefore, the protection scope of the present invention should be based on the protection scope of the claims.

Claims

1. A method for correcting a full fundus laser scanning image, characterized in that: The following steps are involved: S10: establishing a target image and acquiring a plurality of scale full fundus images, wherein the scale full fundus images include a scale with a standard scale and the shooting angles of the scales in the plurality of scale full fundus images are different from each other; S20: respectively obtaining the scale coordinates of the standard scales with the same intervals on the ruler in each of the ruler full fundus images in a preset image coordinate system; S30: performing nonlinear fitting according to the scale coordinates of all the scale full fundus images and the values ​​of the standard scale to obtain a correction equation; S40: determining the coordinate deviation of the pixel point on the target image according to the correction equation; S50: Correcting the distance between the coordinates on the target image according to the obtained coordinate deviation; The specific steps of step S30 are: The first step: normalizing the scale coordinates of the plurality of scale full fundus images and the values ​​of the standard scale; Step 2: Put the normalized scale coordinates and the standard scale values ​​into the preset cubic equation for fitting to obtain the cubic correction equation. The preset cubic equation is expressed as follows: ； The cubic correction equation is expressed as follows: ； ； Among them, vx and vy are the coordinates after correction, and xj and yj are the coordinates on the image; Wherein, before normalizing the scale coordinates of the plurality of scale full fundus images and the values ​​of the standard scale in step S20, the following steps are also included: Step 201: Establishing a functional relationship between world coordinates and scale coordinates; Step 202: Correcting the scale coordinates; Step 203: establishing a functional correspondence between the scale coordinates and the image coordinate system; Step 204: Use bilinear interpolation to convert the scale coordinates into integers.

2. The method for correcting the whole fundus laser scanning image according to claim 1, characterized in that: The functional relationship between the world coordinates and the scale coordinates in step S201 is: ， where t 3×1 is the translation vector, R 3×3 is the rotation matrix, [X c Y c Z c ] T is the camera coordinate, [X w Y w Z w ] T is the world coordinate.

3. The method for correcting the whole fundus laser scanning image according to claim 1, characterized in that: The function formula used to correct the scale coordinates in step 202 is: x(1 + k1m 2 + k2m 4 ) + 2p1xy + p2(m 2 + 2x 2 ) y(1 +k1m 2 + k2m 4 ) + 2p2xy + p1(m 2 + 2x 2 ) In the formula, and is the ideal value of the image point coordinates calculated by the pinhole model, (x, y) is the coordinates of the actual image point; m 2 =x 2 +y 2 , k1 and k2 are radial distortion coefficients, and p1 and p2 are tangential distortion coefficients.

4. The method for correcting the whole fundus laser scanning image according to claim 1, characterized in that: The function corresponding formula of the scale coordinate and the image coordinate system in step 203 is: ； Among them, (u, v) represents the coordinates in the image coordinate system, the origin is located at the pixel point (u0, v0), k×l represents the size of a pixel, f is the focal length of the camera, and θ is the skewness of the camera coordinate system.

5. The method for correcting the whole fundus laser scanning image according to claim 1, characterized in that: The formula for converting the scale coordinates into integers by bilinear interpolation in step 204 is: f(x,y)=a1x+b1y+c1xy+d1.

6. The method for correcting the whole fundus laser scanning image according to claim 1, characterized in that: The normalization method is cosine normalization method.

7. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores program code for executing the full fundus laser scanning image correction method described in any one of claims 1-6.

8. A full fundus laser scanning image correction system, characterized in that: Includes the computer-readable storage medium of claim 7.

Images