CN104867113B - The method and system of perspective image distortion correction - Google Patents

Abstract

The present invention is applicable to the field of image technology, and provides a method and system for correcting image perspective distortion, which is applied to a shooting terminal including two cameras. Taking pictures, acquiring corresponding first images and second images; respectively generating first three-dimensional vector information corresponding to the first image and second three-dimensional vector information corresponding to the second image; according to the obtained first three-dimensional vector information The vector information and the second three-dimensional vector information perform perspective distortion correction on the first image and the second image through a predetermined perspective distortion correction algorithm. Thereby, the present invention can implement perspective distortion correction on a three-dimensional image, and can greatly increase the speed of perspective distortion correction.

Description

Method and system for image perspective distortion correction

technical field

The present invention relates to the field of image capture technology, in particular to a method and system for correcting image perspective distortion.

Background technique

Because the optical system of shooting terminals such as cameras and mobile phones does not work precisely according to the principle of idealized pinhole imaging, there is perspective distortion, and there is an optical distortion error between the actual imaging of the object on the imaging surface of the shooting terminal and the ideal imaging. . In the prior art, there is a technology for correcting lens perspective distortion based on physical distance and direction. It takes several images by focusing the lens, then analyzes the images to determine the focal point to calculate the distance, and then uses the distance parameters to correct the perspective distortion of the image. In the existing technology, it is impossible to take countless photos to calculate the distance, so the correction accuracy is very low and the correction effect is not good; and it takes a long time to take a single focus and exposure each time to take multiple photos, so the speed is slow. In addition, the existing technology can only correct the perspective distortion of a two-dimensional (Two Dimension, 2D) image, but is helpless for the perspective distortion of a three-dimensional (Three Dimension, 3D) image.

In summary, there are obviously inconveniences and defects in the actual use of the prior art, so it is necessary to improve it.

Contents of the invention

In view of the above-mentioned defects, the purpose of the present invention is to provide a method and system for correcting perspective distortion of an image, which can realize perspective distortion correction for three-dimensional images, and can greatly increase the speed of perspective distortion correction.

In order to achieve the above object, the present invention provides a method for correcting image perspective distortion, which is applied to a shooting terminal including two cameras, and the method includes:

Simultaneously photographing the subject through the first camera and the second camera, and acquiring corresponding first images and second images;

respectively generating first three-dimensional vector information corresponding to the first image and second three-dimensional vector information corresponding to the second image;

Perform perspective distortion correction on the first image and the second image through a predetermined perspective distortion correction algorithm according to the obtained first three-dimensional vector information and the second three-dimensional vector information.

According to the method of the present invention, the step of respectively generating the first 3D vector information corresponding to the first image and the second 3D vector information corresponding to the second image includes:

generating the first three-dimensional vector information of each pixel in the first image;

Generate the second three-dimensional vector information of each pixel in the second image.

According to the method of the present invention, the step of respectively generating the first 3D vector information corresponding to the first image and the second 3D vector information corresponding to the second image includes:

Analyzing the overlapping area of the first image and the second image;

Calculate the distance between the original image point corresponding to each pixel point of the first image in the overlapping area and the first camera to obtain the first distance, and according to the first two-dimensional of each pixel point Coordinates and the first distance, generating the first three-dimensional vector information of the first image;

Calculate the distance between the original image point corresponding to each pixel point of the second image in the overlapping area and the second camera to obtain a second distance, and according to the second two-dimensional value of each pixel point Coordinates and the second distance to generate the second three-dimensional vector information of the second image.

According to the method of the present invention, the first image and the second image are performed on the first image and the second image through a predetermined perspective distortion correction algorithm according to the obtained first three-dimensional vector information and the second three-dimensional vector information. The steps of perspective distortion correction include:

calling a predetermined first perspective distortion correction parameter according to the first three-dimensional vector information, and performing perspective distortion correction on the overlapped region of the first image through the perspective distortion correction algorithm;

Calling a predetermined second perspective distortion correction parameter according to the second three-dimensional vector information, and performing perspective distortion correction on the overlapped region of the second image through the perspective distortion correction algorithm.

According to the method of the present invention, the first image and the second image are performed on the first image and the second image through a predetermined perspective distortion correction algorithm according to the obtained first three-dimensional vector information and the second three-dimensional vector information. The steps after perspective distortion correction include:

If an image display instruction is received, determine the type of the image display instruction;

If the image display instruction is a two-dimensional image display instruction, intercepting the corrected first image or the overlapping area in the second image for display;

If the image display instruction is a three-dimensional image display instruction, perform three-dimensional modulation and display on the corrected first image and the second image.

The present invention also provides a system for correcting image perspective distortion, which is applied to a shooting terminal including two cameras, and the system includes:

An image acquisition module, configured to simultaneously photograph the subject through the first camera and the second camera, and acquire corresponding first images and second images;

an information generating module, configured to generate first three-dimensional vector information corresponding to the first image and second three-dimensional vector information corresponding to the second image;

An image correction module, configured to perform perspective distortion correction on the first image and the second image through a predetermined perspective distortion correction algorithm according to the obtained first three-dimensional vector information and the second three-dimensional vector information.

According to the system of the present invention, the information generation module includes:

A first generating submodule, configured to generate the first three-dimensional vector information of each pixel in the first image;

The second generating submodule is configured to generate the second three-dimensional vector information of each pixel in the second image.

According to the system of the present invention, the information generation module includes:

an area analysis submodule, configured to analyze the overlapping area of the first image and the second image;

The first generation sub-module is used to calculate the distance between the original image point corresponding to each pixel point of the first image in the overlapping area and the first camera to obtain the first distance, and according to each The first two-dimensional coordinates and the first distance of each of the pixel points are used to generate the first three-dimensional vector information of the first image;

The second generation sub-module is used to calculate the distance between the original image point corresponding to each pixel point of the second image in the overlapping area and the second camera to obtain the second distance, and according to each The second two-dimensional coordinates of the pixel points and the second distance are used to generate the second three-dimensional vector information of the second image.

According to the system of the present invention, the image correction module includes:

The first correction submodule is configured to call a predetermined first perspective distortion correction parameter according to the first three-dimensional vector information, and perform perspective distortion correction on the overlapped region of the first image through the perspective distortion correction algorithm;

The second correcting submodule is configured to call a predetermined second perspective distortion correction parameter according to the second three-dimensional vector information, and perform perspective distortion correction on the overlapped region of the second image through the perspective distortion correction algorithm.

According to the system of the present invention, it also includes:

An instruction receiving module, configured to determine the type of the image display instruction if an image display instruction is received after performing perspective distortion correction on the first image and the second image;

The first display module is configured to intercept the corrected first image or the overlapping area in the second image for display if the image display instruction is a two-dimensional image display instruction;

The second display module is configured to perform three-dimensional modulation and display on the corrected first image and the second image if the image display instruction is a three-dimensional image display instruction.

When shooting, the shooting terminal of the present invention acquires two images of the object at one time through two cameras, and generates the three-dimensional vector information of the two images respectively, which can greatly shorten the need for multiple focusing shots at different distances in the prior art. The time when the image acquires depth information; then the perspective distortion correction algorithm is called to correct the perspective distortion of the 3D image. Thereby, the present invention can implement perspective distortion correction on a three-dimensional image, and can greatly increase the speed of perspective distortion correction. Preferably, the present invention can capture the three-dimensional vector information of each pixel point in the captured image through two cameras, and perform perspective distortion correction of the three-dimensional image according to the three-dimensional vector information of each pixel point. The accuracy of local large-scale correction for several depth values obtained from photos is greatly improved by the present invention.

Description of drawings

Fig. 1 is a schematic structural diagram of a system for correcting image perspective distortion in the present invention;

Fig. 2 is a schematic structural diagram of a system for correcting preferred image perspective distortion in the present invention;

Fig. 3 is a flow chart of a method for correcting image perspective distortion in the present invention;

Fig. 4 is a flowchart of a method for correcting image perspective distortion in the first embodiment of the present invention;

Fig. 5 is a schematic diagram of the realization of the image perspective distortion correction method in the second embodiment of the present invention.

detailed description

In order to make the object, technical solution and advantages of the present invention clearer, the present invention will be further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described here are only used to explain the present invention, not to limit the present invention.

1 is a schematic structural diagram of a system for correcting image perspective distortion according to the present invention, which is applied to a shooting terminal including two cameras. The system 100 includes an image acquisition module 10, an information generation module 20, and an image correction module 30, wherein:

The image acquisition module 10 is configured to simultaneously photograph the subject through the first camera and the second camera, and acquire corresponding first images and second images. The subject may be any one or more of people, animals, plants, buildings, mountains, water, sky and the like. The first camera and the second camera are two cameras on the same side of the shooting terminal, which can be left and right cameras or up and down cameras. The two cameras can shoot at the same time and the content of the shots is basically the same, that is, the first image and the second The image content of the images is basically the same, and there are many overlapping areas. At present, the two cameras have been widely used in shooting terminals such as cameras, mobile phones, and tablet computers.

The information generation module 20 is configured to respectively generate first 3D vector information corresponding to the first image and second 3D vector information corresponding to the second image. Here, the first three-dimensional vector information can be generated according to the first image, and the first three-dimensional vector information can constitute the first three-dimensional image; the second three-dimensional vector information can be generated according to the second image, and the second three-dimensional vector information can constitute the second three-dimensional vector information. image. Preferably, the three-dimensional vector information in the present invention refers to the plane two-dimensional coordinates of the image and the vector information of the distance between the original image corresponding to each pixel in the image and the camera. The distance is preferably the plane two-dimensional coordinates of the image and the distance between the original image point corresponding to each corresponding pixel in the image and the camera plane. Of course, the distance can also be the plane two-dimensional coordinates of the image and each corresponding pixel in the image. The distance between the corresponding original image point and the center point of the camera. The original image point is the point of the actual object corresponding to the pixel point in the image.

The image correction module 30 is configured to perform perspective distortion correction on the first image and the second image through a predetermined perspective distortion correction algorithm according to the obtained first 3D vector information and second 3D vector information. At least one perspective distortion correction algorithm may be pre-stored in the shooting terminal, such as a lens perspective distortion correction algorithm based on object distance and direction. Since the perspective distortion correction algorithm is a prior art, it will not be repeated here. Specifically, perform perspective distortion correction on the first image according to the perspective distortion correction algorithm and the first 3D vector information; and perform perspective distortion correction on the second image according to the perspective distortion correction algorithm and the second 3D vector information.

When shooting, the shooting terminal of the present invention acquires two images of the object at one time through two cameras, and generates the three-dimensional vector information of the two images respectively, which can greatly shorten the need for multiple focusing shots at different distances in the prior art. The time when the image acquires depth information; then the perspective distortion correction algorithm is called to correct the perspective distortion of the 3D image, so as to realize the perspective distortion correction of the 3D image.

Fig. 2 is a schematic structural diagram of a preferred image perspective distortion correction system of the present invention, which is applied to a shooting terminal including two cameras, and the shooting terminal can be a camera, a mobile phone, a tablet computer, etc., and the system 100 includes at least an image acquisition module 10. Information generation module 20 and image correction module 30, wherein:

The image acquisition module 10 is configured to simultaneously photograph the subject through the first camera and the second camera, and acquire corresponding first images and second images.

The information generation module 20 is configured to respectively generate first 3D vector information corresponding to the first image and second 3D vector information corresponding to the second image. Preferably, the information generating module 20 includes:

The first generating sub-module 21 is configured to generate first three-dimensional vector information of each pixel in the first image.

The second generating sub-module 22 is configured to generate second three-dimensional vector information of each pixel in the second image.

Obtaining three-dimensional vector information through one shot with two cameras greatly shortens the time required to obtain depth information by focusing and shooting multiple photos at different distances in the prior art. The accuracy of pixel-based 3D vector information correction is tens of millions of times higher than that of local large-scale correction based on several depth values obtained from several photos in the existing technology, which can greatly improve the speed and accuracy of perspective distortion correction.

Better yet, the information generating module 30 may also include:

The area analysis sub-module 23 is configured to analyze the overlapping area of the first image and the second image. The analysis of the overlapping area involves image recognition technology. Based on RGB (Red, Green, Blue, red, green and blue) values, brightness, grayscale and other pixel features to find the boundary line between similar points and dissimilar points, the second point can be analyzed. The overlapping area of the first image and the second image.

The first generation sub-module 21 is used to calculate the distance between the original image point corresponding to each pixel point of the first image in the overlapping area and the first camera to obtain the first distance, and according to the first two-dimensional coordinates of each pixel point and the first distance to generate first three-dimensional vector information of the first image, where the first three-dimensional vector information can be represented by (x, y, z). That is, the first 3D vector information in the present invention refers to the plane 2D coordinates of the first image and the vector information of the distance between the original image corresponding to each pixel in the first image and the first camera.

The second generation sub-module 22 is used to calculate the distance from the original image point corresponding to each pixel point of the second image in the overlapping area to the second camera to obtain the second distance, and according to the second two-dimensional coordinates of each pixel point and the second distance to generate second three-dimensional vector information of the second image, where the second three-dimensional vector information can be represented by (x, y, z). That is, the second 3D vector information in the present invention refers to the plane 2D coordinates of the second image and the vector information of the distance between the original image corresponding to each pixel in the second image and the second camera.

The image correction module 30 is configured to perform perspective distortion correction on the first image and the second image through a predetermined perspective distortion correction algorithm according to the obtained first 3D vector information and second 3D vector information.

Preferably, the image correction module 30 includes:

The first correction sub-module 31 is configured to call a predetermined first perspective distortion correction parameter according to the first three-dimensional vector information, and perform perspective distortion correction on the overlapped area of the first image through a perspective distortion correction algorithm.

The second correction sub-module 32 is configured to call a predetermined second perspective distortion correction parameter according to the second three-dimensional vector information, and perform perspective distortion correction on the overlapped area of the second image through a perspective distortion correction algorithm. The second perspective distortion correction parameter may be the same as or different from the first perspective distortion correction parameter.

Even better, the system 100 for correcting image perspective distortion may also include:

The instruction receiving module 40 is configured to determine the type of the image display instruction if an image display instruction is received after performing perspective distortion correction on the first image and the second image. The type of the image display instruction may be a two-dimensional image display instruction or a three-dimensional image display instruction.

The first display module 50 is configured to, if the image display instruction is a two-dimensional image display instruction, intercept and display the overlapped area in the corrected first image or the second image.

The second display module 60 is configured to perform three-dimensional modulation and display on the corrected first image and second image if the image display instruction is a three-dimensional image display instruction. Here, the simultaneous display of the corrected first image and the second image can constitute a three-dimensional image.

The invention describes the technique of capturing the three-dimensional vector information of each pixel point in the photographed image through two cameras, and calling the perspective distortion parameters of the corresponding lens to correct the three-dimensional image distortion, which greatly improves the speed and accuracy of perspective distortion correction, and makes up for the existing There are drawbacks to the technology's inability to perform perspective distortion correction on 3D images. It is also possible to use the corrected two images for three-dimensional display, and the user experience is better.

FIG. 3 is a flow chart of a method for correcting image perspective distortion of the present invention, which is applied to a shooting terminal including two cameras. The method can be implemented by the system 100 for correcting image perspective distortion as shown in FIG. 1 or FIG. 2 , and the method Including:

In step S301, the subject is photographed simultaneously by the first camera and the second camera, and corresponding first images and second images are acquired.

The subject may be any one or more of people, animals, plants, buildings, mountains, water, sky and the like. The first camera and the second camera are two cameras on the same side of the shooting terminal, which can be left and right cameras or up and down cameras. The two cameras can shoot at the same time and the content of the shots is basically the same, that is, the first image and the second The image content of the images is basically the same, and there are many overlapping areas.

Step S302, respectively generating first 3D vector information corresponding to the first image and second 3D vector information corresponding to the second image.

Here, the first three-dimensional vector information can be generated according to the first image, and the first three-dimensional vector information can constitute the first three-dimensional image; the second three-dimensional vector information can be generated according to the second image, and the second three-dimensional vector information can constitute the second three-dimensional vector information. image. In this step, preferably, the first three-dimensional vector information of each pixel in the first image is generated, and the second three-dimensional vector information of each pixel in the second image is generated. The accuracy of pixel-based 3D vector information correction is tens of millions of times higher than that of local large-scale correction based on several depth values obtained from several photos in the existing technology, which can greatly improve the speed and accuracy of perspective distortion correction. That is, the three-dimensional vector information in the present invention refers to the plane two-dimensional coordinates of the image and the vector information of the distance between the original image corresponding to each pixel in the image and the camera. The distance is preferably the plane two-dimensional coordinates of the image and the distance between the original image point corresponding to each corresponding pixel in the image and the camera plane. Of course, the distance can also be the plane two-dimensional coordinates of the image and each corresponding pixel in the image. The distance between the corresponding original image point and the center point of the camera. The original image point is the point of the actual object corresponding to the pixel point in the image.

Step S303 , performing perspective distortion correction on the first image and the second image through a predetermined perspective distortion correction algorithm according to the obtained first 3D vector information and second 3D vector information.

At least one perspective distortion correction algorithm may be pre-stored in the shooting terminal, such as a lens perspective distortion correction algorithm based on object distance and direction. Since the perspective distortion correction algorithm is a prior art, it will not be repeated here. Specifically, perform perspective distortion correction on the first image according to the perspective distortion correction algorithm and the first 3D vector information; and perform perspective distortion correction on the second image according to the perspective distortion correction algorithm and the second 3D vector information.

FIG. 4 is a flowchart of a method for correcting image perspective distortion in the first embodiment of the present invention, which is applied to a shooting terminal including two cameras. The method can be implemented by the system 100 for correcting image perspective distortion as shown in FIG. 2 , so The methods described include:

In step S401, the subject is photographed simultaneously by the first camera and the second camera, and corresponding first images and second images are acquired.

Step S402, analyzing the overlapping area of the first image and the second image. The analysis of the overlapping area involves image recognition technology. Based on pixel features such as RGB value, brightness, and gray scale, the boundary line between similar points and dissimilar points can be found, and the overlapping area of the first image and the second image can be analyzed.

Step S403, calculate the distance between the original image point corresponding to each pixel point of the first image in the overlapping area and the first camera, obtain the first distance, and generate The first three-dimensional vector information of the first image.

That is, the first three-dimensional vector information in the present invention refers to the plane two-dimensional coordinates of the first image and the vector information of the distance between the corresponding pixel point and the first camera in the first image, and the first three-dimensional vector information can be used (x, y , z) to represent.

Step S404, call a predetermined first perspective distortion correction parameter according to the first 3D vector information, and perform perspective distortion correction on the overlapping area of the first image through a perspective distortion correction algorithm.

Step S405, calculate the distance between the original image point corresponding to each pixel point of the second image in the overlapping area and the second camera, obtain the second distance, and generate Second 3D vector information of the second image.

That is, the second three-dimensional vector information in the present invention refers to the plane two-dimensional coordinates of the second image and the vector information of the distance between the original image point corresponding to each pixel in the second image and the second camera, and the second three-dimensional vector information can be Use (x, y, z) to represent.

In step S406, a predetermined second perspective distortion correction parameter is invoked according to the second three-dimensional vector information, and perspective distortion correction is performed on the overlapping area of the second image through a perspective distortion correction algorithm.

The second perspective distortion correction parameter may be the same as or different from the first perspective distortion correction parameter.

Step S407, receiving an image display instruction.

The type of the image display instruction may be a two-dimensional image display instruction or a three-dimensional image display instruction.

Step S408, determine the type of the image display instruction, if the image display instruction is a two-dimensional image display instruction, execute step S409, and if the image display instruction is a three-dimensional image display instruction, execute step S410.

Step S409, if the image display instruction is a two-dimensional image display instruction, intercepting the corrected first image or the overlapping area in the second image for display.

Step S410, if the image display instruction is a 3D image display instruction, perform 3D modulation and display on the corrected first image and second image.

Here, the simultaneous display of the corrected first image and the second image can constitute a three-dimensional image.

Fig. 5 is a schematic diagram of the realization of the image perspective distortion correction method in the second embodiment of the present invention, which is applied to a shooting terminal including left and right cameras, mainly including:

1. Simultaneously photograph the subject through the first camera and the second camera, and obtain corresponding first images and second images. two

1. Turn on the two camera modes, and compose the subject through the display screen;

2. Specifically, the left and right cameras take pictures of the subject at the same time, and obtain and save two left and right images, that is, image L (ie, the first image) and image R (ie, the second image);

2. Generating first three-dimensional vector information corresponding to the first image and second three-dimensional vector information corresponding to the second image respectively.

Analyze the overlapping area of image L and image R. For the overlapping area of the two images L and R (the snowflake point area in Figure 5), call the pre-stored algorithm in the shooting terminal based on each pixel to perform geometric and mathematical calculations, and obtain each pixel in the image. The original image point corresponding to each pixel is at a distance z from the camera, and the 3D vector information document of the generated image is as follows:

(x1,y1,z1);(x2,y1,z2)………………(xn,y1,zn)

(x1, y2, zn+1); (x2, y2, zn+2)……………(xn, y2, z2n)

. . .

. . .

(x1,ym,z(m-1)n+1);(x2,ym,z(m-1)n+2)……(xn,ym,zmn)

Stage this document. There are three coordinate systems here: the first coordinate of the image L, the second coordinate of the image R, and the third coordinate of the overlapping image formed by the overlapping parts of the two images. Wherein the first coordinate of the image L and the second coordinate of the image R are a two-dimensional coordinate system, and the third coordinate of the overlapped image is a three-dimensional coordinate system. The origin of the X-axis in the third coordinate (x, y, z) is the abscissa x1 of the first column of coincident pixels in the first coordinate of the image L, and the origin of the Y-axis in the third coordinate (x, y, z) It is the vertical coordinate ym of the pixel point in the bottom line in the overlapping image. In each coordinate system, the coordinate origin refers to the last pixel point in the lower left corner of each image (image L, image R or coincident image). Remember the critical points of the overlapping and non-overlapping parts of the left and right images L and R as shown in the x1 and x2 data on the abscissa. Because the coordinates of images R and L are not uniform, x1 or x2 is needed for conversion. If the coordinates of image L are used as a reference, x1 will be used, and if the coordinates of image R are used as a reference, x2 will be used. This embodiment is based on image L The coordinates of are used as an example to illustrate. The coordinate origin of the third coordinate is transformed based on the coordinate origin of the first coordinate or the second coordinate, and the conversion uses x1 or x2.

3. Perform perspective distortion correction on the first image and the second image through a predetermined perspective distortion correction algorithm according to the obtained first 3D vector information and second 3D vector information.

According to the generated 3D vector information, the perspective distortion correction coefficient table provided by the original lens manufacturer is called and the perspective distortion correction is performed on the overlapped area (the oblique line in the figure) in the two images through the calibration algorithm (the non-overlapping area correction value is 0 by default). During the correction process, the distortion coefficient M related to the three-dimensional vector information (x, y, z) corresponds to the pixels of the image L captured by the left camera, and M corresponds to the pixels of the image R captured by the right camera; finally, only the corrected left and right images L are saved _J , R _J , delete other temporary data or files.

If you only need to display the image in two dimensions in the later stage, you can cut out the overlapping area (the oblique line in the figure) in any of the two images and display it; _J , R _J can perform three-dimensional modulation and play simultaneously.

In summary, when the shooting terminal of the present invention is shooting, it acquires two images of the subject at one time through two cameras, and generates the three-dimensional vector information of the two images respectively, which can greatly shorten the distance required by the prior art. The time to obtain depth information by focusing on multiple images at a time; then call the perspective distortion correction algorithm to correct the perspective distortion of the 3D image. Thereby, the present invention can implement perspective distortion correction on a three-dimensional image, and can greatly increase the speed of perspective distortion correction. Preferably, the present invention can capture the three-dimensional vector information of each pixel point in the captured image through two cameras, and perform perspective distortion correction of the three-dimensional image according to the three-dimensional vector information of each pixel point. The accuracy of local large-scale correction for several depth values obtained from photos is greatly improved by the present invention.

Certainly, the present invention also can have other multiple embodiments, without departing from the spirit and essence of the present invention, those skilled in the art can make various corresponding changes and deformations according to the present invention, but these corresponding Changes and deformations should belong to the scope of protection of the appended claims of the present invention.

Claims (6)

1. A method for image perspective distortion correction, characterized in that it is applied to a shooting terminal comprising two cameras, and the method comprises: Simultaneously photographing the subject through the first camera and the second camera, and acquiring corresponding first images and second images; respectively generating first three-dimensional vector information corresponding to the first image and second three-dimensional vector information corresponding to the second image; Perform perspective distortion correction on the first image and the second image through a predetermined perspective distortion correction algorithm according to the obtained first three-dimensional vector information and the second three-dimensional vector information; The step of respectively generating the first three-dimensional vector information corresponding to the first image and the second three-dimensional vector information corresponding to the second image includes: generating the first three-dimensional vector information of each pixel in the first image; generating the second three-dimensional vector information of each pixel in the second image; The step of respectively generating the first 3D vector information corresponding to the first image and the second 3D vector information corresponding to the second image includes: Analyzing the overlapping area of the first image and the second image; Calculate the distance between the original image point corresponding to each pixel point of the first image in the overlapping area and the first camera to obtain the first distance, and according to the first two-dimensional of each pixel point Coordinates and the first distance, generating the first three-dimensional vector information of the first image; Calculate the distance between the original image point corresponding to each pixel point of the second image in the overlapping area and the second camera to obtain a second distance, and according to the second two-dimensional value of each pixel point Coordinates and the second distance to generate the second three-dimensional vector information of the second image. 2. The method according to claim 1, characterized in that, according to the obtained first three-dimensional vector information and the second three-dimensional vector information, through a predetermined perspective distortion correction algorithm, the first image The step of performing perspective distortion correction with the second image comprises: calling a predetermined first perspective distortion correction parameter according to the first three-dimensional vector information, and performing perspective distortion correction on the overlapped region of the first image through the perspective distortion correction algorithm; Calling a predetermined second perspective distortion correction parameter according to the second three-dimensional vector information, and performing perspective distortion correction on the overlapped region of the second image through the perspective distortion correction algorithm. 3. The method according to any one of claims 1-2, characterized in that, according to the obtained first three-dimensional vector information and the second three-dimensional vector information, through a predetermined perspective distortion correction algorithm, the The steps of performing perspective distortion correction on the first image and the second image include: If an image display instruction is received, determine the type of the image display instruction; If the image display instruction is a two-dimensional image display instruction, intercepting the corrected first image or the overlapping area in the second image for display; If the image display instruction is a three-dimensional image display instruction, perform three-dimensional modulation and display on the corrected first image and the second image. 4. A system for image perspective distortion correction, characterized in that it is applied to a shooting terminal including two cameras, and the system includes: An image acquisition module, configured to simultaneously photograph the subject through the first camera and the second camera, and acquire corresponding first images and second images; an information generating module, configured to generate first three-dimensional vector information corresponding to the first image and second three-dimensional vector information corresponding to the second image; An image correction module, configured to perform perspective distortion correction on the first image and the second image through a predetermined perspective distortion correction algorithm according to the obtained first three-dimensional vector information and the second three-dimensional vector information; The information generation module includes: A first generating submodule, configured to generate the first three-dimensional vector information of each pixel in the first image; A second generating submodule, configured to generate the second three-dimensional vector information of each pixel in the second image; The information generation module includes: an area analysis submodule, configured to analyze the overlapping area of the first image and the second image; The first generation sub-module is used to calculate the distance between the original image point corresponding to each pixel point of the first image in the overlapping area and the first camera to obtain the first distance, and according to each The first two-dimensional coordinates and the first distance of each of the pixel points are used to generate the first three-dimensional vector information of the first image; The second generation sub-module is used to calculate the distance between the original image point corresponding to each pixel point of the second image in the overlapping area and the second camera to obtain the second distance, and according to each The second two-dimensional coordinates of the pixel points and the second distance are used to generate the second three-dimensional vector information of the second image. 5. The system according to claim 4, wherein the image correction module comprises: The first correction submodule is configured to call a predetermined first perspective distortion correction parameter according to the first three-dimensional vector information, and perform perspective distortion correction on the overlapped region of the first image through the perspective distortion correction algorithm; The second correcting submodule is configured to call a predetermined second perspective distortion correction parameter according to the second three-dimensional vector information, and perform perspective distortion correction on the overlapped region of the second image through the perspective distortion correction algorithm. 6. The system according to any one of claims 4-5, further comprising: An instruction receiving module, configured to determine the type of the image display instruction if an image display instruction is received after performing perspective distortion correction on the first image and the second image; The first display module is configured to intercept the corrected first image or the overlapping area in the second image for display if the image display instruction is a two-dimensional image display instruction; The second display module is configured to perform three-dimensional modulation and display on the corrected first image and the second image if the image display instruction is a three-dimensional image display instruction.