A low deformation control method for ultra-wide-angle lens with large depth of field
By setting up an image correction device in a large depth of field ultra-wide-angle lens, establishing the correspondence between the grid imaging picture and the original grid and forming a mapping mechanism, the image distortion problem caused by unreasonable deformation control of wide-angle lenses is solved, and the image imaging quality is improved.
Patent Information
- Application Number
- CN202411186706.1
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2024-08-28
- Publication Date
- 2024-11-19
- Estimated Expiration
- 2044-08-28
AI Technical Summary
In the prior art, when using wide-angle lenses, unreasonable deformation control leads to distortion of image imaging effects, especially in large depth of field ultra-wide-angle lenses, which affects image quality.
By setting up image correction devices inside the ultra-wide-angle lens with a large depth of field, a preset grid picture is formed and photographed, a correspondence between the grid imaging picture and the original grid is established, and a mapping mechanism is formed to correct the image and ensure that each grid is The stretching ratio is reasonable.
The corrected image distortion degree is reduced, the image imaging effect is improved, and the actual situation of deformation degree at different positions is reasonably processed, avoiding distortion caused by stretching in the same proportion.
Smart Images

Figure CN118707720B_ABST
Abstract
Description
Technical Field
[0001] The invention relates to the field of optical technology, and in particular to a low deformation control method for an ultra-wide-angle lens with a large depth of field. Background Art
[0002] Lens distortion in photography refers to optical aberrations that occur when a camera lens fails to accurately reproduce the image of the subject being photographed. This results in a distorted (distorted) image that appears stretched or compressed, depending on the type of distortion. There are two main types of lens distortion: barrel distortion and pincushion distortion. Barrel distortion occurs when straight lines near the edges of an image appear to bend outward, creating a bulging effect. Pincushion distortion, on the other hand, occurs when straight lines near the edges of an image appear to bend inward, creating a squeezing effect.
[0003] When using a wide-angle lens, these distortions will be more obvious. Therefore, the deformation needs to be controlled. However, the existing control methods will stretch the picture to a certain extent. When the stretching ratio is not properly controlled, it is easy to cause distortion in some parts of the picture, thereby affecting the image imaging effect. Summary of the invention
[0004] In order to solve the above technical problems, a low deformation control method for an ultra-wide-angle lens with a large depth of field is provided. The technical solution solves the problem raised in the above background technology that when a wide-angle lens is used, these distortions will be more obvious. Therefore, the deformation needs to be controlled, but the existing control method will stretch the picture to a certain extent. When the stretching ratio is not properly controlled, it is easy to cause distortion in some parts of the picture, thereby affecting the image imaging effect.
[0005] In order to achieve the above purpose, the technical solution adopted by the present invention is:
[0006] A low deformation control method for a large depth of field ultra-wide angle lens, comprising:
[0007] An image correction device is provided inside the ultra-wide-angle lens with a large depth of field;
[0008] A preset grid screen is formed, wherein the grid located at the center of the preset grid screen is a reference grid, the reference grid is black, and all grids other than the reference grid are white, and the preset grid screen is composed of grids of equal size tiled;
[0009] The preset grid picture is photographed, the reference grid is located within the shooting sight, the preset grid picture is larger than the shooting sight range, the shooting angle is perpendicular to the preset grid picture, and a grid imaging picture of the preset grid picture captured by an ultra-wide-angle lens with a large depth of field is obtained;
[0010] In the preset grid picture, an original grid picture corresponding to the grid imaging picture is obtained, the grid in the original grid picture is used as the second grid, and the grid in the grid imaging picture is used as the first grid;
[0011] Establishing a correspondence between the grid imaging picture and the grids in the corresponding original grid picture;
[0012] forming a mapping mechanism that maps the first grid to a grid of equal size to the second grid, and pairing the mapping mechanism with the first grid mapping;
[0013] Aggregating at least one first grid and its corresponding mapping mechanism to obtain a correction mechanism;
[0014] The correction mechanism is stored in an image correction device, and the image correction device uses the correction mechanism to correct an image captured by an ultra-wide-angle lens with a large depth of field.
[0015] Preferably, forming a preset grid screen comprises the following steps:
[0016] Divide the preset scenery surface using equally spaced horizontal lines and equally spaced vertical lines, wherein the spacing between adjacent horizontal lines is a preset distance, and the spacing between adjacent vertical lines is a preset distance;
[0017] A grid formed by at least one horizontal line and a vertical line forms a preset grid picture on the preset scenery surface.
[0018] Preferably, in the preset grid picture, obtaining the original grid picture corresponding to the grid imaging picture comprises the following steps:
[0019] Identify the position of the black grid in the grid imaging image as the calibration grid;
[0020] Identifying grids at the edge of the grid imaging picture to obtain at least one first edge grid;
[0021] Get the path coordinates (q, p) from the first edge grid to the calibration grid. The path coordinates (q, p) satisfy that the calibration grid moves |q| grids horizontally and |p| grids vertically, then the calibration grid coincides with the first edge grid, where |q| is the absolute value of q, and |p| is the absolute value of p. When q is greater than 0, move horizontally to the right, when q is less than 0, move horizontally to the left, when p is greater than 0, move vertically upward, and when p is less than 0, move vertically downward;
[0022] In the preset grid screen, the reference grid is moved according to the path coordinates to obtain a second edge grid;
[0023] The area enclosed by at least one second edge grid in the preset grid picture forms an original grid picture corresponding to the grid imaging picture.
[0024] Preferably, establishing a correspondence between the grid imaging picture and the grids in the corresponding original grid picture comprises the following steps:
[0025] Counting the path coordinates from the first grid to the calibration grid in the grid imaging picture as the first path coordinates;
[0026] Counting the path coordinates from the second grid to the reference grid in the original grid image as the second path coordinates;
[0027] A corresponding relationship is established between the first grid and the second grid whose first path coordinates are equal to the second path coordinates.
[0028] Preferably, the mapping mechanism for mapping the first grid to a grid of the same size as the second grid comprises the following steps:
[0029] Coordinate modeling of grid imaging images;
[0030] Based on the upper and lower edges of the first grid, a first fitting function is formed,
[0031] Based on the left and right edges of the first grid, forming a second fitting function;
[0032] Using the first fitting function and the second fitting function to perform coordinate transformation on the coordinates of the points in the first grid, to obtain a first transformed grid;
[0033] Using the coordinate transformation formula, the first transformed grid is mapped to a grid of the same size as the second grid.
[0034] Preferably, forming a first fitting function based on the upper and lower edges of the first grid comprises the following steps:
[0035] Uniformly select at least one upper identification point on the upper edge of the first grid, and obtain the minimum value of the ordinate of the upper identification point as the upper eigenvalue;
[0036] Subtract the upper eigenvalue from the ordinate of the upper identification point to obtain the upper identification point of the feature;
[0037] Fitting the abscissa and ordinate of the identification point on at least one feature to obtain an upper fitting function, wherein the abscissa of the identification point on the feature is an independent variable, and the ordinate of the identification point on the feature is a dependent variable;
[0038] Uniformly select at least one lower identification point on the lower edge of the first grid, and obtain the minimum value of the ordinate of the lower identification point as the lower eigenvalue;
[0039] Subtract the lower eigenvalue from the ordinate of the lower identification point to obtain the characteristic lower identification point;
[0040] Fitting the abscissa and ordinate of at least one identification point under the feature to obtain a lower fitting function, wherein the abscissa of the identification point under the feature is an independent variable, and the ordinate of the identification point under the feature is a dependent variable;
[0041] The lower fitting function and the upper fitting function are averaged to obtain the first fitting function.
[0042] Preferably, forming a second fitting function based on the left and right edges of the first grid comprises the following steps:
[0043] Uniformly select at least one right identification point on the right edge of the first grid, and obtain the minimum value of the horizontal coordinate of the right identification point as the right eigenvalue;
[0044] Subtract the right eigenvalue from the horizontal coordinate of the right identification point to obtain the characteristic right identification point;
[0045] Fitting the ordinate and abscissa of at least one characteristic right identification point to obtain a right fitting function, wherein the ordinate of the characteristic right identification point is an independent variable, and the abscissa of the characteristic right identification point is a dependent variable;
[0046] Uniformly select at least one left identification point on the left edge of the first grid, and obtain the minimum value of the horizontal coordinate of the left identification point as the left eigenvalue;
[0047] Subtract the left eigenvalue from the horizontal coordinate of the left identification point to obtain the characteristic left identification point;
[0048] Fitting the ordinate and abscissa of at least one left feature identification point to obtain a left fitting function, wherein the ordinate of the left feature identification point is an independent variable, and the abscissa of the left feature identification point is a dependent variable;
[0049] The left fitting function and the right fitting function are averaged to obtain the second fitting function.
[0050] Preferably, using the first fitting function and the second fitting function to transform the coordinates of the points in the first grid to obtain the first transformed grid comprises the following steps:
[0051] The coordinates (x, y) of the points in the first grid are transformed to obtain (xf(y), yg(x)), where f(y) is the second fitting function and g(x) is the first fitting function.
[0052] Preferably, the step of using a coordinate transformation formula to map the first transformed grid to a grid of the same size as the second grid comprises the following steps:
[0053] The coordinate transformation formula includes a first coordinate transformation formula and a second coordinate transformation formula, the first coordinate transformation formula is used to transform the abscissa of the first transformation grid, and the second coordinate transformation formula is used to transform the ordinate of the first transformation grid;
[0054] The first coordinate transformation formula is as follows: ,
[0055] Wherein, a is the minimum value of the abscissa of the first transformed grid, b is the maximum value of the abscissa of the first transformed grid, c is the minimum value of the abscissa of the first grid, and d is the minimum value of the abscissa of the first grid. is the abscissa of the point in the first transformed grid, is the horizontal coordinate of the point in the first grid, and For the corresponding relationship;
[0056] The second coordinate transformation formula is as follows: ,
[0057] Wherein, h is the minimum value of the ordinate of the first transformed grid, i is the maximum value of the ordinate of the first transformed grid, j is the minimum value of the ordinate of the first grid, k is the minimum value of the ordinate of the first grid, is the ordinate of the point in the first transformed grid, is the ordinate of the point in the first grid, and For the corresponding relationship.
[0058] Preferably, the image correction device uses a correction mechanism to correct an image captured by an ultra-wide-angle lens with a large depth of field, comprising the following steps:
[0059] Acquire an actual imaging picture, divide the actual imaging picture according to the grid division in the grid imaging picture, obtain at least one actual grid, and place the first grid and the actual grid at the same relative position;
[0060] When correcting the actual grid, the correction is performed using a mapping mechanism of the first grid corresponding to the actual grid.
[0061] Compared with the prior art, the present invention has the following beneficial effects:
[0062] By performing grid division, different mapping mechanisms are formed for different grids, so that the captured image can be divided in the same way, and the same mapping mechanism is used for correction in grids with the same relative position. Thus, a more reasonable stretching ratio can be assigned to each grid. This method takes into account the actual situation that the deformation degree of different positions in the image is different, instead of stretching all areas at the same ratio. It can reduce the distortion of the corrected image, thereby ensuring the image imaging effect. BRIEF DESCRIPTION OF THE DRAWINGS
[0063] Figure 1 It is a schematic flow chart of a low-distortion control method for a large-depth-of-field ultra-wide-angle lens according to the present invention;
[0064] Figure 2 A schematic diagram of a process of forming a preset grid screen according to the present invention;
[0065] Figure 3 A schematic diagram of a process of obtaining an original grid picture corresponding to a grid imaging picture in a preset grid picture of the present invention;
[0066] Figure 4 A schematic diagram of a process of establishing a corresponding relationship between a grid imaging picture and a grid in a corresponding original grid picture according to the present invention;
[0067] Figure 5 A schematic flow chart of a mapping mechanism for mapping a first grid to a grid of the same size as a second grid according to the present invention;
[0068] Figure 6 It is a schematic diagram of a process of forming a first fitting function based on the upper and lower edges of the first grid according to the present invention;
[0069] Figure 7 It is a schematic diagram of a process of forming a second fitting function based on the left and right edges of the first grid according to the present invention;
[0070] Figure 8 The present invention is a flow chart of the image correction device using the correction mechanism to correct an image taken by an ultra-wide-angle lens with a large depth of field. DETAILED DESCRIPTION
[0071] The following description is used to disclose the present invention so that those skilled in the art can implement the present invention. The preferred embodiments described below are only examples, and those skilled in the art may think of other obvious variations.
[0072] Reference Figure 1 As shown, a low deformation control method for a large depth of field ultra-wide angle lens comprises:
[0073] An image correction device is provided inside the ultra-wide-angle lens with a large depth of field;
[0074] A preset grid screen is formed, wherein the grid located at the center of the preset grid screen is a reference grid, the reference grid is black, and all grids other than the reference grid are white, and the preset grid screen is composed of grids of equal size tiled;
[0075] The preset grid picture is photographed, the reference grid is located within the shooting sight, the preset grid picture is larger than the shooting sight range, the shooting angle is perpendicular to the preset grid picture, and a grid imaging picture of the preset grid picture captured by an ultra-wide-angle lens with a large depth of field is obtained;
[0076] In the preset grid picture, an original grid picture corresponding to the grid imaging picture is obtained, the grid in the original grid picture is used as the second grid, and the grid in the grid imaging picture is used as the first grid;
[0077] Establishing a correspondence between the grid imaging picture and the grids in the corresponding original grid picture;
[0078] forming a mapping mechanism that maps the first grid to a grid of equal size to the second grid, and pairing the mapping mechanism with the first grid mapping;
[0079] Aggregating at least one first grid and its corresponding mapping mechanism to obtain a correction mechanism;
[0080] The correction mechanism is stored in an image correction device, and the image correction device uses the correction mechanism to correct an image captured by an ultra-wide-angle lens with a large depth of field.
[0081] In this scheme, a correspondence is established between the grid imaging screen and the grids in the corresponding original grid screen. The original grid screen is an image in the preset grid screen that corresponds to the grid imaging screen. The original grid screen is composed of very small equal-sized grids, and the grid imaging screen is obtained by imaging after being captured by a wide-angle lens. Therefore, the grids in the grid imaging screen are deformed curved-edge grids, but because the grids are very small, the deformation degrees on the upper and lower sides of the grids can be regarded as the same, and the deformation degrees on the left and right sides of the grids can also be regarded as the same. Therefore, an approximate mapping mechanism is established for each grid, so that the curved-edge grids in the grid imaging screen can be stretched to the same size as the grids in the original grid screen. Since the deformation degree of each grid is different, the corresponding mapping mechanisms are also different. Therefore, its stretching process takes into account the different deformation conditions of different positions of the wide-angle lens, so that during the stretching process, the image can be restored to a higher degree.
[0082] Reference Figure 2 As shown, forming a preset grid screen includes the following steps:
[0083] Divide the preset scenery surface using equally spaced horizontal lines and equally spaced vertical lines, wherein the spacing between adjacent horizontal lines is a preset distance, and the spacing between adjacent vertical lines is a preset distance;
[0084] A grid formed by at least one horizontal line and a vertical line forms a preset grid picture on the preset scenery surface.
[0085] Reference Figure 3 As shown, in the preset grid picture, obtaining the original grid picture corresponding to the grid imaging picture includes the following steps:
[0086] Identify the position of the black grid in the grid imaging image as the calibration grid;
[0087] Identifying grids at the edge of the grid imaging picture to obtain at least one first edge grid;
[0088] Get the path coordinates (q, p) from the first edge grid to the calibration grid. The path coordinates (q, p) satisfy that the calibration grid moves |q| grids horizontally and |p| grids vertically, then the calibration grid coincides with the first edge grid, where |q| is the absolute value of q, and |p| is the absolute value of p. When q is greater than 0, move horizontally to the right, when q is less than 0, move horizontally to the left, when p is greater than 0, move vertically upward, and when p is less than 0, move vertically downward;
[0089] In the preset grid screen, the reference grid is moved according to the path coordinates to obtain a second edge grid;
[0090] The area enclosed by at least one second edge grid in the preset grid picture forms an original grid picture corresponding to the grid imaging picture.
[0091] Because the grids in the grid imaging picture are distorted, it is difficult to obtain the original grid picture corresponding to the grid imaging picture in the preset grid picture by using the coordinate establishment method. Considering that the grids after wide-angle lens imaging are distorted, the number and relative positions of the grids have not changed, that is, after the corresponding grids in the preset grid picture and the grid imaging picture move synchronously horizontally and vertically, the grids achieved are grids with consistent relative positions. Therefore, the original grid picture corresponding to the grid imaging picture can be obtained in the preset grid picture.
[0092] Reference Figure 4 As shown, establishing a correspondence between a grid imaging picture and a grid in a corresponding original grid picture includes the following steps:
[0093] Counting the path coordinates from the first grid to the calibration grid in the grid imaging picture as the first path coordinates;
[0094] Counting the path coordinates from the second grid to the reference grid in the original grid image as the second path coordinates;
[0095] A corresponding relationship is established between the first grid and the second grid whose first path coordinates are equal to the second path coordinates.
[0096] The purpose is to establish a corresponding relationship between the grid imaging picture corresponding to the same image and the grid in the original grid picture, thereby establishing a mapping mechanism between the corresponding grid imaging picture and the grid in the original grid picture, so that targeted deformation control can be performed.
[0097] Reference Figure 5 As shown, the mapping mechanism for mapping the first grid to a grid of the same size as the second grid includes the following steps:
[0098] Coordinate modeling of grid imaging images;
[0099] Based on the upper and lower edges of the first grid, a first fitting function is formed,
[0100] Based on the left and right edges of the first grid, forming a second fitting function;
[0101] Using the first fitting function and the second fitting function to perform coordinate transformation on the coordinates of the points in the first grid, to obtain a first transformed grid;
[0102] Using the coordinate transformation formula, the first transformed grid is mapped to a grid of the same size as the second grid.
[0103] The first grid is a curved-edge square. Since the first grid is very small, the deformation degree of its upper and lower edges can be considered consistent, and the deformation degree of its left and right edges can be considered consistent. Therefore, when adjusting, first perform vertical adjustment, that is, make the upper and lower edges become approximate straight lines, and perform the same transformation on the vertical coordinates of the points in the first grid. Secondly, perform horizontal adjustment, that is, make the effective edge become an approximate straight line, and perform the same transformation on the horizontal coordinates of the points in the first grid. Thus, the first transformed grid is obtained. The first transformed grid is an approximate square, but the size of the first transformed grid may be inconsistent with the size of the second grid, because the first grid is stretched and imaged by a wide-angle lens, and the first transformed grid obtained is also stretched. Therefore, the first transformed grid is scaled so that the side length of the scaled first transformed grid is consistent with the side length of the second grid, thereby completing the deformation control.
[0104] Reference Figure 6 As shown, based on the upper and lower edges of the first grid, forming a first fitting function includes the following steps:
[0105] Uniformly select at least one upper identification point on the upper edge of the first grid, and obtain the minimum value of the ordinate of the upper identification point as the upper eigenvalue;
[0106] Subtract the upper eigenvalue from the ordinate of the upper identification point to obtain the upper identification point of the feature;
[0107] Fitting the abscissa and ordinate of the identification point on at least one feature to obtain an upper fitting function, wherein the abscissa of the identification point on the feature is an independent variable, and the ordinate of the identification point on the feature is a dependent variable;
[0108] Uniformly select at least one lower identification point on the lower edge of the first grid, and obtain the minimum value of the ordinate of the lower identification point as the lower eigenvalue;
[0109] Subtract the lower eigenvalue from the ordinate of the lower identification point to obtain the characteristic lower identification point;
[0110] Fitting the abscissa and ordinate of at least one identification point under the feature to obtain a lower fitting function, wherein the abscissa of the identification point under the feature is an independent variable, and the ordinate of the identification point under the feature is a dependent variable;
[0111] The lower fitting function and the upper fitting function are averaged to obtain the first fitting function.
[0112] There is actually a certain difference in the deformation degree of the upper and lower edges of the first grid. In order to reduce the error, the average is taken to obtain the first fitting function, which can reduce the adjustment error.
[0113] Reference Figure 7 As shown, based on the left and right edges of the first grid, forming the second fitting function includes the following steps:
[0114] Uniformly select at least one right identification point on the right edge of the first grid, and obtain the minimum value of the horizontal coordinate of the right identification point as the right eigenvalue;
[0115] Subtract the right eigenvalue from the horizontal coordinate of the right identification point to obtain the characteristic right identification point;
[0116] Fitting the ordinate and abscissa of at least one characteristic right identification point to obtain a right fitting function, wherein the ordinate of the characteristic right identification point is an independent variable, and the abscissa of the characteristic right identification point is a dependent variable;
[0117] Uniformly select at least one left identification point on the left edge of the first grid, and obtain the minimum value of the horizontal coordinate of the left identification point as the left eigenvalue;
[0118] Subtract the left eigenvalue from the horizontal coordinate of the left identification point to obtain the characteristic left identification point;
[0119] Fitting the ordinate and abscissa of at least one left feature identification point to obtain a left fitting function, wherein the ordinate of the left feature identification point is an independent variable, and the abscissa of the left feature identification point is a dependent variable;
[0120] The left fitting function and the right fitting function are averaged to obtain the second fitting function.
[0121] There is actually a certain difference in the degree of deformation of the left and right edges of the first grid. In order to reduce the error, the average is taken to obtain the second fitting function, which can reduce the adjustment error.
[0122] Using the first fitting function and the second fitting function to transform the coordinates of the points in the first grid to obtain the first transformed grid comprises the following steps:
[0123] The coordinates (x, y) of the points in the first grid are transformed to obtain (xf(y), yg(x)), where f(y) is the second fitting function and g(x) is the first fitting function.
[0124] For the coordinates (x, y) of each point in the first grid, it will be displaced as the first grid is deformed. In order to eliminate this deformation, (x, y) needs to be processed. Substitute x into the first fitting function to obtain g(x). The corresponding g(x) describes the degree of deformation of the upper and lower edges of the first grid, which is longitudinal. Therefore, in order to eliminate the longitudinal deformation, g(x) needs to be subtracted from the longitudinal coordinate y. The processing of the horizontal coordinate is similar, thereby making the edge of the first grid linear.
[0125] Using the coordinate transformation formula, mapping the first transformed grid to a grid of the same size as the second grid comprises the following steps:
[0126] The coordinate transformation formula includes a first coordinate transformation formula and a second coordinate transformation formula, the first coordinate transformation formula is used to transform the abscissa of the first transformation grid, and the second coordinate transformation formula is used to transform the ordinate of the first transformation grid;
[0127] The first coordinate transformation formula is as follows:
[0128]
[0129] Wherein, a is the minimum value of the abscissa of the first transformed grid, b is the maximum value of the abscissa of the first transformed grid, c is the minimum value of the abscissa of the first grid, and d is the minimum value of the abscissa of the first grid. is the abscissa of the point in the first transformed grid, is the horizontal coordinate of the point in the first grid, and For the corresponding relationship;
[0130] The second coordinate transformation formula is as follows: ,
[0131] Wherein, h is the minimum value of the ordinate of the first transformed grid, i is the maximum value of the ordinate of the first transformed grid, j is the minimum value of the ordinate of the first grid, k is the minimum value of the ordinate of the first grid, is the ordinate of the point in the first transformed grid, is the ordinate of the point in the first grid, and For the corresponding relationship.
[0132] Reference Figure 8As shown, the image correction device uses a correction mechanism to correct an image taken by a large depth of field ultra-wide-angle lens, including the following steps:
[0133] Acquire an actual imaging picture, divide the actual imaging picture according to the grid division in the grid imaging picture, obtain at least one actual grid, and place the first grid and the actual grid at the same relative position;
[0134] When correcting the actual grid, the correction is performed using a mapping mechanism of the first grid corresponding to the actual grid.
[0135] Furthermore, the present solution also proposes a storage medium having a computer-readable program stored thereon, and when the computer-readable program is called, the low-distortion control method for the ultra-wide-angle lens with a large depth of field is executed.
[0136] It is understandable that the storage medium may be a magnetic medium, such as a floppy disk, a hard disk, or a magnetic tape; an optical medium, such as a DVD; or a semiconductor medium, such as a solid state drive (SSD).
[0137] In summary, the advantages of the present invention are: by performing grid division, different mapping mechanisms are formed for different grids, so that the captured image can be divided in the same way, and the same mapping mechanism is used for correction in grids with the same relative position, thereby a more reasonable stretching ratio can be assigned to each grid. This method takes into account the actual situation that the degree of deformation at different positions in the imaged image is different, rather than stretching all areas at the same ratio, which can reduce the distortion of the corrected image and thus ensure the image imaging effect.
[0138] The above shows and describes the basic principles, main features and advantages of the present invention. Those skilled in the art should understand that the present invention is not limited to the above embodiments. The above embodiments and descriptions only describe the principles of the present invention. The present invention may be subject to various changes and improvements without departing from the spirit and scope of the present invention. These changes and improvements fall within the scope of the present invention. The scope of protection claimed by the present invention is defined by the attached claims and their equivalents.
Claims
1. A low deformation control method for a large depth of field ultra-wide angle lens, characterized in that: include: An image correction device is provided inside the ultra-wide-angle lens with a large depth of field; A preset grid screen is formed, wherein the grid located at the center of the preset grid screen is a reference grid, the reference grid is black, and all grids other than the reference grid are white, and the preset grid screen is composed of grids of equal size tiled; The preset grid screen is photographed, the reference grid is located within the shooting sight, the preset grid screen is larger than the shooting sight, the shooting angle is perpendicular to the preset grid screen, and a grid imaging screen of the preset grid screen captured by an ultra-wide-angle lens with a large depth of field is obtained; In the preset grid picture, an original grid picture corresponding to the grid imaging picture is obtained, the grid in the original grid picture is used as the second grid, and the grid in the grid imaging picture is used as the first grid; Establishing a correspondence between the grid imaging picture and the grids in the corresponding original grid picture; forming a mapping mechanism that maps the first grid to a grid of equal size to the second grid, and pairing the mapping mechanism with the first grid mapping; Aggregating at least one first grid and its corresponding mapping mechanism to obtain a correction mechanism; The correction mechanism is stored in an image correction device, and the image correction device uses the correction mechanism to correct the image taken by the ultra-wide-angle lens with a large depth of field; The mapping mechanism for mapping the first grid to a grid of the same size as the second grid comprises the following steps: Coordinate modeling of grid imaging images; Based on the upper and lower edges of the first grid, a first fitting function is formed, Based on the left and right edges of the first grid, forming a second fitting function; Using the first fitting function and the second fitting function to perform coordinate transformation on the coordinates of the points in the first grid, to obtain a first transformed grid; Using a coordinate transformation formula, mapping the first transformed grid to a grid of the same size as the second grid; The forming of a first fitting function based on the upper and lower edges of the first grid comprises the following steps: Uniformly select at least one upper identification point on the upper edge of the first grid, and obtain the minimum value of the ordinate of the upper identification point as the upper eigenvalue; Subtract the upper eigenvalue from the ordinate of the upper identification point to obtain the upper identification point of the feature; Fitting the abscissa and ordinate of the identification point on at least one feature to obtain an upper fitting function, wherein the abscissa of the identification point on the feature is an independent variable, and the ordinate of the identification point on the feature is a dependent variable; Uniformly select at least one lower identification point on the lower edge of the first grid, and obtain the minimum value of the ordinate of the lower identification point as the lower eigenvalue; Subtract the lower eigenvalue from the ordinate of the lower identification point to obtain the characteristic lower identification point; Fitting the abscissa and ordinate of at least one feature identification point to obtain a lower fitting function, wherein the abscissa of the feature identification point is an independent variable and the ordinate of the feature identification point is a dependent variable; The lower fitting function and the upper fitting function are averaged to obtain the first fitting function; The forming of the second fitting function based on the left and right edges of the first grid comprises the following steps: Uniformly select at least one right identification point on the right edge of the first grid, and obtain the minimum value of the horizontal coordinate of the right identification point as the right eigenvalue; Subtract the right eigenvalue from the horizontal coordinate of the right identification point to obtain the characteristic right identification point; Fitting the ordinate and abscissa of at least one characteristic right identification point to obtain a right fitting function, wherein the ordinate of the characteristic right identification point is an independent variable, and the abscissa of the characteristic right identification point is a dependent variable; Uniformly select at least one left identification point on the left edge of the first grid, and obtain the minimum value of the horizontal coordinate of the left identification point as the left eigenvalue; Subtract the left eigenvalue from the horizontal coordinate of the left identification point to obtain the characteristic left identification point; Fitting the ordinate and abscissa of at least one characteristic left identification point to obtain a left fitting function, wherein the ordinate of the characteristic left identification point is an independent variable, and the abscissa of the characteristic left identification point is a dependent variable; The left fitting function and the right fitting function are averaged to obtain the second fitting function; The method of using the first fitting function and the second fitting function to transform the coordinates of the points in the first grid to obtain the first transformed grid comprises the following steps: Transform the coordinates (x, y) of the points in the first grid to obtain (xf(y), yg(x)), where f(y) is the second fitting function and g(x) is the first fitting function; The method of using the coordinate transformation formula to map the first transformed grid to a grid of the same size as the second grid comprises the following steps: The coordinate transformation formula includes a first coordinate transformation formula and a second coordinate transformation formula, the first coordinate transformation formula is used to transform the abscissa of the first transformation grid, and the second coordinate transformation formula is used to transform the ordinate of the first transformation grid; The first coordinate transformation formula is as follows: , Wherein, a is the minimum value of the abscissa of the first transformed grid, b is the maximum value of the abscissa of the first transformed grid, c is the minimum value of the abscissa of the first grid, and d is the minimum value of the abscissa of the first grid. is the abscissa of the point in the first transformed grid, is the horizontal coordinate of the point in the first grid, and For the corresponding relationship; The second coordinate transformation formula is as follows: , Wherein, h is the minimum value of the ordinate of the first transformed grid, i is the maximum value of the ordinate of the first transformed grid, j is the minimum value of the ordinate of the first grid, k is the minimum value of the ordinate of the first grid, is the ordinate of the point in the first transformed grid, is the ordinate of the point in the first grid, and For the corresponding relationship.
2. The low deformation control method of a large depth of field ultra-wide angle lens according to claim 1, characterized in that: The forming of the preset grid screen comprises the following steps: Divide the preset scenery surface using equally spaced horizontal lines and equally spaced vertical lines, wherein the spacing between adjacent horizontal lines is a preset distance, and the spacing between adjacent vertical lines is a preset distance; A grid formed by at least one horizontal line and a vertical line forms a preset grid picture on the preset scenery surface.
3. The low deformation control method of a large depth of field ultra-wide angle lens according to claim 2, characterized in that: The step of obtaining the original grid picture corresponding to the grid imaging picture in the preset grid picture comprises the following steps: Identify the position of the black grid in the grid imaging image as the calibration grid; Identifying grids at the edge of the grid imaging picture to obtain at least one first edge grid; Get the path coordinates (q, p) from the first edge grid to the calibration grid. The path coordinates (q, p) satisfy that the calibration grid moves |q| grids horizontally and |p| grids vertically, then the calibration grid coincides with the first edge grid, where |q| is the absolute value of q, and |p| is the absolute value of p. When q is greater than 0, move horizontally to the right, when q is less than 0, move horizontally to the left, when p is greater than 0, move vertically upward, and when p is less than 0, move vertically downward; In the preset grid screen, the reference grid is moved according to the path coordinates to obtain a second edge grid; The area enclosed by at least one second edge grid in the preset grid picture forms an original grid picture corresponding to the grid imaging picture.
4. The low deformation control method of a large depth of field ultra-wide angle lens according to claim 3, characterized in that: The step of establishing a correspondence between the grid imaging picture and the grids in the corresponding original grid picture comprises the following steps: Counting the path coordinates from the first grid to the calibration grid in the grid imaging picture as the first path coordinates; Counting the path coordinates from the second grid to the reference grid in the original grid image as the second path coordinates; A corresponding relationship is established between the first grid and the second grid whose first path coordinates are equal to the second path coordinates.
5. The low deformation control method of a large depth of field ultra-wide angle lens according to claim 4, characterized in that: The image correction device uses a correction mechanism to correct an image taken by an ultra-wide-angle lens with a large depth of field, including the following steps: Acquire an actual imaging picture, divide the actual imaging picture according to the grid division in the grid imaging picture, obtain at least one actual grid, and place the first grid and the actual grid at the same relative position; When correcting the actual grid, the correction is performed using a mapping mechanism of the first grid corresponding to the actual grid.
