Method for calibrating binocular camera calibration parameter, apparatus, computer device, and medium

Abstract

A method for calibrating a binocular camera calibration parameter, an apparatus, a computer device, and a medium. The method comprises: acquiring a binocular disparity map and / or a binocular depth map; performing epipolar alignment detection on the binocular disparity map and / or the binocular depth map to obtain an epipolar alignment error; and on the basis of the epipolar alignment error and a first preset threshold, determining a target calibration parameter to be calibrated of a binocular camera. In the solution, in the process of calibrating the binocular camera calibration parameter, the epipolar alignment error is obtained by performing epipolar alignment detection on the binocular disparity map and / or the binocular depth map, and the target calibration parameter of the binocular camera is determined on the basis of the epipolar alignment error, thereby not only simplifying the calculation of the epipolar alignment error but also improving the precision of the epipolar alignment error, thus enhancing the accuracy of calibration of the calibration parameter.

Description

Binocular camera calibration method and device, computer device and medium

[0001] Cross-reference to Related Applications

[0002] The present application claims priority to the Chinese patent application No. 202411171533.6, filed on August 23, 2024, entitled "Binocular camera calibration method and device, computer device and medium", the content of which is incorporated herein by reference in its entirety.

[0003] TECHNICAL FIELD

[0004] The present application relates to the technical field of camera parameter monitoring, in particular to a binocular camera calibration method and device, computer device and medium. BACKGROUND

[0005] A binocular camera generally includes a left camera and a right camera, and the two cameras are respectively used to simulate the left eye and the right eye of a person, so as to realize positioning or binocular depth calculation. During the use of the binocular camera by a user, the rigidity of the binocular camera structure will change due to changes in the use environment or collisions, so that the initial calibration parameters of the binocular camera when it is shipped are no longer applicable, thereby causing the binocular depth calculation to fail. Therefore, it is necessary to continuously monitor the calibration parameters of the binocular camera to calibrate the calibration parameters of the binocular camera.

[0006] At present, for the calibration of the calibration parameters of the binocular camera, the binocular camera needs to be set on an XR device, and the calibration parameters of the binocular camera are transmitted to the XR device to enable the XR device to perform calibration of the calibration parameters of the binocular camera.

[0007] However, the above method needs to increase additional transmission bandwidth and computing power, resulting in high calibration parameter calibration cost. SUMMARY

[0008] The main purpose of the present application is to provide a binocular camera calibration method and device, computer device and medium, which do not need to increase additional transmission bandwidth and computing power, thereby reducing the cost of calibration parameter calibration.

[0009] In order to achieve the above purpose, in a first aspect, the present application provides a binocular camera calibration method, comprising:

[0010] obtaining a binocular disparity map and / or a binocular depth map, wherein the binocular disparity map and / or the binocular depth map are determined based on initial calibration parameters of a binocular camera to be calibrated and left and right eye images obtained by the binocular camera to be calibrated;

[0011] The epipolar alignment detection is performed on the binocular disparity map and / or the binocular depth map to obtain an epipolar alignment error, including at least one of the following:

[0012] Based on a pixel point in a first disparity map in the binocular disparity map, a corresponding pixel point of a second disparity map in the binocular disparity map is calculated, and the epipolar alignment error is obtained according to the pixel point in the first disparity map and the corresponding pixel point of the second disparity map;

[0013] Based on the confidence of the binocular depth map, the epipolar alignment error is obtained

[0014] Based on the epipolar alignment error and a first preset threshold, a target calibration parameter of the binocular camera to be calibrated is determined.

[0015] In an embodiment, the binocular disparity map includes a first disparity map and a second disparity map.

[0016] The epipolar alignment detection is performed on the binocular disparity map to obtain an epipolar alignment error, including:

[0017] The disparity and coordinates of all pixel points in the first disparity map are obtained;

[0018] Based on the disparity and coordinates of all pixel points, the disparity of a first pixel point and the disparity of a second pixel point in the second disparity map are calculated, wherein the first pixel point is a pixel point in the second disparity map corresponding to each pixel point in all pixel points, and the second pixel point is a pixel point adjacent to the first pixel point in the second disparity map;

[0019] The epipolar alignment error is obtained according to the disparity of the first pixel point and the disparity of the second pixel point.

[0020] In an embodiment, based on the disparity and coordinates of all pixel points, the disparity of a first pixel point and the disparity of a second pixel point in the second disparity map are calculated, including:

[0021] Based on the disparity and coordinates of all pixel points, the coordinates of the first pixel point are calculated;

[0022] The disparity of the first pixel point is obtained from the second disparity map according to the coordinates of the first pixel point;

[0023] Based on the coordinates of the first pixel point, the coordinates of the second pixel point are calculated;

[0024] The disparity of the second pixel point is obtained from the second disparity map according to the coordinates of the second pixel point.

[0025] In an embodiment, the epipolar alignment error is obtained according to the disparity of the first pixel point and the disparity of the second pixel point, including:

[0026] determine the first target pixel point corresponding to each pixel point in all the pixel points based on the disparity of the first pixel point and the disparity of the second pixel point in the second disparity map;

[0027] count the first target pixel points corresponding to each pixel point to obtain the number of the first target pixel points;

[0028] calculate the ratio of the number of the first target pixel points to the number of all the pixel points in the first disparity map, and take the ratio as the epipolar alignment error.

[0029] In an embodiment, the determining the first target pixel point corresponding to each pixel point in all the pixel points based on the disparity of the first pixel point and the disparity of the second pixel point in the second disparity map comprises:

[0030] respectively calculate the difference between the disparity of each pixel point and the disparity of the first pixel point and the disparity of the second pixel point to obtain a first difference value and a second difference value;

[0031] respectively calculate the absolute value of the first difference value and the absolute value of the second difference value to obtain a first absolute value and a second absolute value;

[0032] if there is an absolute value less than the second preset threshold value in the first absolute value and the second absolute value, take the pixel point corresponding to the absolute value as the first target pixel point corresponding to each pixel point.

[0033] In an embodiment, the binocular depth map comprises a first depth map and a second depth map.

[0034] perform epipolar alignment detection on the binocular depth map to obtain an epipolar alignment error, comprising:

[0035] calculate a first confidence ratio corresponding to the first depth map and a second confidence ratio corresponding to the second depth map;

[0036] take the smallest confidence ratio in the first confidence ratio and the second confidence ratio as the epipolar alignment error.

[0037] In an embodiment, the calculating the first confidence ratio corresponding to the first depth map comprises:

[0038] traverse all the pixel points in the first depth map to obtain the pixel points in the first depth map whose confidence is greater than a third preset threshold value to obtain second target pixel points;

[0039] count the second target pixel points to obtain the number of the second target pixel points;

[0040] obtain the resolution of the first depth map;

[0041] A ratio of the number of the second target pixel points to a resolution of the first depth map is calculated to obtain a first confidence ratio corresponding to the first depth map.

[0042] In an embodiment, the binocular disparity map is obtained by:

[0043] The left and right images are obtained by the binocular camera to be calibrated, and initial calibration parameters of the binocular camera to be calibrated are obtained, wherein the initial calibration parameters include intrinsic parameters, extrinsic parameters and de-distortion parameters.

[0044] The left and right images are preprocessed to obtain left and right distortion images.

[0045] The left and right distortion images are de-distorted by using the de-distortion parameters to obtain left and right de-distortion images.

[0046] The left and right de-distortion images are epipolar rectified by using the intrinsic parameters and the extrinsic parameters to obtain left and right epipolar rectification images.

[0047] The left and right epipolar rectification images are matched by using a stereo matching algorithm to obtain the binocular disparity map.

[0048] In an embodiment, the binocular depth map is obtained by:

[0049] The binocular disparity map is processed by using the intrinsic parameters and the extrinsic parameters to obtain the binocular depth map.

[0050] In an embodiment, the target calibration parameters of the binocular camera to be calibrated are determined based on the epipolar alignment error and a first preset threshold, including:

[0051] If the epipolar alignment error is less than or equal to the first preset threshold, the initial calibration parameters of the binocular camera to be calibrated are not invalid, and the initial calibration parameters are taken as the target calibration parameters.

[0052] If the epipolar alignment error is greater than the first preset threshold, the initial calibration parameters of the binocular camera to be calibrated are invalid, and the initial calibration parameters are updated to obtain the target calibration parameters.

[0053] In an embodiment, if the epipolar alignment error is greater than the first preset threshold, the initial calibration parameters of the binocular camera to be calibrated are invalid, and the initial calibration parameters are updated to obtain the target calibration parameters, including:

[0054] If the epipolar alignment error is greater than the first preset threshold, the initial calibration parameters of the binocular camera to be calibrated are invalid, and the initial calibration parameters are updated.

[0055] If a difference between the updated initial calibration parameters and the initial calibration parameters is less than a third preset threshold, the initial calibration parameters are taken as the target calibration parameters.

[0056] In a second aspect, the embodiments of the present application provide a binocular camera calibration device, comprising:

[0057] an acquisition module configured to acquire a binocular disparity map and / or a binocular depth map, wherein the binocular disparity map and / or the binocular depth map are determined based on initial calibration parameters of a binocular camera to be calibrated and left and right eye images acquired by the binocular camera to be calibrated;

[0058] a detection module configured to perform epipolar alignment detection on the binocular disparity map and / or the binocular depth map to obtain an epipolar alignment error, the epipolar alignment error comprising at least one of:

[0059] calculating, based on a pixel point of a first disparity map in the binocular disparity map, a pixel point corresponding to a second disparity map in the binocular disparity map, and obtaining the epipolar alignment error according to the pixel point of the first disparity map and the pixel point corresponding to the second disparity map;

[0060] obtaining the epipolar alignment error based on a confidence of the binocular depth map;

[0061] a parameter calibration module configured to determine target calibration parameters of the binocular camera to be calibrated based on the epipolar alignment error and a first preset threshold.

[0062] In a third aspect, the embodiments of the present application provide a computer device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, and the processor implements the steps of any of the above methods when executing the computer program.

[0063] In a fourth aspect, the embodiments of the present application provide a computer readable storage medium, which stores a computer program, and the computer program is executable on a processor to implement the steps of any of the above methods.

[0064] In a fifth aspect, the embodiments of the present application provide a computer program product, comprising a computer program, and the computer program is executable on a processor to implement the steps of any of the above methods.

[0065] The embodiment of the application provides a binocular camera calibration method, a binocular camera calibration device, computer equipment and a medium. In the calibration process of the binocular camera calibration parameter, the epipolar alignment error is obtained by performing epipolar alignment detection on the binocular disparity map and / or the binocular depth map, and the target calibration parameter of the binocular camera is determined based on the epipolar alignment error. The calculation of the epipolar alignment error is simplified, the accuracy of the epipolar alignment error is improved, and the accuracy of the calibration parameter calibration is improved. In addition, the calibration process of the binocular camera calibration parameter can be directly performed in the binocular camera, the additional transmission bandwidth and calculation power are avoided, the cost of the calibration parameter calibration is reduced, and the calculation results of the binocular depth calculation, i.e., the binocular disparity map and / or the binocular depth map, are reused to calibrate whether the epipolar lines are aligned in real time, so that whether the calibration parameter of the camera is invalid is determined, and the calibration parameter of the invalid camera is calibrated. The result can be obtained in real time online, the calculation amount of the invalid detection link is saved, the detection result is more in line with the requirement of the binocular depth calculation, and the self-calibration calculation result can also be fed back from the binocular depth calculation to realize the feedback loop. BRIEF DESCRIPTION OF DRAWINGS

[0066] The accompanying drawings, which form a part of this application, are included to provide a further understanding of the application and are incorporated in and constitute a part of this application. The embodiments illustrated in the drawings provide explanations of the application and do not constitute an undue limitation on the application. In the drawings:

[0067] FIG. 1 is a flowchart of a binocular camera calibration method according to an embodiment of the application;

[0068] FIG. 2 is a flowchart of another binocular camera calibration method according to an embodiment of the application;

[0069] FIG. 3 is a flowchart of an epipolar alignment error calculation method according to an embodiment of the application;

[0070] FIG. 4 is a structural diagram of a binocular camera calibration device according to an embodiment of the application;

[0071] FIG. 5 is a schematic diagram of computer equipment according to an embodiment of the application. Embodiments of the application

[0072] To make the purposes, technical solutions and advantages of the embodiments of the application clearer, the technical solutions in the embodiments of the application will be described clearly and completely below with reference to the drawings in the embodiments of the application. Obviously, the described embodiments are only part of the embodiments of the application, rather than all the embodiments of the application. Based on the embodiments in the application, all other embodiments obtained by those skilled in the art without creative labor fall within the scope of protection of the application.

[0073] The terms "first", "second", "third", "fourth" and the like in the description and in the claims of the present application, if any, are used for distinguishing between similar objects and not necessarily for describing a specific sequential or chronological order. It is to be understood that the use of these terms herein is to be construed as interchangeable in order to describe the embodiments of the present application.

[0074] In the embodiments of the present application, the term "module" or "unit" refers to a computer program or a part of a computer program with a predetermined function, and works together with other related parts to achieve a predetermined target, and can be implemented entirely or partially by using software, hardware (such as a processing circuit or a memory) or a combination thereof. Similarly, one processor (or multiple processors or memories) can be used to implement one or more modules or units. In addition, each module or unit can be a part of an integral module or unit that includes the functions of the module or unit.

[0075] It should be understood that, in various embodiments of the present application, the magnitude of the serial number of each process does not mean the order of execution, and the execution order of each process should be determined according to its function and inherent logic, and should not constitute any limitation on the implementation process of the embodiments of the present application.

[0076] It should be understood that, in the present application, "include" and "have" and any variations thereof are intended to cover non-exclusive inclusion, for example, a process, method, system, product or device that includes a series of steps or units does not have to be limited to those clearly listed steps or units, but can include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0077] It should be understood that, in the present application, "multiple" means two or more. "And / or" is only a description of the relationship between the associated objects, which means that there can be three relationships, for example, A and / or B can mean: A exists alone, A and B exist together, and B exists alone. The character " / " generally represents that the associated objects before and after are in an "or" relationship. "Including A, B and C", "including A, B, C" means that A, B and C are all included, "including A, B or C" means that one of A, B and C is included, "including A, B and / or C" means that any one or any two or three of A, B and C is included.

[0078] It should be understood that, in the present application, “B corresponding to A”, “B corresponding to A”, “A corresponding to B” or “B corresponding to A” means that B is associated with A, and B can be determined according to A. Determining B according to A does not mean that B is determined only according to A, but B can also be determined according to A and / or other information. The matching of A and B means that the similarity of A and B is greater than or equal to a preset threshold.

[0079] Depending on the context, “if” as used herein can be interpreted as “when” or “upon” or “in response to determining” or “in response to detecting”.

[0080] The data involved in the present application can be data authorized by the tested person or fully authorized by all parties, and the collection, transmission, use, etc. of the data meet the requirements of relevant laws, regulations and standards of relevant countries and regions. The embodiments / examples of the present application can be combined with each other.

[0081] The technical solutions of the present application will be described in detail below with specific examples. The following specific examples can be combined with each other, and the same or similar concepts or processes can not be described in detail in some examples.

[0082] Please refer to FIG. 1, which is a flowchart of a binocular camera calibration method according to an embodiment of the present application. As shown in FIG. 1, the method comprises the following steps:

[0083] Step S101: Obtain a binocular disparity map and / or a binocular depth map.

[0084] The binocular disparity map comprises a first disparity map and a second disparity map, and the binocular depth map comprises a first depth map and a second depth map.

[0085] For obtaining the binocular disparity map, the left and right eye images are obtained by the binocular camera to be calibrated, and the initial calibration parameters of the binocular camera to be calibrated are obtained. The left and right eye images are preprocessed to obtain left and right eye distortion images. Then, the distortion correction parameters are used to correct the distortion of the left and right eye distortion images to obtain left and right eye distortion-corrected images. Then, the intrinsic and extrinsic parameters are used to perform epipolar correction on the left and right eye distortion-corrected images to obtain left and right eye epipolar correction images. Finally, a stereo matching algorithm is used to match the left and right eye epipolar correction images to obtain the binocular disparity map.

[0086] The binocular camera is calibrated before leaving the factory, and the binocular camera has initial calibration parameters after calibration. The initial calibration parameters include intrinsic parameters, extrinsic parameters and de-distortion parameters. The intrinsic parameters are determined by the binocular camera itself, which describes the internal structure parameters of the binocular camera, such as focal length, image center position, etc., and the intrinsic parameters are not affected by the external environment. The extrinsic parameters include rotation and translation between the two camera coordinate systems in the binocular camera. The de-distortion parameters include the radial distortion coefficient and the tangential distortion coefficient of the binocular camera, which are correction coefficients for transforming the actual imaging point position to the ideal imaging point position.

[0087] After obtaining the initial calibration parameters of the binocular camera to be calibrated, the left and right eye images can be obtained by the two cameras of the binocular camera to be calibrated for a certain scene or target, wherein the left and right eye images include left eye images and right eye images, the left eye images are images obtained by the left camera of the binocular camera to be calibrated for a certain scene or target, and the right eye images are images obtained by the right camera of the binocular camera to be calibrated for a certain scene or target.

[0088] After obtaining the initial calibration parameters of the binocular camera to be calibrated and the left and right eye images, the binocular disparity map and the binocular depth map corresponding to the left target image can be obtained based on the initial calibration parameters of the binocular camera to be calibrated and the left and right eye images.

[0089] As shown in FIG. 2, after obtaining the left eye image and the right eye image, the left eye image and the right eye image are preprocessed respectively to obtain the left eye distortion image and the right eye distortion image, wherein the preprocessing methods include but are not limited to denoising, white balance and other conventional image processing methods, and the preprocessing can be performed in an image signal processor (Image Signal Processing, ISP).

[0090] After obtaining the left eye distortion image and the right eye distortion image, the left eye distortion image and the right eye distortion image are corrected by using the de-distortion parameters respectively to obtain the left eye de-distortion image and the right eye de-distortion image. The distortion parameters can also be obtained by calculating the feature points on the calibration board, detecting the image edge estimation, and matching the feature points.

[0091] Then, the left eye de-distortion image and the right eye de-distortion image are rectified by using the intrinsic parameters and the extrinsic parameters respectively to obtain the left eye rectified image and the right eye rectified image. The rectification can align the epipolar lines of the left eye de-distortion image and the right eye de-distortion image, and make the corresponding pixel points in the left eye de-distortion image and the right eye de-distortion image on the same line, which facilitates the search for corresponding pixel points on the same epipolar line when performing stereo matching on the left eye rectified image and the right eye rectified image, and improves the search and matching efficiency of the corresponding pixel points.

[0092] Finally, the left epipolar rectification image and the right epipolar rectification image are matched by using a stereo matching algorithm to obtain a first disparity map and a second disparity map. The stereo matching algorithm includes but is not limited to a feature-based matching algorithm, a block-based matching algorithm, a deep learning-based matching algorithm, and the like.

[0093] It should be noted that the process of obtaining the first disparity map and the second disparity map by using the stereo matching algorithm, and the process of obtaining the epipolar alignment error by performing epipolar alignment detection on the binocular disparity map in step S102 can be performed in a binocular depth algorithm module. The binocular depth algorithm can include various algorithm schemes, such as SGM, BM, SGBM, Raft stereo, Hitnet, and various binocular algorithms that can calculate binocular disparity, or binocular algorithms that can calculate confidence.

[0094] When the binocular depth algorithm is SGM, the left epipolar rectification image and the right epipolar rectification image are processed by the SGM algorithm module, and the SGM algorithm module directly outputs the epipolar alignment error.

[0095] After obtaining the first disparity map and the second disparity map, the first disparity map and the second disparity map can also be processed by using intrinsic parameters and extrinsic parameters to obtain a first depth map and a second depth map. The algorithm of the first depth map and the second depth map includes but is not limited to a convolutional neural network method, a deep learning and stereo vision fusion method, a baseline triangulation method, and the like. The binocular disparity map is determined by a binocular matching algorithm, including SGM, SGBM, Raft Stereo, Hitnet, and the like.

[0096] The embodiments of the present application perform denoising, distortion correction, epipolar rectification, and the like on the left and right images based on the initial calibration parameters of the binocular camera, so that the precision of the obtained binocular disparity map and binocular depth map is higher, to improve the calibration effect of the calibration parameters.

[0097] Step S102: Perform epipolar alignment detection on the binocular disparity map and / or the binocular depth map to obtain an epipolar alignment error.

[0098] For the determination of the epipolar alignment error, the present application is implemented in two ways. The first way is to perform epipolar alignment detection on the binocular disparity map, and the other way is to perform epipolar alignment detection on the binocular depth map.

[0099] The first mode: the epipolar alignment error is obtained by detecting the epipolar alignment of the binocular disparity map. All pixel points in the first disparity map are traversed. For each pixel point in all pixel points in the first disparity map, the disparity and the coordinate of each pixel point are obtained. Then, the disparity of the first pixel point and the disparity of the second pixel point in the second disparity map are calculated based on the disparity and the coordinate of each pixel point. The first target pixel point corresponding to each pixel point is counted, the number of the first target pixel point is obtained, and finally, the ratio of the number of the first target pixel point to the number of all pixel points in the first disparity map is calculated, and the ratio is taken as the epipolar alignment error.

[0100] The first pixel point is a pixel point in the second disparity map corresponding to each pixel point, and the second pixel point is a pixel point adjacent to the first pixel point in the second disparity map. The first target pixel point corresponding to each pixel point is determined based on the disparity of the first pixel point and the disparity of the second pixel point in the second disparity map.

[0101] The disparity of the first pixel point and the disparity of the second pixel point in the second disparity map are calculated based on the disparity and the coordinate of each pixel point. The coordinate of the first pixel point is calculated based on the disparity and the coordinate of each pixel point. The disparity of the first pixel point is obtained from the second disparity map according to the coordinate of the first pixel point. The coordinate of the second pixel point is calculated based on the coordinate of the first pixel point. The disparity of the second pixel point is obtained from the second disparity map according to the coordinate of the second pixel point.

[0102] The first target pixel point corresponding to each pixel point is determined based on the disparity of the first pixel point and the disparity of the second pixel point in the second disparity map. The difference between the disparity of each pixel point and the disparity of the first pixel point and the disparity of the second pixel point is calculated to obtain a first difference value and a second difference value. The absolute value of the first difference value and the absolute value of the second difference value are calculated to obtain a first absolute value and a second absolute value. If there is an absolute value less than a second preset threshold value in the first absolute value and the second absolute value, the pixel point corresponding to the absolute value is taken as the first target pixel point corresponding to each pixel point. The second preset threshold value can be set according to specific conditions, which is not limited here.

[0103] As shown in FIG. 3, first, each pixel point (i.e. each pixel) in all pixel points in the first disparity map is traversed, and the disparity d_L and the coordinate (u, v) of each pixel point are obtained by querying the first disparity map. After obtaining the disparity d_L and the coordinate (u, v) of a certain pixel point in the first disparity map, the coordinate (u-d_L, v) of the pixel point (i.e. the first pixel point) corresponding to the certain pixel point on the second disparity map can be calculated based on the disparity d_L of the certain pixel point, and the coordinates of the pixel points adjacent to the first pixel point on the second disparity map, such as the coordinate (u-d_L-1, v) of the pixel point adjacent to the left side of the first pixel point and the coordinate (u-d_L+1, v) of the pixel point adjacent to the right side of the first pixel point.

[0104] After obtaining the coordinate (u-d_L, v) of the first pixel point, the coordinate (u-d_L-1, v) of the pixel point adjacent to the left side of the first pixel point, and the coordinate (u-d_L+1, v) of the pixel point adjacent to the right side of the first pixel point, the corresponding disparities can be queried from the second disparity map based on the above coordinates, such as the disparity d_LR2 of the first pixel point, the disparity d_LR1 of the pixel point adjacent to the left side of the first pixel point, and the disparity d_LR3 of the pixel point adjacent to the right side of the first pixel point.

[0105] It should be noted that the pixel points adjacent to the first pixel point can be the left and right pixel points of the first pixel point, and can also include the upper and lower pixel points. Here, no specific limitation is made, and adjustment can be made according to the specific situation.

[0106] Then, the absolute values of the differences between the disparity d_L of each pixel point in the first disparity map and the disparity d_LR2 of the first pixel point in the second disparity map, the absolute values of the differences between the disparity d_L of each pixel point in the first disparity map and the disparity d_LR1 of the pixel point adjacent to the left side of the first pixel point, and the absolute values of the differences between the disparity d_L of each pixel point in the first disparity map and the disparity d_LR3 of the pixel point adjacent to the right side of the first pixel point are calculated respectively.

[0107] The second preset threshold is 1, and it is necessary to compare whether there is an absolute value less than 1 in all the calculated absolute values. If there is, the pixel point corresponding to the absolute value can be taken as the first target pixel point corresponding to each pixel point. For example, the absolute values of the differences between the disparity d_L of each pixel point in the first disparity map and the disparity d_LR2 of the first pixel point in the second disparity map, and the absolute values of the differences between the disparity d_L of each pixel point in the first disparity map and the disparity d_LR1 of the pixel point adjacent to the left side of the first pixel point are less than 1, and then the first pixel point in the second disparity map and the pixel point adjacent to the left side of the first pixel point can be taken as the first target pixel point.

[0108] It can be seen that the first target pixel point corresponding to the pixel point with a disparity of d_L in the first disparity map is two, that is, the first pixel point in the second disparity map and the pixel point adjacent to the left of the first pixel point. According to the above manner, each pixel point in the first disparity map is traversed in a loop, the first target pixel point corresponding to each pixel point is obtained, and then the first target pixel point corresponding to each pixel point is counted to obtain the number of first target pixel points GoodMatch_count.

[0109] When all the pixel points in the first disparity map are traversed, the epipolar alignment error GoodMatch_percentage can be calculated. Wherein, the epipolar alignment error GoodMatch_percentage = the number of first target pixel points GoodMatch_count / the number of all pixel points in the first disparity map Total_pixel_number. Since the number of all pixel points in the first disparity map is the same as the number of all pixel points in the second disparity map, Total_pixel_number here can also be the number of all pixel points in the second disparity map.

[0110] In addition, it needs to be explained that in the manner of obtaining the epipolar alignment error by detecting the epipolar alignment of the binocular disparity map, all the pixel points in the second disparity map can be traversed, for each pixel point in all the pixel points in the second disparity map, the disparity and coordinates of each pixel point are obtained, and then based on the disparity and coordinates of each pixel point, the disparity of the first pixel point in the first disparity map and the disparity of the second pixel point are calculated, the first target pixel point corresponding to each pixel point is counted to obtain the number of first target pixel points, and finally the ratio of the number of first target pixel points to the number of all pixel points in the first disparity map (or the second disparity map) is calculated, and the ratio is taken as the epipolar alignment error. The specific implementation in the above steps is similar to the specific implementation of calculating the epipolar alignment error by first traversing all the pixel points in the first disparity map, which will not be repeated here.

[0111] Compared with the prior art, the application can real-time calibrate whether the epipolar line is aligned by multiplexing the intermediate calculation result of binocular depth calculation-disparity map, and then judge whether the calibration parameters of the binocular camera are invalid, so as to calibrate the target calibration parameters of the invalid calibration parameters. Not only can the result be given in real time, but also the calculation amount of the additional invalid detection link is saved, and the detection result is more in line with the requirements of binocular depth calculation. The results of calibration parameter invalidity detection and online self-calibration calculation are returned to the invalidity detection in binocular depth calculation, realizing feedback closed loop.

[0112] The second mode is: performing epipolar alignment detection on the binocular depth map to obtain an epipolar alignment error, and the minimum confidence ratio between a first confidence ratio corresponding to the first depth map and a second confidence ratio corresponding to the second depth map is taken as the epipolar alignment error.

[0113] The first confidence ratio corresponding to the first depth map is calculated by: first, traversing all the pixel points in the first depth map to obtain second target pixel points with a confidence greater than a third preset threshold in all the pixel points in the first depth map; then, counting the second target pixel points to obtain a number of the second target pixel points; then, obtaining a resolution of the first depth map; and finally, calculating a ratio of the number of the second target pixel points to the resolution of the first depth map to obtain the first confidence ratio corresponding to the first depth map. The third preset threshold can be set according to specific conditions, which is not limited here.

[0114] The second confidence ratio corresponding to the second depth map is calculated by: first, traversing all the pixel points in the second depth map to obtain third target pixel points with a confidence greater than a fourth preset threshold in all the pixel points in the second depth map; then, counting the third target pixel points to obtain a number of the third target pixel points; then, obtaining a resolution of the second depth map; and finally, calculating a ratio of the number of the third target pixel points to the resolution of the second depth map to obtain the second confidence ratio corresponding to the second depth map. The fourth preset threshold can be set according to specific conditions, which is not limited here.

[0115] Specifically, taking the first depth map as an example, first, all the pixel points in the first depth map are traversed to obtain second target pixel points with a confidence greater than a third preset threshold in all the pixel points in the first depth map.

[0116] Then, the second target pixel points are counted to obtain a number of the second target pixel points HighConf_left, and a resolution of the first depth map Total_pixel_number is obtained. Finally, a ratio of the number of the second target pixel points to the resolution of the first depth map is calculated to obtain the first confidence ratio corresponding to the first depth map HighConf_Percentage_left.

[0117] The manner of calculating the second confidence ratio corresponding to the second depth map is similar to the manner of calculating the first confidence ratio corresponding to the first depth map, which is not described here.

[0118] When the first confidence percentage HighConf_Percentage_left corresponding to the first depth map and the second confidence percentage HighConf_Percentage_right corresponding to the second depth map are obtained, the one with the smaller confidence is selected as the epipolar alignment error.

[0119] In the calibration of the binocular camera calibration parameters, the epipolar alignment error is obtained by detecting the epipolar alignment of the binocular disparity map or the binocular depth map, and the target calibration parameters of the binocular camera are determined based on the epipolar alignment error. The calculation of the epipolar alignment error is simplified, the accuracy of the epipolar alignment error is improved, and the accuracy of the calibration of the calibration parameters is improved.

[0120] The third way is to detect the epipolar alignment of the binocular disparity map and the binocular depth map to obtain the epipolar alignment error. The way of detecting the epipolar alignment of the binocular disparity map to obtain the epipolar alignment error and detecting the epipolar alignment of the binocular depth map to obtain the epipolar alignment error is similar to the above embodiments, which will not be repeated here.

[0121] When the two epipolar alignment errors are obtained, the final epipolar alignment error can be calculated by processing the two epipolar alignment errors, such as mean value calculation, average taking, etc.

[0122] The application improves the accuracy of the epipolar alignment error and the accuracy of the calibration of the calibration parameters by detecting the epipolar alignment of the binocular disparity map and the binocular depth map and determining the target calibration parameters of the binocular camera based on the two epipolar alignment errors obtained.

[0123] Step S103: determining the target calibration parameters of the binocular camera to be calibrated based on the epipolar alignment error and the first preset threshold.

[0124] To determine the target calibration parameters of the binocular camera to be calibrated based on the epipolar alignment error and the first preset threshold, the epipolar alignment error and the first preset threshold need to be compared. If the epipolar alignment error is less than or equal to the first preset threshold, the initial calibration parameters of the binocular camera to be calibrated are not invalid, and the initial calibration parameters are used as the target calibration parameters. If the epipolar alignment error is greater than the first preset threshold, the initial calibration parameters of the binocular camera to be calibrated are invalid, and the initial calibration parameters are updated to obtain the target calibration parameters. The first preset threshold can be set according to specific conditions, which will not be limited here.

[0125] If the epipolar line alignment error is greater than the first preset threshold, the initial calibration parameter of the binocular camera to be calibrated is invalid, and the initial calibration parameter is updated to obtain a target calibration parameter, including: if the epipolar line alignment error is greater than the first preset threshold, the initial calibration parameter of the binocular camera to be calibrated is invalid, and the initial calibration parameter is updated; if the difference between the updated initial calibration parameter and the initial calibration parameter is less than a third preset threshold, the initial calibration parameter is taken as the target calibration parameter. The third preset threshold can be set according to specific conditions, which is not limited here.

[0126] In combination with FIG. 2, if the epipolar line alignment error is less than or equal to the first preset threshold, it indicates that the initial calibration parameter of the binocular camera to be calibrated is not invalid, and the current calibration parameter is maintained, that is, the initial calibration parameter is maintained, that is, the current initial calibration parameter is taken as the target calibration parameter.

[0127] If the epipolar line alignment error is greater than the first preset threshold, it indicates that the initial calibration parameter of the binocular camera to be calibrated is invalid, and the initial calibration parameter needs to be updated. However, it should be noted here that since the online calibration of the calibration parameter is different from the offline calibration, the online calibration of the calibration has no rich scene features, therefore, after updating the initial calibration parameter, the updated calibration parameter is compared with the initial calibration parameter, if the difference between the updated calibration parameter and the initial calibration parameter is less than the third preset threshold, the update of the initial calibration parameter is abandoned to avoid picture flickering. If the difference between the updated calibration parameter and the initial calibration parameter is greater than or equal to the third preset threshold, the updated calibration parameter needs to be verified to ensure the accuracy and reliability of the calibration result, and then the updated calibration parameter is taken as the target calibration parameter.

[0128] It should be noted that after updating the initial calibration parameter, the updated initial calibration parameter, that is, the target calibration parameter, needs to be fed back to the left de-distortion image and the right de-distortion image, and the left de-distortion image and the right de-distortion image are corrected based on the target calibration parameter in the subsequent calibration parameter calibration process.

[0129] The embodiment of the application provides a binocular camera calibration method, a binocular camera calibration device, computer equipment and a medium. In the calibration process of the binocular camera, the epipolar alignment error is obtained by performing epipolar alignment detection on a binocular disparity map and / or a binocular depth map, and the target calibration parameter of the binocular camera is determined based on the epipolar alignment error. The calculation of the epipolar alignment error is simplified, the accuracy of the epipolar alignment error is improved, and the accuracy of the calibration of the calibration parameter is improved. In addition, the calibration process of the calibration parameter of the binocular camera can be directly performed in the binocular camera, the additional transmission bandwidth and calculation power are avoided, the cost of the calibration of the calibration parameter is reduced, and the calculation result of the binocular depth calculation, that is, the binocular disparity map and / or the binocular depth map, is reused to calibrate whether the epipolar lines are aligned in real time, so that whether the calibration parameter of the camera is invalid is determined, and the calibration parameter of the invalid camera is calibrated. The result can be obtained in real time and online, the calculation amount of the invalid detection link is saved, the detection result is more in line with the requirement of the binocular depth calculation, and the self-calibration calculation result can also realize feedback and closed loop from the binocular depth calculation.

[0130] It should be understood that the size of the serial number of each step in the above embodiment does not mean the order of execution, and the execution order of each process should be determined according to its function and inherent logic, and should not constitute any limitation on the implementation process of the embodiment of the application.

[0131] The following is the device embodiment of the application. For details not described in detail, reference can be made to the corresponding method embodiments described above.

[0132] FIG. 4 shows a structure schematic diagram of a binocular camera calibration device provided by the embodiment of the application. For the convenience of description, only the part related to the embodiment of the application is shown. The binocular camera calibration device includes an acquisition module 401, a detection module 402 and a parameter calibration module 403, and the details are as follows:

[0133] The acquisition module 401 is configured to acquire a binocular disparity map and / or a binocular depth map. The binocular disparity map and / or the binocular depth map is determined based on the initial calibration parameter of the binocular camera to be calibrated and the left and right eye images acquired by the binocular camera to be calibrated.

[0134] The detection module 402 is configured to perform epipolar alignment detection on the binocular disparity map and / or the binocular depth map to obtain an epipolar alignment error. The epipolar alignment error includes at least one of the following:

[0135] Based on the pixel points of the first disparity map in the binocular disparity map, the pixel points corresponding to the second disparity map in the binocular disparity map are calculated, and the epipolar alignment error is obtained according to the pixel points of the first disparity map and the pixel points corresponding to the second disparity map.

[0136] obtaining the epipolar alignment error based on a confidence of the binocular depth map;

[0137] The calibration module 403 is configured to determine a target calibration parameter of the binocular camera to be calibrated based on the epipolar alignment error and a first preset threshold.

[0138] In an embodiment, the binocular disparity map includes a first disparity map and a second disparity map.

[0139] The detection module 402 is further configured to perform epipolar alignment detection on the binocular disparity map to obtain an epipolar alignment error, including:

[0140] obtaining disparity and coordinates of all pixel points in the first disparity map;

[0141] calculating disparity of a first pixel point and disparity of a second pixel point in the second disparity map based on the disparity and the coordinates of all the pixel points, wherein the first pixel point is a pixel point in the second disparity map corresponding to each of the pixel points, and the second pixel point is a pixel point adjacent to the first pixel point in the second disparity map;

[0142] obtaining the epipolar alignment error according to the disparity of the first pixel point and the disparity of the second pixel point.

[0143] In an embodiment, the detection module 402 is further configured to calculate the coordinates of the first pixel point based on the disparity and the coordinates of all the pixel points.

[0144] obtaining the disparity of the first pixel point from the second disparity map according to the coordinates of the first pixel point;

[0145] calculating the coordinates of the second pixel point based on the coordinates of the first pixel point;

[0146] obtaining the disparity of the second pixel point from the second disparity map according to the coordinates of the second pixel point.

[0147] In an embodiment, the detection module 402 is further configured to determine a first target pixel point corresponding to each of the pixel points based on the disparity of the first pixel point and the disparity of the second pixel point in the second disparity map.

[0148] counting the first target pixel points corresponding to each of the pixel points to obtain a number of the first target pixel points;

[0149] calculating a ratio of the number of the first target pixel points to a number of all the pixel points in the first disparity map, and taking the ratio as the epipolar alignment error.

[0150] In an embodiment, the detection module 402 is further configured to calculate a first difference value and a second difference value by calculating a difference between the disparity of each of the pixel points and the disparity of the first pixel point and the disparity of the second pixel point, respectively.

[0151] respectively, to obtain a first absolute value and a second absolute value;

[0152] If there is an absolute value less than the second preset threshold value in the first absolute value and the second absolute value, the pixel point corresponding to the absolute value is taken as the first target pixel point corresponding to each pixel point.

[0153] In an embodiment, the binocular depth map includes a first depth map and a second depth map.

[0154] The detection module 402 is further configured to calculate a first confidence ratio corresponding to the first depth map and a second confidence ratio corresponding to the second depth map.

[0155] The smallest confidence ratio between the first confidence ratio and the second confidence ratio is taken as the epipolar error.

[0156] In an embodiment, the detection module 402 is further configured to traverse all pixel points in the first depth map, obtain pixel points in the first depth map whose confidence is greater than a third preset threshold value, and obtain second target pixel points.

[0157] The second target pixel points are counted to obtain a number of the second target pixel points.

[0158] The resolution of the first depth map is obtained.

[0159] The ratio of the number of the second target pixel points to the resolution of the first depth map is calculated to obtain the first confidence ratio corresponding to the first depth map.

[0160] In an embodiment, the obtaining module 401 is further configured to obtain left and right eye images by using a binocular camera to be calibrated and obtain initial calibration parameters of the binocular camera to be calibrated, wherein the initial calibration parameters include intrinsic parameters, extrinsic parameters, and de-distortion parameters.

[0161] The left and right eye images are preprocessed to obtain left and right eye distortion images.

[0162] The left and right eye distortion images are de-distorted by using the de-distortion parameters to obtain left and right eye de-distortion images.

[0163] The left and right eye de-distortion images are epipolar corrected by using the intrinsic parameters and the extrinsic parameters to obtain left and right eye epipolar correction images.

[0164] The left and right eye epipolar correction images are matched by using a stereo matching algorithm to obtain a binocular disparity map.

[0165] In an embodiment, the obtaining module 401 is further configured to process the binocular disparity map by using the intrinsic parameters and the extrinsic parameters to obtain a binocular depth map.

[0166] In an embodiment, the parameter calibration module 403 is further configured to: if the epipolar line alignment error is less than or equal to a first preset threshold, the initial calibration parameter of the binocular camera to be calibrated is not invalid, and the initial calibration parameter is taken as the target calibration parameter; and if the epipolar line alignment error is greater than the first preset threshold, the initial calibration parameter of the binocular camera to be calibrated is invalid, the initial calibration parameter is updated, and the target calibration parameter is obtained.

[0167] In an embodiment, the parameter calibration module 403 is further configured to: if the epipolar line alignment error is greater than the first preset threshold, the initial calibration parameter of the binocular camera to be calibrated is invalid, and the initial calibration parameter is updated; and if a difference between the updated initial calibration parameter and the initial calibration parameter is less than a third preset threshold, the initial calibration parameter is taken as the target calibration parameter.

[0168] FIG. 5 of the present application provides a schematic diagram of a computer device. As shown in FIG. 5, the computer device 5 of this embodiment includes a processor 501, a memory 502, and a computer program 503 stored in the memory 502 and executable on the processor 501. The processor 501 implements the steps in each of the binocular camera calibration parameter calibration method embodiments described above when executing the computer program 503, such as steps 101 to 103 shown in FIG. 1. Alternatively, the processor 501 implements the functions of each module / unit in each of the binocular camera calibration parameter calibration apparatus embodiments described above when executing the computer program 503, such as the functions of the modules / units 401 to 403 shown in FIG. 4.

[0169] The present application also provides a readable storage medium having a computer program stored therein, the computer program being executable by a processor to implement the binocular camera calibration parameter calibration method provided by any of the embodiments described above.

[0170] The readable storage medium can be a computer storage medium or a communication medium. The communication medium includes any medium that facilitates the transfer of computer program from one place to another. The computer storage medium can be any available medium that can be accessed by a general or special purpose computer. For example, the readable storage medium is coupled to the processor, so that the processor can read information from the readable storage medium and write information to the readable storage medium. Of course, the readable storage medium can also be an integral part of the processor. The processor and the readable storage medium can be located in an application specific integrated circuit (ASIC). In addition, the ASIC can be located in a user equipment. Of course, the processor and the readable storage medium can also exist as discrete components in a communication device. The readable storage medium can be a read only memory (ROM), a random access memory (RAM), a CD-ROM, a magnetic tape, a floppy disk, and an optical data storage device, etc.

[0171] The application also provides a computer program product, which comprises execution instructions stored in a readable storage medium. At least one processor of the device can read the execution instructions from the readable storage medium, and the at least one processor executes the execution instructions to enable the device to implement the binocular camera calibration method provided in the various embodiments described above.

[0172] In the embodiments of the above device, it should be understood that the processor can be a central processing unit (CPU), and can also be other general-purpose processors, digital signal processors (DSP), application specific integrated circuits (ASIC), etc. The general-purpose processor can be a microprocessor or the processor can also be any conventional processor, etc. The steps of the method disclosed in the application can be directly embodied as execution completed by a hardware processor, or executed by a combination of hardware and software modules in the processor.

[0173] The above embodiments are only used to illustrate the technical solutions of the present application, but not limit them; although the present application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that they can still modify the technical solutions recorded in the foregoing embodiments, or make equivalent replacements for some technical features; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present application.

Claims

1. A method of calibrating parameters of a binocular camera calibration, wherein, The method comprises: obtaining a binocular disparity map and / or a binocular depth map, wherein the binocular disparity map and the binocular depth map are determined based on initial calibration parameters of a binocular camera to be calibrated and binocular images obtained by the binocular camera to be calibrated; performing epipolar alignment detection on the binocular disparity map and / or the binocular depth map to obtain an epipolar alignment error, including at least one of: based on a pixel point in a first disparity map in the binocular disparity map, calculating a corresponding pixel point of a second disparity map in the binocular disparity map, and according to the pixel point in the first disparity map and the corresponding pixel point of the second disparity map, obtaining the epipolar alignment error; based on a confidence of the binocular depth map, obtaining the epipolar alignment error; based on the epipolar alignment error and a first preset threshold, determining target calibration parameters of the binocular camera to be calibrated.

2. The method of claim 1, wherein, The binocular disparity map comprises a first disparity map and a second disparity map; The method further comprises: obtaining disparity and coordinates of all pixel points in the first disparity map; based on the disparity and coordinates of all pixel points, calculating disparity of a first pixel point and disparity of a second pixel point in the second disparity map, wherein the first pixel point is a pixel point in the second disparity map corresponding to each pixel point in all pixel points, and the second pixel point is a pixel point adjacent to the first pixel point in the second disparity map; according to the disparity of the first pixel point and the disparity of the second pixel point, obtaining the epipolar alignment error.

3. The method of calibrating parameters of a binocular camera according to claim 2, wherein, The method further comprises: based on the disparity and coordinates of all pixel points, calculating coordinates of the first pixel point; obtaining the disparity of the first pixel point from the second disparity map according to the coordinates of the first pixel point; based on the coordinates of the first pixel point, calculating coordinates of the second pixel point; obtaining the disparity of the second pixel point from the second disparity map according to the coordinates of the second pixel point.

4. The method according to claim 2, wherein the method further comprises: based on the disparity of the first pixel point and the disparity of the second pixel point in the second disparity map, determining a first target pixel point corresponding to each pixel point in all pixel points; counting the first target pixel points corresponding to each pixel point to obtain a number of the first target pixel points; calculating a ratio of the number of the first target pixel points to a number of all pixel points in the first disparity map, and taking the ratio as the epipolar alignment error.

5. The method of calibrating parameters of a binocular camera according to claim 4, wherein, the method further comprises: respectively calculating a difference value of the disparity of each pixel point and the disparity of the first pixel point and the disparity of the second pixel point to obtain a first difference value and a second difference value. respectively, to obtain a first absolute value and a second absolute value; if there is an absolute value less than a second preset threshold value in the first absolute value and the second absolute value, the pixel point corresponding to the absolute value is taken as a first target pixel point corresponding to each pixel point.

6. The method of calibrating parameters of a binocular camera according to claim 1, wherein, The binocular depth map includes a first depth map and a second depth map. The epipolar alignment detection on the binocular depth map includes: calculating a first confidence ratio corresponding to the first depth map and a second confidence ratio corresponding to the second depth map; the minimum confidence ratio in the first confidence ratio and the second confidence ratio is taken as the epipolar alignment error.

7. The method of calibrating parameters of a binocular camera according to claim 6, wherein, The calculation of the first confidence ratio corresponding to the first depth map includes: traversing all pixel points in the first depth map to obtain pixel points in the first depth map whose confidence is greater than a third preset threshold value, to obtain second target pixel points; counting the second target pixel points to obtain the number of the second target pixel points; obtaining the resolution of the first depth map; calculating the ratio of the number of the second target pixel points to the resolution of the first depth map to obtain the first confidence ratio corresponding to the first depth map.

8. The method of claim 1, wherein, The binocular disparity map includes: obtaining left and right eye images by the to-be-calibrated binocular camera and obtaining initial calibration parameters of the to-be-calibrated binocular camera, wherein the initial calibration parameters include intrinsic parameters, extrinsic parameters and de-distortion parameters; preprocessing the left and right eye images to obtain left and right eye distortion images; performing distortion correction on the left and right eye distortion images by using the de-distortion parameters to obtain left and right eye de-distortion images; performing epipolar correction on the left and right eye de-distortion images by using the intrinsic parameters and the extrinsic parameters to obtain left and right eye epipolar correction images; performing matching processing on the left and right eye epipolar correction images by using a stereo matching algorithm to obtain the binocular disparity map.

9. The method of calibrating parameters of a binocular camera according to claim 8, wherein, The binocular depth map includes: processing the binocular disparity map by using the intrinsic parameters and the extrinsic parameters to obtain the binocular depth map.

10. The method of claim 1, wherein, The determination of the target calibration parameters of the to-be-calibrated binocular camera based on the epipolar alignment error and a first preset threshold value includes: if the epipolar alignment error is less than or equal to the first preset threshold value, the initial calibration parameters of the to-be-calibrated binocular camera are not invalid, and the initial calibration parameters are taken as the target calibration parameters; if the epipolar alignment error is greater than the first preset threshold value, the initial calibration parameters of the to-be-calibrated binocular camera are invalid, and the initial calibration parameters are updated to obtain the target calibration parameters.

11. The method of calibrating parameters of a binocular camera according to claim 10, wherein, The updating of the initial calibration parameters when the epipolar alignment error is greater than the first preset threshold value and the initial calibration parameters of the to-be-calibrated binocular camera are invalid to obtain the target calibration parameters includes: if the epipolar alignment error is greater than the first preset threshold value, the initial calibration parameters of the to-be-calibrated binocular camera are invalid, and the initial calibration parameters are updated; If a difference between the updated initial calibration parameter and the initial calibration parameter is less than a third preset threshold, the initial calibration parameter is taken as the target calibration parameter.

12. A binocular camera calibration apparatus, wherein, The method comprises the steps of: obtaining a binocular disparity map and a binocular depth map, wherein the binocular disparity map and the binocular depth map are determined based on an initial calibration parameter of a binocular camera to be calibrated and left and right eye images obtained by the binocular camera to be calibrated; detecting epipolar alignment of the binocular disparity map or the binocular depth map to obtain an epipolar alignment error, the epipolar alignment error comprising at least one of: calculating a pixel point corresponding to a second disparity map in the binocular disparity map based on a pixel point of a first disparity map in the binocular disparity map, and obtaining the epipolar alignment error according to the pixel point of the first disparity map and the pixel point corresponding to the second disparity map; obtaining the epipolar alignment error based on a confidence of the binocular depth map; determining a target calibration parameter of the binocular camera to be calibrated based on the epipolar alignment error and a first preset threshold.

13. A computer device, wherein, comprise a memory and one or more processors in communication with the memory; the memory stores instructions executable by the one or more processors, and the instructions are executed by the one or more processors to cause the one or more processors to implement the binocular camera calibration parameter calibration method according to any one of claims 1 to 11.

14. A computer readable storage medium, wherein, comprise a program or instructions, when the program or instructions are run on a computer, the binocular camera calibration parameter calibration method according to any one of claims 1 to 10 is implemented.

15. A computer program product, wherein, comprise a computer program, when the computer program is executed by a processor, the binocular camera calibration parameter calibration method according to any one of claims 1 to 10 is implemented.

Images