Method, device and equipment for automatically deleting abnormal pictures for calibration and medium
By calculating and comparing the average projection error values of chart image obtained by the camera module, filtering and deleting images with large errors, the adverse effects of abnormal pictures on camera module calibration are solved, and calibration accuracy and accuracy are improved.
Patent Information
- Application Number
- CN202411302315.1
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-09-18
- Publication Date
- 2025-01-03
AI Technical Summary
When the existing camera module acquires calibration pictures, abnormal pictures appear due to external factors, which in turn affects the calibration accuracy of the camera module.
By acquiring multiple chart images, calculate the average and total average of projection errors of each image, compare the difference with the preset threshold, delete images with large errors, and ensure calibration accuracy.
The calibration accuracy and accuracy of the camera module are improved, and the adverse effects of abnormal pictures on calibration are avoided.
Smart Images
Figure CN119251311A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of camera calibration, and in particular to a method, device, equipment and medium for automatically deleting abnormal images used for calibration. Background Art
[0002] In image measurement or other applications of computer vision, the calibration of imaging parameters is a very critical link and a basic problem of computer vision systems. The accuracy of the calibration directly affects the performance of subsequent visual tasks.
[0003] Although manufacturers provide most of the parameters of the camera and the corresponding camera lens, in actual application, users often need to recalibrate the camera according to actual usage, obtain specific and accurate internal parameters, and obtain the orientation information of the camera space in order to obtain distortion-free images to meet the needs of users.
[0004] At present, the method for internal reference calibration is Zhang's calibration method. First, take a picture of the calibration board to obtain the calibration picture. The calibration board is a chessboard composed of black and white rectangles. Then extract the corner point information of the calibration picture. The corner point refers specifically to the inner corner point on the calibration board. These corner points do not touch the edge of the calibration board. Then, a series of operations are performed on the obtained corner point information to finally complete the calibration of the module internal reference.
[0005] Therefore, the calibration image is critical to the accuracy of the calibration of the camera module. However, when the existing camera module obtains the calibration image, due to the influence of external factors, the obtained calibration image becomes abnormal, which leads to inaccurate calibration of the camera module. Summary of the invention
[0006] The purpose of the embodiments of the present application is to propose a method, device, equipment and medium for automatically deleting abnormal images for calibration to solve the problem of adverse effects of abnormal images on camera module calibration and improve calibration accuracy.
[0007] In order to solve the above technical problems, the embodiment of the present application provides a method for automatically deleting abnormal images for calibration, which adopts the following technical solution, including: Step 100: using the camera module to be calibrated to obtain a plurality of chart images, and using the plurality of chart images to calibrate the camera module to obtain the internal and external parameters of the camera module; Step 200: Obtain all actual feature points of the chart, and project the actual feature points onto the chart image using the internal and external parameters of the camera module to obtain predicted positions of the actual feature points; Step 300: obtaining the actual position of the actual feature point, calculating a first mean value of the projection error between the predicted position and the actual position, and calculating a second mean value of the projection error of all chart images, and calculating a difference between the first mean value and the second mean value of the chart images; Step 400: compare the difference with a preset threshold, and if the difference is greater than the preset threshold, delete the chart image; Step 500: Repeat steps 200-400 until all chart images are determined.
[0008] Furthermore, the method for automatically deleting abnormal images for calibration also includes: Step 600: determine whether the area of the deleted chart image is covered by other chart images; Step 700: If there is no other chart image coverage, it is determined that the calibration has failed.
[0009] Furthermore, the step 100 includes: Step 110, making an actual chart, and fixing the chart on a calibration board; Step 120: fix the camera module, adjust the focal length and exposure parameters of the camera module to obtain a clear plane image, shoot the chart from different angles, and obtain the multiple chart images.
[0010] Furthermore, after step 120, step 100 further includes: Step 130: Obtain predicted feature points of the chart image using a corner point detection method; Step 140: acquiring actual feature points corresponding to the predicted feature points according to the actual size and feature point layout of the chart; Step 150: Obtain the intrinsic and extrinsic parameters of the camera module according to the actual feature points and the predicted feature points using the Zhang Zhengyou camera calibration algorithm.
[0011] Furthermore, the step 200 includes: Step 210: projecting the actual feature points onto a corresponding chart image according to the internal and external parameters of the camera module; Step 220: Obtain the predicted position of the actual feature point in the image coordinate system, and calculate the predicted coordinates of the predicted position.
[0012] Furthermore, the step 300 includes: Step 310: Calculate and obtain the actual coordinates of the actual feature points using an image processing algorithm; Step 320: Calculate the projection error between the predicted coordinates and the actual coordinates using a Euclidean distance algorithm; Step 330: Calculate the projection errors of all actual feature points on the chart, and obtain the first mean value.
[0013] Furthermore, after step 330, step 300 further includes: Step 340: Obtain a first mean value of projection errors of all actual feature points corresponding to all chart images; Step 350: Calculate a second mean of the projection errors of all chart images according to the first mean.
[0014] In order to solve the above technical problems, the embodiment of the present application further provides a device for automatically deleting abnormal images for calibration, which adopts the method for automatically deleting abnormal images for calibration described in the first aspect, including: A calibration module, used to obtain a plurality of chart images using the camera module to be calibrated, and calibrate the camera module using the plurality of chart images to obtain the internal and external parameters of the camera module; An acquisition module, used to acquire all actual feature points of the chart and project the actual feature points onto the chart image using the internal and external parameters of the camera module to acquire predicted positions of the actual feature points; a calculation module, configured to calculate a first mean value of the projection error between the predicted position and the actual position according to the actual position of the actual feature point, calculate a second mean value of the projection error of all chart images, and calculate a difference between the first mean value and the second mean value of the chart image, The deletion module is used to compare the difference with a preset threshold value, and if the difference is greater than the preset threshold value, delete the chart image.
[0015] In order to solve the above technical problems, an embodiment of the present application also provides a computer device, including a memory and a processor, wherein the memory stores computer-readable instructions, and when the processor executes the computer-readable instructions, the steps of the method for automatically deleting calibrated abnormal images as described above are implemented.
[0016] In order to solve the above technical problems, an embodiment of the present application also provides a computer-readable storage medium, on which computer-readable instructions are stored. When the computer-readable instructions are executed by a processor, the steps of the method for automatically deleting calibrated abnormal images as described above are implemented.
[0017] Compared with the prior art, the embodiments of the present application mainly have the following technical effects: by projecting the actual feature points of the calibrated chart onto the chart image, and calculating the first mean of the projection error of each chart image, and then obtaining the second mean of the projection error of all chart images, the difference between the first mean and the second mean is compared with a preset threshold, so as to screen out the chart images with larger errors and delete them, avoiding the deletion of pictures with normal errors, and solving the problem of adverse effects of abnormal pictures on the calibration of the camera module, so that when calibrating again, the calibration precision and accuracy of the camera module are improved. BRIEF DESCRIPTION OF THE DRAWINGS
[0018] In order to more clearly illustrate the scheme in the present application, a brief introduction is given below to the drawings required for use in the description of the embodiments of the present application. Obviously, the drawings described below are some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.
[0019] Figure 1 A flowchart of an embodiment of a method for automatically deleting abnormal images for calibration according to the present application; Figure 2 yes Figure 1 A flowchart of a specific implementation of step 100; Figure 3 yes Figure 2 A flowchart of a specific implementation method after step 120; Figure 4 yes Figure 1 A flowchart of a specific implementation of step 200; Figure 5 yes Figure 1 A flowchart of a specific implementation of step 300; Figure 6 yes Figure 5 A flowchart of a specific implementation of step 330; Figure 7 yes Figure 1 A flowchart of a specific implementation method after step 500; Figure 8 It is a structural schematic diagram of an embodiment of a device for automatically deleting abnormal images for calibration according to the present application; Fig. 9 It is a structural diagram of an embodiment of a computer device according to the present application. DETAILED DESCRIPTION
[0020] The following will be combined with the accompanying drawings in the invention to clearly and completely describe the technical solutions in the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments in the present invention, other embodiments obtained by ordinary technicians in this field without creative work are all within the scope of protection of the present invention.
[0021] Reference to "embodiments" herein means that a particular feature, structure, or characteristic described in conjunction with the embodiments may be included in at least one embodiment of the present application. The appearance of the phrase in various locations in the specification does not necessarily refer to the same embodiment, nor is it an independent or alternative embodiment that is mutually exclusive with other embodiments. It is explicitly and implicitly understood by those skilled in the art that the embodiments described herein may be combined with other embodiments.
[0022] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by technicians in the technical field to which this application belongs; the terms used in the specification of the application are only for the purpose of describing specific embodiments and are not intended to limit this application; the terms "including" and "having" in the specification and claims of this application and the above-mentioned figure descriptions and any variations thereof are intended to cover non-exclusive inclusions.
[0023] In addition, the terms "first" and "second" are used for descriptive purposes only and should not be understood as indicating or implying relative importance or implicitly indicating the number of the indicated technical features. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of the features. In the description of this application, the meaning of "plurality" is two or more, unless otherwise clearly and specifically defined.
[0024] The method for automatically deleting abnormal images for calibration provided in the embodiment of the present application can be applied to a computer device with the ability to run instructions or programs. Among them, the computer device can be, but is not limited to, various servers, workstations, personal computers, laptops, etc. Running on different computing devices is only a difference in the execution subject of the scheme. People in this field can foresee that running on different computing devices can produce the same technical effect. The present invention is described in detail below through specific embodiments.
[0025] The purpose of the embodiments of the present application is to propose a method, device, equipment and medium for automatically deleting abnormal images for calibration to solve the problem of adverse effects of abnormal images on camera module calibration and improve calibration accuracy.
[0026] In order to solve the above technical problems, the embodiment of the present application provides a method for automatically deleting abnormal images for calibration, which adopts the following technical solutions: Figure 1 , Figure 1 A flowchart of an embodiment of a method for automatically deleting abnormal images for calibration according to the present application; comprising: S100: using the camera module to be calibrated to obtain a plurality of chart images, and using the plurality of chart images to calibrate the camera module to obtain the internal and external parameters of the camera module.
[0027] In this embodiment, when the camera module is calibrated using a Chart, a Chart of known dimensions and features is photographed using the camera module to be calibrated, and a correspondence between the Chart image coordinates and the real-world Chart coordinates is established to determine the internal and external parameters of the camera module.
[0028] The internal parameters of the camera module mainly include focal length: which determines the viewing angle and magnification of the camera. A shorter focal length can capture a wider field of view, while a longer focal length can magnify distant objects; principal point coordinates: a point on the image plane, the intersection of light on the image plane after passing through the center of the lens. Radial distortion coefficient: Due to the shape of the lens, the light will bend when passing through the lens, resulting in radial distortion of the image. The radial distortion coefficient is used to correct this distortion. Tangential distortion coefficient: In addition to radial distortion, the lens may also cause tangential distortion. The tangential distortion coefficient is used to correct this type of distortion. External parameters refer to the rotation matrix and translation matrix of each chart image.
[0029] In some optional embodiments, such as Figure 2 , Figure 2 yes Figure 1 A flowchart of a specific implementation method of S100; S100 includes: S110, making an actual chart and fixing the chart on a calibration plate; S120, fixing the camera module, adjusting the focal length and exposure parameters of the camera module to obtain a clear plane image, shooting the chart from different angles, and obtaining multiple chart images.
[0030] In this embodiment, the Chart can be selected as a checkerboard or a circular array. Ensure that the size of the Chart is accurately known and has enough feature points to improve the accuracy of calibration. Print the Chart on a flat paper or make it into a physical calibration plate. Fix the camera module in a stable position and adjust parameters such as focus and exposure to obtain a clear image. Take multiple Chart images from different angles and positions to cover the field of view of the entire camera module as much as possible. Ensure that the Chart remains still during the shooting process and that the feature points of the Chart in the image are clearly visible.
[0031] In this embodiment, multiple chart images are obtained to prepare for calibration of the camera module. By shooting the chart from different angles and obtaining multiple chart images, calibration accuracy is improved.
[0032] In some optional embodiments, such as Figure 3 , Figure 3 yes Figure 2 A flowchart of a specific implementation method after S120 in the figure; after S120, S100 also includes: S130, using the corner point detection method to obtain the predicted feature points of the chart image; S140, according to the actual size of the chart and the layout of the feature points, obtaining the actual feature points corresponding to the predicted feature points; S150, according to the actual feature points and the predicted feature points and using the Zhang Zhengyou camera calibration algorithm to obtain the internal and external parameters of the camera module.
[0033] In this embodiment, a corner point detection algorithm is used to identify and detect feature points of the Chart image, that is, to predict feature points.
[0034] The chart can be a checkerboard chart or a circular array chart. For a checkerboard chart, the corners of the checkerboard can be detected; for a circular array chart, the center of the circle can be detected. Get the position (coordinates) of the predicted feature point in the image coordinate system. Based on the known size of the chart and the layout of the actual feature points, determine the position (coordinates) of the corresponding actual feature points in the real-world coordinate system.
[0035] In this embodiment, the camera module adopts Zhang Zhengyou calibration method to solve the intrinsic and extrinsic parameters of the camera according to the matching relationship of feature points.
[0036] In this embodiment, the camera module is calibrated by obtaining the predicted feature points of the chart image and comparing them with the positions (coordinates) of the corresponding actual feature points, thereby calculating the intrinsic and extrinsic parameters of the camera module.
[0037] S200, obtaining all actual feature points of the chart, and projecting the actual feature points onto the chart image using the internal and external parameters of the camera module to obtain the predicted positions of the actual feature points.
[0038] In some optional embodiments, such as Figure 4 , Figure 4 yes Figure 1 A flowchart of a specific implementation of S200, S200 includes: S210, projecting the actual feature points onto the corresponding chart image according to the internal and external parameters of the camera module; S220, obtaining the predicted position of the actual feature points in the image coordinate system, and calculating the predicted coordinates of the predicted position.
[0039] In this embodiment, a reprojection error algorithm is used to project the actual feature points of the chart in the three-dimensional world onto the chart image plane through the camera module to be calibrated, and the deviation between the predicted feature points and the corresponding actual feature point positions (coordinates) detected in the actual chart is calculated.
[0040] S300, obtaining the actual position of the actual feature point, calculating the first mean of the projection error between the predicted position and the actual position, and calculating the second mean of the projection error of all chart images, and calculating the difference between the first mean and the second mean of the chart images.
[0041] In some optional embodiments, such as Figure 5 , Figure 5 yes Figure 1 A flowchart of a specific implementation of S300; S300 includes: S310, using an image processing algorithm to calculate and obtain the actual coordinates of the actual feature points; S320, using a Euclidean distance algorithm to calculate the projection error between the predicted coordinates and the actual coordinates; S330, calculating the projection errors of all actual feature points on the chart, and calculating and obtaining the first mean.
[0042] It is worth noting that, in other implementations, the actual coordinates may also be calculated based on the actual size and three-dimensional spatial position of the actual chart.
[0043] In this embodiment, when calculating the reprojection error, it can be assumed that the coordinates of the predicted feature point are (x0, y0), and the actual coordinates of the corresponding actual feature point are (x1, y1), then the Euclidean distance r between the two is: r=sqrt((x1-x0)*(x1-x0)+(y1-y0)*(y1-y0))(1), Each predicted feature point corresponds to an r. For example, if there are 88 points in a whole image, the first mean is (r1+r2+r3…r 88 ) / 88, the first mean is taken as the projection error err of this image. The smaller the err value is, the more accurate the calibrated data is.
[0044] In some optional embodiments, such as Figure 6 , Figure 6 yes Figure 5A flowchart of a specific implementation of S330; after S330, S300 also includes: S340, obtaining a first mean value of the projection errors of all actual feature points corresponding to all chart images; S350, calculating a second mean value of the projection errors of all chart images based on the first mean value.
[0045] Get the first mean (err) of each image, calculate the projection error of all images, that is, the second mean (aveerr), and then calculate the difference between the first mean (err) and the second mean (aveerr) of each image. If it is greater than a certain set value, the image is automatically deleted. Finally, recalibrate the remaining corner points.
[0046] S400: Compare the difference with a preset threshold value, and if the difference is greater than the preset threshold value, delete the chart image.
[0047] If the difference of some chart images is slightly larger, but does not reach the set value, it will not have any impact. Generally, the projection error of images near the four corners will be larger. If the projection error of some images is relatively large, it will directly affect the accuracy of this calibration data, so these images need to be deleted.
[0048] When deleting, make sure that this area is covered by feature points of other chart images: if this area is not covered by other images, the calibration result may be inaccurate. For example, if only 10 chart images are taken, two in the middle of each of the four corners, and the two in the upper left corner are deleted, the feature point position will not cover the upper left corner area.
[0049] If the chart image with a relatively large projection error is directly deleted, it will lead to the accidental deletion of the image. Because some abnormal chart images with excessively large projection errors produce inaccurate calibration, the projection error value of the normal chart image will be raised (image with normal error). In this embodiment, the actual feature points of the calibrated chart are projected onto the chart image, and the first mean of the projection error of each chart image is calculated, and then the second mean of the projection error of all chart images is obtained. The difference between the first mean and the second mean is compared with a preset threshold, so as to screen out the chart images with large errors and delete them, so as to avoid the deletion of images with normal errors, and solve the problem of the adverse effect of abnormal images on the calibration of the camera module, so that when the camera module can be calibrated again, the calibration precision and accuracy of the camera module can be improved.
[0050] S500, repeat S200-400 until all chart images are determined.
[0051] In some optional embodiments, such as Figure 7 , Figure 7 yes Figure 1 A flowchart of a specific implementation method after S500 in FIG. 1 ; the method for automatically deleting calibrated abnormal images also includes: S600, determine whether the area of the deleted chart image is covered by other chart images; S700, if no other chart image covers it, determine that the calibration fails. When the area of the deleted image is not covered by feature points of other images, the structure is determined to be NG, and the loop ends.
[0052] In order to solve the above technical problems, the embodiment of the present application also provides a device 800 for automatically deleting abnormal images for calibration, such as Figure 8 , Figure 8 1 is a schematic diagram of a structure of an embodiment of an apparatus 800 for automatically deleting abnormal images for calibration according to the present application; the method for automatically deleting abnormal images for calibration according to the first aspect is adopted, comprising: The calibration module 801 is used to obtain multiple chart images using the camera module to be calibrated, and calibrate the camera module using the multiple chart images to obtain the internal and external parameters of the camera module; The acquisition module 802 is used to project the actual feature points onto the chart image according to all the actual feature points of the acquired chart and using the internal and external parameters of the camera module to acquire the predicted positions of the actual feature points; The calculation module 803 is used to calculate the first mean of the projection error between the predicted position and the actual position according to the actual position of the actual feature point, calculate the second mean of the projection error of all charts, and calculate the difference between the first mean and the second mean of the chart image. The deletion module 804 is used to compare the difference with a preset threshold, and if the difference is greater than the preset threshold, delete the chart image.
[0053] In order to solve the above technical problems, an embodiment of the present application also provides a computer device, including a memory and a processor, wherein the memory stores computer-readable instructions, and when the processor executes the computer-readable instructions, the steps of the method for automatically deleting calibrated abnormal images are implemented as described above.
[0054] In order to solve the above technical problems, an embodiment of the present application also provides a computer-readable storage medium, on which computer-readable instructions are stored. When the computer-readable instructions are executed by a processor, the steps of the above-mentioned method for automatically deleting calibrated abnormal images are implemented.
[0055] In order to solve the above problems, an embodiment of the present application further proposes a computer device, which adopts the following technical solution: it includes a processor, a network module and a memory, and the processor and the memory are interconnected through the network module.
[0056] The computer device may be a computer, server, workstation or other device, or a mobile phone, tablet, vehicle-mounted mobile terminal or other device with program execution capability. The internal structure diagram of the computer device may be as follows: Fig. 9 As shown, Fig. 9 It is a structural diagram of an embodiment of a computer device according to the present application. The computer device includes a processor, a memory and a network module. Among them, the processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile and / or volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, instructions or codes. The internal memory provides an environment for the operation of the operating system and instructions or codes in the non-volatile storage medium. When the instructions or codes are executed by the processor, the functions or steps of a method for fine-tuning the above-mentioned pre-trained model are implemented. The network module of the computer device may include a network interface and / or a wireless network module, and the computer device may communicate with other devices or service platforms through the network module. In addition, the computer device may also include a display screen and an input device, etc.
[0057] The memory is used to store a computer program, the computer program includes program instructions, the processor is configured to call the program instructions, and the processor implements the steps of the above-mentioned method for efficient fine-tuning of adaptive parameters when executing the instructions or codes.
[0058] In order to solve the above problems, the embodiment of the present application further proposes a computer-readable storage medium, which adopts the following technical solution: The computer-readable storage medium stores a computer program, which includes program instructions. When the program instructions are executed by a processor, the processor executes a method for automatically deleting abnormal images for calibration.
[0059] The present application also provides a computer-readable storage medium storing a computer program, wherein the computer program includes program instructions, and when the program instructions are executed by a processor, Figures 1 to 7 The method for automatically deleting abnormal images for calibration provided in each step can be specifically referred to the implementation method provided in the above steps, which will not be repeated here.
[0060] The computer-readable storage medium may be an abnormal image automatic deletion device 800 for calibration provided in any of the aforementioned embodiments or an internal storage unit of the terminal device, such as a hard disk or memory of a computer device. The computer-readable storage medium may also be an external storage device of the computer device, such as a plug-in hard disk, a smart media card (SMC), a secure digital (SD) card, a flash card, etc., provided on the computer device.
[0061] Furthermore, the computer-readable storage medium may also include both an internal storage unit of the computer device and an external storage device. The computer-readable storage medium is used to store the computer program and other programs and data required by the computer device. The computer-readable storage medium may also be used to temporarily store data that has been output or is to be output.
[0062] However, it should be understood that it is not required to implement all the components shown, and more or fewer components may be implemented instead. Among them, those skilled in the art can understand that the computer device here is a device that can automatically perform numerical calculations and / or information processing according to pre-set or stored instructions, and its hardware includes but is not limited to microprocessors, application specific integrated circuits (ASIC), field-programmable gate arrays (FPGA), digital signal processors (DSP), embedded devices, etc.
[0063] The computer device may be a computing device such as a desktop computer, a notebook, a PDA, a cloud server, etc. The computer device may interact with a user through a keyboard, a mouse, a remote controller, a touch pad, or a voice control device.
[0064] Compared with the prior art, the embodiments of the present application mainly have the following technical effects: by projecting the actual feature points of the calibrated chart onto the chart image, and calculating the first mean of the projection error of each chart image, and then obtaining the second mean of the projection error of all chart images, the difference between the first mean and the second mean is compared with a preset threshold, so as to screen out the chart images with larger errors and delete them, avoiding the deletion of pictures with normal errors, and solving the problem of adverse effects of abnormal pictures on the calibration of the camera module, so that when calibrating again, the calibration precision and accuracy of the camera module are improved.
[0065] The above are only preferred embodiments of the present invention and are not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the protection scope of the present invention.
Claims
1. A method for automatically deleting abnormal images for calibration, characterized in that it includes: Step 100: using the camera module to be calibrated to obtain a plurality of chart images, and using the plurality of chart images to calibrate the camera module to obtain the internal and external parameters of the camera module; Step 200: Obtain all actual feature points of the chart, and project the actual feature points onto the chart image using the internal and external parameters of the camera module to obtain predicted positions of the actual feature points; Step 300: obtaining the actual position of the actual feature point, calculating a first mean value of the projection error between the predicted position and the actual position, calculating a second mean value of the projection error of all chart images, and calculating a difference between the first mean value and the second mean value of the chart images; Step 400: compare the difference with a preset threshold, and if the difference is greater than the preset threshold, delete the chart image; Step 500: Repeat steps 200-400 until all chart images are determined.
2. The method for automatically deleting abnormal images for calibration according to claim 1, characterized in that: The method for automatically deleting abnormal images for calibration also includes: Step 600: determine whether the area of the deleted chart image is covered by other chart images; Step 700: If there is no other chart image coverage, it is determined that the calibration has failed.
3. The method for automatically deleting abnormal images for calibration according to claim 2, characterized in that: The step 100 comprises: Step 110, making an actual chart, and fixing the chart on a calibration board; Step 120: fix the camera module, adjust the focal length and exposure parameters of the camera module to obtain a clear plane image, shoot the chart from different angles, and obtain the multiple chart images.
4. The method for automatically deleting abnormal images for calibration according to claim 3, characterized in that: After step 120, step 100 further includes: Step 130: Obtain predicted feature points of the chart image using a corner point detection method; Step 140: acquiring actual feature points corresponding to the predicted feature points according to the actual size and feature point layout of the chart; Step 150: According to the actual feature points and the predicted feature points, and using the Zhang Zhengyou camera calibration algorithm, obtain the intrinsic and extrinsic parameters of the camera module.
5. The method for automatically deleting abnormal images for calibration according to claim 2, characterized in that: The step 200 comprises: Step 210: projecting the actual feature points onto a corresponding chart image according to the internal and external parameters of the camera module; Step 220: Obtain the predicted position of the actual feature point in the image coordinate system, and calculate the predicted coordinates of the predicted position.
6. The method for automatically deleting abnormal images for calibration according to claim 2, characterized in that: The step 300 includes: Step 310: Calculate and obtain the actual coordinates of the actual feature points using an image processing algorithm; Step 320: Calculate the projection error between the predicted coordinates and the actual coordinates using a Euclidean distance algorithm; Step 330: Calculate the projection errors of all actual feature points on the chart, and obtain the first mean value.
7. The method for automatically deleting abnormal images for calibration according to claim 6, characterized in that: After step 330, step 300 further includes: Step 340: Obtain a first mean value of projection errors of all actual feature points corresponding to all chart images; Step 350: Obtain a second mean value of the projection errors of all chart images according to the first mean value calculation.
8. A device for automatically deleting abnormal images for calibration, using the method for automatically deleting abnormal images for calibration according to any one of claims 1 to 7, characterized in that: include: A calibration module, used to obtain a plurality of chart images using the camera module to be calibrated, and calibrate the camera module using the plurality of chart images to obtain the internal and external parameters of the camera module; An acquisition module, used to project the actual feature points onto the chart image based on all the actual feature points of the acquired chart and using the internal and external parameters of the camera module to acquire the predicted positions of the actual feature points; a calculation module, configured to calculate a first mean value of the projection error between the predicted position and the actual position according to the actual position of the actual feature point, calculate a second mean value of the projection error of all chart images, and calculate a difference between the first mean value and the second mean value of the chart image, The deletion module is used to compare the difference with a preset threshold value, and if the difference is greater than the preset threshold value, delete the chart image.
9. A computer device, characterized in that: It includes a memory and a processor, wherein the memory stores computer-readable instructions, and when the processor executes the computer-readable instructions, the steps of the method for automatically deleting calibrated abnormal images as described in any one of claims 1 to 7 are implemented.
10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores computer-readable instructions, and when the computer-readable instructions are executed by a processor, the steps of the method for automatically deleting calibrated abnormal images as described in any one of claims 1 to 7 are implemented.