A method for fish-eye camera distortion correction

Abstract

This invention provides a method for fisheye camera distortion correction. The method includes: transforming the coordinates of the distortion-free image from the pixel coordinate system to the camera coordinate system; normalizing the coordinates of the camera coordinate system using the absolute value of z; calculating the distance from the optical center of the image to the distortion-free projection point using the normalized coordinates; calculating the angle between the incident ray and the optical axis based on the distance; calculating the distance from the optical center of the image to the distortion projection point based on a model polynomial; deriving the coordinates of the distortion projection point based on similar triangles; and mapping the coordinates of the distortion projection point to the corrected image coordinates to obtain the corrected distortion-free image. The fisheye camera distortion correction method provided by this invention is an improvement on the existing OpenCV fisheye distortion correction model. It achieves the restoration of image information with a field of view greater than or equal to 180 degrees after distortion correction without additional steps or recalibrating intrinsic parameters, thus ensuring the completeness and effectiveness of the image information after distortion correction.

Description

Technical Field

[0001] This invention relates to the field of image processing technology, and in particular to a method for distortion correction of a fisheye camera. Background Technology

[0002] A fisheye lens is a short focal length camera lens with a wide field of view (FOV of 180 degrees or more). Its front lens resembles the bulging eye of a fish, hence the name. The human eye's lens is flattened and round, allowing it to see objects at greater distances. A fisheye lens, however, is spherical, thus, while it can only see relatively close objects, it offers a wider field of view, meaning it sees a broader range of objects.

[0003] Fisheye lenses are widely used in various fields due to their advantages, but the images captured by fisheye lenses have very severe distortion, making them appear unnatural to the human eye. Therefore, it is necessary to correct the distorted images captured by fisheye lenses to make them visually acceptable to humans.

[0004] However, the existing fisheye camera distortion correction algorithm of OpenCV (a cross-platform computer vision and machine learning software library released under the Apache 2.0 license) cannot recover image information with a field of view (FOV) greater than or equal to 180 degrees, resulting in severe loss of image information after distortion correction.

[0005] The equidistant projection model is a commonly used model for fisheye lens distortion correction. CN201610817788 determines the effective region edge of the fisheye image by obtaining the radial brightness gradient map corresponding to the fisheye image, and then performs mapping based on the center and radius of this effective edge and the field of view of the fisheye lens. However, this scheme requires accurate radial brightness gradient and a clear understanding of the field of view of the fisheye lens, making its implementation conditions quite stringent. Summary of the Invention

[0006] The purpose of this invention is to provide a method for distortion correction of fisheye cameras, which can directly use the basic calibration results of OpenCV and can restore image information with a field of view of greater than or equal to 180 degrees after distortion correction without recalibrating the intrinsic parameters, thereby ensuring the integrity and effectiveness of the image information after distortion correction.

[0007] To solve the above-mentioned technical problems, the present invention provides a method for distortion correction of a fisheye camera, comprising:

[0008] Transform the coordinates of the distortion-free image from the pixel coordinate system to the camera coordinate system;

[0009] The coordinates of the camera coordinate system are normalized using the absolute value of z;

[0010] The distance r from the optical center of the image to the distortion-free projection point is calculated using normalized coordinates;

[0011] The angle θ between the incident ray and the optical axis is calculated based on the distance r.

[0012] The distance d(θ) from the optical center of the image to the distorted projection point is calculated based on the model polynomial.

[0013] The coordinates of the distorted projection points are derived from similar triangles; and

[0014] The coordinates of the distorted projection points are mapped to the coordinates of the corrected image to obtain the corrected, distortion-free image.

[0015] Optionally, the coordinates of the pixels in the fisheye image in the camera coordinate system are (x, y, z). The coordinates (a, b) of the distortion-free projection point in the normalized coordinate system are obtained according to a = x / abs(z) and b = y / abs(z).

[0016] Optionally, the distance r from the optical center of the image to the distortion-free projection point can be calculated using normalized coordinates:

[0017]

[0018] Optionally, the angle θ between the incident ray and the optical axis can be calculated based on the distance r as: θ = atan(r).

[0019] Optionally, if z < 0, then θ = π - atan(r).

[0020] Optionally, the distance d(θ) from the optical center of the image to the distorted projection point can be calculated based on the model polynomial as follows:

[0021] d(θ)=θ+k1θ 3 +k2θ 5 +k3θ 7 +k4θ 9

[0022] Among them, k1, k2, k3 and k4 are calibration parameters.

[0023] Optionally, the coordinates of the distorted projection points are derived based on similar triangles, and the coordinates of the distorted projection points are transformed from the normalized coordinate system to the pixel coordinate system.

[0024] Optionally, the coordinates of the distorted projection point in the pixel coordinate system are (u, v).

[0025]

[0026] Among them, f x fy c x With c y These are calibration parameters.

[0027] Optionally, the method for mapping the coordinates of the distorted projection point to the coordinates of the corrected image includes: filling the pixel information of the coordinates (u,v) of the distorted projection point into the corresponding coordinates (j,i) of the corrected image.

[0028] In summary, the fisheye camera distortion correction method provided by this invention first transforms the coordinates of the distortion-free image from the pixel coordinate system to the camera coordinate system. Then, the coordinates in the camera coordinate system are normalized using the absolute value of z. Next, the distance r from the optical center of the image to the distortion-free projection point is calculated using the normalized coordinates. Then, the angle θ between the incident ray and the optical axis is calculated based on the distance r. Next, the distance d(θ) from the optical center of the image to the distortion projection point is calculated using a model polynomial. Then, the coordinates of the distortion projection point are derived using similar triangles. Finally, the coordinates of the distortion projection point are mapped to the corrected image coordinates to obtain the corrected distortion-free image. This fisheye camera distortion correction method provided by this invention is an improvement on the existing OpenCV fisheye distortion correction model. It requires no additional steps and can directly use the existing OpenCV calibration results. It can restore image information with a field of view greater than or equal to 180 degrees after distortion correction without recalibrating intrinsic parameters, thus ensuring the completeness and effectiveness of the image information after distortion correction.

[0029] The fisheye camera distortion correction method provided by this invention uses the absolute value of z for normalization to ensure the correct coordinate mapping interval. If z < 0, then θ = π - atan(r), and the correct distortion curve segment is applied. Therefore, the effective area of ​​the fisheye image can be recovered. The field of view angle that can be recovered is significantly improved compared with the original OpenCV algorithm. Moreover, it is simple to calculate and easy to port. Attached Figure Description

[0030] Those skilled in the art will understand that the accompanying drawings are provided to better understand the invention and do not constitute any limitation on the scope of the invention. Wherein:

[0031] Figure 1 This is a schematic diagram of a fisheye lens imaging model.

[0032] Figure 2a This is a schematic diagram of a fisheye image.

[0033] Figure 2b This is a schematic diagram of an image corrected by rotating it 90 degrees to the left using the existing OpenCV fisheye correction algorithm.

[0034] Figure 2cThis is a schematic diagram of an image corrected by rotating it 90 degrees to the right using the existing OpenCV fisheye correction algorithm.

[0035] Figure 3 It is a curve of the distance d(θ) from the optical center O of the image to the distorted projection point P.

[0036] Figure 4 This is a flowchart of a method for fisheye camera distortion correction provided in one embodiment of the present invention.

[0037] Figure 5a This is a schematic diagram of a fisheye image.

[0038] Figure 5b This is a schematic diagram of an image corrected by rotating 90 degrees to the left using the fisheye camera distortion correction method provided in an embodiment of the present invention.

[0039] Figure 5c This is a schematic diagram of an image corrected by rotating 90 degrees to the right using the fisheye camera distortion correction method provided in an embodiment of the present invention.

[0040] Figure 6a The image shown on the left is a schematic diagram of the image after correction using the existing OpenCV fisheye correction algorithm.

[0041] Figure 6b This is a schematic diagram on the left side of an image after distortion correction using the fisheye camera distortion correction method provided in an embodiment of the present invention.

[0042] Figure 7a The image shown on the right is a schematic diagram of the image after correction using the existing OpenCV fisheye correction algorithm.

[0043] Figure 7b The image shown on the right is a schematic diagram of the image after distortion correction using the fisheye camera distortion correction method provided in an embodiment of the present invention. Detailed Implementation

[0044] Fisheye cameras offer a wider field of view compared to pinhole cameras, but this comes with significant image distortion. In product applications, fisheye images typically need to be corrected to restore them to a distortion-free image that conforms to visual perception.

[0045] The Kannala-Brandt model used in OpenCV can effectively simulate image distortion from fisheye cameras. Figure 1 This is a schematic diagram of a fisheye camera imaging model, such as... Figure 1As shown, OcXcYcZc is the fisheye camera coordinate system, Oxy is the fisheye image plane coordinate system, and the distance between the origin of the fisheye camera coordinate system and the origin of the fisheye image plane coordinate system is the focal length f of the fisheye camera. θ represents the angle between the incident ray and the optical axis of a pixel in the fisheye image captured by the fisheye camera. Assuming that the projection point corresponding to one pixel in the fisheye image captured by the fisheye camera in the fisheye camera coordinate system is Pc, then the distortion-free projection point in the fisheye image plane coordinate system corresponding to this projection point Pc is P′, and the actual projection point (i.e., the distorted projection point) is P.

[0046] Assuming a polynomial relationship exists between the distance d(θ) from the optical center O of the image to the distorted projection point P and the angle θ between the incident light and the optical axis, the model formula is as follows:

[0047]

[0048] a = x / z, b = y / z

[0049]

[0050] θ = atan(r)

[0051] d(θ)=θ+k1θ 3 +k2θ 5 +k3θ 7 +k4θ 9

[0052] Among them, k1, k2, k3, k4 and f x f y c x With c y These are all calibration parameters. The coordinates of the projection point Pc are (x, y, z), the distance from the image optical center O to the distortion-free projection point P is r, and the coordinates of the distortion-free projection point P′ are (u, v). The coordinates of the distortion-free projection point P′ can be obtained using the above formula.

[0053] However, using this model to correct fisheye distortion is an inverse mapping process: first, the pinhole model is applied to obtain the distortion-free projection point, and then the distortion relation is applied to obtain the corresponding distorted projection point. The former has a singularity when z=0, which makes it impossible to obtain the distortion-free projection point when the field of view angle is close to 180 degrees, because the distortion-free image cannot be recovered.

[0054] Figure 2a This is a schematic diagram of a fisheye image. Figure 2b This is a schematic diagram of an image corrected by rotating it 90 degrees to the left using the existing OpenCV fisheye correction algorithm. Figure 2c This is a schematic diagram of an image corrected by rotating it 90 degrees to the right using the existing OpenCV fisheye correction algorithm. Please refer to it. Figures 2a to 2cAs shown, when using existing OpenCV fisheye correction algorithms to attempt to recover image information from fisheye images with visual angles greater than or equal to 180 degrees on both sides, the image reaches the recovery limit at a visual angle of 180 degrees. Content exceeding 180 degrees represents a flip of image information within a visual angle of 180 degrees on the other side. Figure 2b The image on the left side of the middle page, and Figure 2c The images on the right are all flipped versions of image information.

[0055] To address the aforementioned problems, the inventors conducted research. Analysis of the model formula clearly shows that d(θ) generally increases within the range of the angle θ between the incident ray and the optical axis. Figure 3 It is a curve of the distance d(θ) from the optical center O of the image to the distorted projection point P. Figure 3 As can be seen, with the increase of the angle θ, the distance d(θ) from the projection point to the optical center of the image also gradually increases, and the image exhibits a circular distortion with inner expansion (slow change of θ) and outer compression (rapid change of θ). The linear fit of the d(θ) curve determines the maximum angle θ that the model can correct.

[0056] In the existing OpenCV formula, because the distance r ≥ 0 from the optical center O of the image to the distortion-free projection point P, the solution of atan(r) is always in the interval [0, π / 2]. When θ > π / 2 (viewing angle is greater than 180 degrees), the distortion information d(θ) curve is not effective. Moreover, when z < 0, the (a, b) direction and the (x, y) direction are reversed, which leads to the incorrect index (u, v) in the interval during inverse mapping, resulting in image misalignment and flipping.

[0057] Further research has led to the present invention providing a method for fisheye camera distortion correction. By simply modifying the original formula, the large-angle information in the distortion model can be effectively utilized to successfully recover image information beyond a 180-degree field of view.

[0058] This invention provides a method for distortion correction in fisheye cameras, comprising:

[0059] Transform the coordinates of the distortion-free image from the pixel coordinate system to the camera coordinate system;

[0060] The coordinates of the camera coordinate system are normalized using the absolute value of z;

[0061] The distance r from the optical center of the image to the distortion-free projection point is calculated using normalized coordinates;

[0062] The angle θ between the incident ray and the optical axis is calculated based on the distance r.

[0063] The distance d(θ) from the optical center of the image to the distorted projection point is calculated based on the model polynomial.

[0064] The coordinates of the distorted projection points are derived from similar triangles; and

[0065] The coordinates are mapped to the coordinates of the corrected image to obtain the corrected, distortion-free image.

[0066] The fisheye camera distortion correction method provided by this invention is based on the existing OpenCV fisheye distortion correction model and is improved. It does not require additional steps and can directly use the existing OpenCV basic calibration results. It can restore the image information with a field of view of greater than or equal to 180 degrees after distortion correction without recalibrating the intrinsic parameters, thereby ensuring the integrity and effectiveness of the image information after distortion correction.

[0067] To make the objectives, advantages, and features of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings and specific embodiments. It should be noted that the drawings are all in a very simplified form and are not drawn to scale, and are only used to facilitate and clarify the explanation of the embodiments of this invention. Furthermore, the structures shown in the drawings are often part of the actual structures. In particular, different figures may emphasize different aspects and may sometimes use different scales.

[0068] As used in this invention, the singular forms “a,” “an,” and “the” include plural objects; the term “or” is generally used to mean “and / or”; the term “a number” is generally used to mean “at least one”; the term “at least two” is generally used to mean “two or more”; furthermore, the terms “first,” “second,” and “third” are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of indicated technical features. Therefore, features defined as "first," "second," and "third" may explicitly or implicitly include one or at least two of those features. The term "proximal" usually refers to the end closer to the operator, and the term "distal" usually refers to the end closer to the patient. "One end" and "the other end," as well as "proximal" and "distal," usually refer to two corresponding parts, which include not only the endpoints. The terms "installation," "connection," and "joining" should be interpreted broadly. For example, they can be fixed connections, detachable connections, or integral connections; they can be mechanical connections or electrical connections; they can be direct connections or indirect connections through an intermediate medium; they can be internal connections between two components or the interaction between two components.

[0069] Furthermore, as used in this invention, the phrase "one element is disposed on another element" generally only indicates that there is a connection, coupling, cooperation, or transmission relationship between the two elements, and the connection, coupling, cooperation, or transmission between the two elements can be direct or indirect through an intermediate element. It should not be construed as indicating or implying a spatial positional relationship between the two elements, i.e., one element can be located arbitrarily inside, outside, above, below, or to one side of the other element, unless otherwise explicitly stated. Those skilled in the art can understand the specific meaning of the above terms in this invention according to the specific circumstances.

[0070] Figure 4 This is a flowchart of a method for fisheye camera distortion correction provided in one embodiment of the present invention. Please refer to it. Figure 4 As shown, the present invention provides a method for distortion correction of a fisheye camera, comprising:

[0071] S1: Transform the coordinates of the distortion-free image from the pixel coordinate system to the camera coordinate system;

[0072] S2: Normalize the coordinates of the camera coordinate system using the absolute value of z;

[0073] S3: Calculate the distance r from the optical center of the image to the distortion-free projection point using normalized coordinates;

[0074] S4: Calculate the angle θ between the incident ray and the optical axis based on the distance r;

[0075] S5: Calculate the distance d(θ) from the optical center of the image to the distorted projection point based on the model polynomial;

[0076] S6: Derive the coordinates of the distorted projection points based on similar triangles; and

[0077] S7: Map the coordinates of the distorted projection points to the coordinates of the corrected image to obtain the corrected, distortion-free image.

[0078] Next, we will combine Figure 1 and Figure 4 As shown, a method for fisheye camera distortion correction provided by an embodiment of the present invention will be described in detail.

[0079] In step S1, the coordinates of the distortion-free image are transformed from the pixel coordinate system to the camera coordinate system.

[0080] In this step, the coordinates of the distortion-free image are transformed from the pixel coordinate system to the camera coordinate system. Since the coordinate transformation is performed using existing known methods, this invention will not describe it in detail.

[0081] Please refer to Figure 1As shown, in the camera coordinate system OcXcYcZc, the coordinates of a pixel in the fisheye image are (x, y, z). That is, the coordinates of a certain pixel Pc in the fisheye image are (x, y, z) in the camera coordinate system. The angle between this pixel Pc and the optical axis is θ, that is, the angle between the incident ray and the optical axis is θ.

[0082] In step S2, the coordinates of the camera coordinate system are normalized using the absolute value of z.

[0083] Specifically, the coordinates of the camera coordinate system are normalized to obtain the fisheye image plane coordinate system oxyz, that is, the normalized coordinate system oxyz. In this normalized coordinate system, the coordinates of the distortion-free projection point P′ are (a, b), where a = x / abs(z), b = y / abs(z), and abs(z) is the absolute value of z.

[0084] In this step, normalization using the absolute value of z ensures the correct coordinate mapping range.

[0085] In step S3, the distance r from the optical center of the image to the distortion-free projection point is calculated using normalized coordinates.

[0086] Please refer to Figure 1 As shown, given that the coordinates of the distortion-free projection point P′ are (a, b), the distance r from the image optical center O to the distortion-free projection point P′ can be calculated using the formula... Perform the calculation.

[0087] In step S4, the angle θ between the incident ray and the optical axis is calculated based on the distance r.

[0088] Specifically, in Figure 1 In this context, the angle θ between the incident ray and the optical axis is: θ = atan(r). Furthermore, if z < 0, then θ = π - atan(r).

[0089] In this step, the case of z < 0 was considered, and the correct distortion curve segment was applied.

[0090] The fisheye camera distortion correction method provided by this invention uses the absolute value of z for normalization to ensure the correct coordinate mapping interval. If z < 0, then θ = π - atan(r), and the correct distortion curve segment is applied. Therefore, the effective area of ​​the fisheye image can be recovered. The field of view angle that can be recovered is significantly improved compared with the original OpenCV algorithm. Moreover, it is simple to calculate and easy to port.

[0091] In step S5, the distance d(θ) from the optical center of the image to the distortion projection point is calculated based on the model polynomial.

[0092] The distance d(θ) from the optical center O of the image to the distorted projection point P is:

[0093] d(θ)=θ+k1θ 3 +k2θ 5 +k3θ 7 +k4θ 9

[0094] Among them, k1, k2, k3 and k4 are calibration parameters.

[0095] In step S6, the coordinates of the distorted projection point P are derived based on similar triangles.

[0096] Please refer to Figure 1 As shown, the coordinates of the distorted projection point P are derived based on similar triangles, and these coordinates are then transformed from the normalized coordinate system to the pixel coordinate system. Finally, the coordinates of the distorted projection point P in the pixel coordinate system are obtained as (u, v).

[0097]

[0098] Among them, f x f y c x With c y These are calibration parameters.

[0099] In step S7, the coordinates of the distorted projection point are mapped to the coordinates of the corrected image to obtain the corrected, distortion-free image. Specifically, the pixel information of the coordinates (u,v) of the distorted projection point P is filled into the corresponding coordinates (j,i) of the corrected image to obtain the corrected, distortion-free image.

[0100] The fisheye camera distortion correction method provided by this invention is based on the existing OpenCV fisheye distortion correction model and is improved. No additional steps are required. The existing OpenCV basic calibration results can be directly used. The image information with a field of view of greater than or equal to 180 degrees after distortion correction can be restored without recalibrating the intrinsic parameters, thereby ensuring the integrity and effectiveness of the image information after distortion correction.

[0101] Figure 5a This is a schematic diagram of a fisheye image. Figure 5b This is a schematic diagram of an image corrected by rotating 90 degrees to the left using the fisheye camera distortion correction method provided in an embodiment of the present invention. Figure 5c This is a schematic diagram of an image corrected by rotating 90 degrees to the right using the fisheye camera distortion correction method provided in an embodiment of the present invention. Please refer to it. Figures 5a to 5cAs shown, normalization using the absolute value of z ensures the correct coordinate mapping interval. If z < 0, then θ = π - atan(r), applying the correct distortion curve segment. The limiting angle of correction is affected by the results of calibration parameters k1, k2, k3, and k4. Under the premise of effective calibration (maxθ > π / 2), recalibration is unnecessary, allowing the improved fisheye distortion correction algorithm to successfully recover image information with a visual field angle greater than or equal to 180 degrees.

[0102] Figure 6a This is a schematic diagram on the left side of the image after correction using the existing OpenCV fisheye correction algorithm. Figure 6b This is a schematic diagram on the left side of an image after distortion correction using the fisheye camera distortion correction method provided in an embodiment of the present invention. Figure 7a The image shown on the right is a schematic diagram of the image after correction using the existing OpenCV fisheye correction algorithm. Figure 7b This is a schematic diagram on the right side of an image after distortion correction using the fisheye camera distortion correction method provided in an embodiment of the present invention. (Comparison) Figure 6a and Figure 6b ,as well as Figure 7a and Figure 7b It can be observed that the fisheye camera distortion correction method described in this invention significantly improves the field of view restored compared to the existing OpenCV fisheye correction algorithm.

[0103] In summary, the fisheye camera distortion correction method provided by this invention first transforms the coordinates of the distortion-free image from the pixel coordinate system to the camera coordinate system. Then, the coordinates in the camera coordinate system are normalized using the absolute value of z. Next, the distance r from the optical center of the image to the distortion-free projection point is calculated using the normalized coordinates. Then, the angle θ between the incident ray and the optical axis is calculated based on the distance r. Next, the distance d(θ) from the optical center of the image to the distortion projection point is calculated using a model polynomial. Then, the coordinates of the distortion projection point are derived using similar triangles. Finally, the coordinates of the distortion projection point are mapped to the corrected image coordinates to obtain the corrected distortion-free image. This fisheye camera distortion correction method provided by this invention is an improvement on the existing OpenCV fisheye distortion correction model. It requires no additional steps and can directly use the existing OpenCV calibration results. It can restore image information with a field of view greater than or equal to 180 degrees after distortion correction without recalibrating intrinsic parameters, thus ensuring the completeness and effectiveness of the image information after distortion correction.

[0104] The fisheye camera distortion correction method provided by this invention uses the absolute value of z for normalization to ensure the correct coordinate mapping interval. If z < 0, then θ = π - atan(r), and the correct distortion curve segment is applied. Therefore, the effective area of ​​the fisheye image can be recovered. The field of view angle that can be recovered is significantly improved compared with the original OpenCV algorithm. Moreover, it is simple to calculate and easy to port.

[0105] The above description is merely a description of preferred embodiments of the present invention and is not intended to limit the scope of the present invention. Any person skilled in the art can make possible changes and modifications to the technical solutions of the present invention by utilizing the methods and techniques disclosed above without departing from the spirit and scope of the present invention. Therefore, any simple modifications, equivalent changes and alterations made to the above embodiments based on the technical essence of the present invention without departing from the content of the technical solutions of the present invention shall fall within the protection scope of the technical solutions of the present invention.

Claims

1. A method for distortion correction in a fisheye camera, characterized in that, include: Transform the coordinates of the distortion-free image from the pixel coordinate system to the camera coordinate system; The coordinates of the camera coordinate system are normalized using the absolute value of z; The distance r from the optical center of the image to the distortion-free projection point is calculated using normalized coordinates; The angle θ between the incident ray and the optical axis is calculated based on the distance r; wherein, the angle θ between the incident ray and the optical axis is: θ = atan(r), and if z < 0, then θ = π - atan(r); The distance d(θ) from the optical center of the image to the distorted projection point is calculated based on the model polynomial. The coordinates of the distorted projection points are derived from similar triangles; and The coordinates of the distorted projection points are mapped to the coordinates of the corrected image to obtain the corrected, distortion-free image.

2. The method for fisheye camera distortion correction according to claim 1, characterized in that, The coordinates of the pixels in the fisheye image in the camera coordinate system are (x, y, z). According to a = x / abs(z) and b = y / abs(z), the coordinates of the distortion-free projection point in the normalized coordinate system are obtained as (a, b).

3. The method for fisheye camera distortion correction according to claim 2, characterized in that, The distance r from the optical center of the image to the distortion-free projection point is calculated using normalized coordinates: 。 4. The method for fisheye camera distortion correction according to claim 3, characterized in that, The distance d(θ) from the optical center of the image to the distorted projection point is calculated based on the model polynomial as follows: in, , , and These are calibration parameters.

5. The method for fisheye camera distortion correction according to claim 4, characterized in that, The coordinates of the distorted projection point are derived based on similar triangles, and the coordinates of the distorted projection point are transformed from the normalized coordinate system to the pixel coordinate system.

6. The method for fisheye camera distortion correction according to claim 5, characterized in that, The coordinates of the distorted projection point in the pixel coordinate system are (u, v). in, , , and These are calibration parameters.

7. The method for distortion correction of a fisheye camera according to claim 6, characterized in that, The method for mapping the coordinates of the distorted projection point to the coordinates of the corrected image includes: filling the pixel information of the coordinates (u,v) of the distorted projection point into the corresponding coordinates (j,i) of the corrected image.

Images