Image feature matching method, computer device and storage medium
By combining color difference and distance difference in image feature matching, the error problem caused by illumination and occlusion in low-texture region matching is solved, thus improving the accuracy and reliability of image feature matching.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- HON HAI PRECISION INDUSTRY CO LTD
- Filing Date
- 2022-05-31
- Publication Date
- 2026-06-30
AI Technical Summary
Existing image stereo matching techniques are easily affected by factors such as lighting, occlusion, and low-texture areas in the image scene when matching low-texture regions, leading to incorrect matching results. This is especially true when an entire region is low-texture and of the same color, where relying solely on color for judgment can easily cause errors.
Image feature matching is performed by combining color difference values and distance differences. Weak texture regions are identified through edge detection algorithms, feature points are detected, feature points are matched using color difference values, and positional differences are calculated to determine the matching point.
It effectively improves the accuracy of image feature matching, avoids judgment errors caused by similar colors, and improves the reliability of matching.
Smart Images
Figure CN117197503B_ABST
Abstract
Description
Technical Field
[0001] This application relates to the field of image processing technology, and in particular to an image feature matching method, a computer device, and a storage medium. Background Technology
[0002] Stereo image matching primarily involves establishing pixel correspondences between two 2D images, calculating the disparity between them, and acquiring the disparity image. Existing matching techniques often produce errors due to factors such as lighting conditions, occlusion, and low-texture areas in the image scene. This is especially true when matching low-texture areas; in a region where the entire area is low-texture and of the same color, small-scale matching decisions can be misleading, and relying solely on color to determine a successful match can lead to significant errors. Summary of the Invention
[0003] In view of the above, it is necessary to provide an image feature matching method, computer device and storage medium that can combine color difference value and distance difference to perform image feature matching, thereby effectively improving the accuracy of image feature matching.
[0004] The image feature matching method includes: acquiring a first image and a second image;
[0005] The first weak texture region of the first image and the second weak texture region of the second image are determined based on the edge detection algorithm.
[0006] Detect the first feature point of the first weak texture region and the second feature point of the second weak texture region;
[0007] The first feature point and the second feature point are matched based on the color difference value, and the corresponding point with the smallest color difference value with the first feature point is determined from the second feature point as the target corresponding point;
[0008] Determine the positional difference between the target corresponding point and the first feature point;
[0009] The matching point of the first feature point is determined based on the positional difference.
[0010] Optionally, determining the first weak texture region of the first image and the second weak texture region of the second image based on the edge detection algorithm includes:
[0011] The edge detection algorithm is used to determine the first object edge of the first image, the first object edge is marked, and the region within the first object edge is taken as the first weak texture region. The edge detection algorithm includes the Canny algorithm.
[0012] The edge detection algorithm is used to determine the edge of the second object in the second image, the edge of the second object is marked, and the region within the edge of the second object is taken as the second weak texture region.
[0013] Optionally, detecting the first feature point of the first weak texture region and the second feature point of the second weak texture region includes:
[0014] The Harris corner points of the first image are detected as the first feature points;
[0015] The Harris corner points of the second image are detected as the second feature points.
[0016] Optionally, the step of matching the first feature point and the second feature point based on the color difference value, and determining the corresponding point with the smallest color difference value to the first feature point from the second feature points as the target corresponding point, includes:
[0017] The first image is set as the reference image, and the corresponding point of any first feature point and the number of the corresponding points are determined from the second feature points according to the epipolar constraint.
[0018] When the number of corresponding points is greater than a preset value, the color difference value between any first feature point and each corresponding point is calculated, and the corresponding point with the smallest color difference value is taken as the target corresponding point.
[0019] Optionally, determining the positional difference between the target corresponding point and the first feature point includes:
[0020] Calculate the first distance parameter between the first feature point and the edge of the first object;
[0021] Calculate the second distance parameter between the target point and the edge of the second object;
[0022] The positional difference is calculated based on the first distance parameter and the second distance parameter.
[0023] Optionally, the first distance parameter and the second distance parameter each include a preset number of directional distance values.
[0024] Optionally, calculating the position difference based on the first distance parameter and the second distance parameter includes:
[0025] Calculate the position difference for each direction between the first distance parameter and the second distance parameter;
[0026] The position difference is obtained by weighting the position difference corresponding to each direction according to the preset weights.
[0027] Optionally, determining the matching point of the first feature point based on the positional difference includes:
[0028] Determine whether the positional difference is within a preset threshold range, and use the target corresponding point in the second feature point that is within the threshold range as the matching point of the first feature point.
[0029] The computer-readable storage medium stores at least one instruction, which, when executed by a processor, implements the image feature matching method.
[0030] The computer device includes a memory and at least one processor, the memory storing at least one instruction which, when executed by the at least one processor, implements the image feature matching and checking method.
[0031] Compared to existing technologies, the image feature matching method, computer device, and storage medium described above can combine color difference values and distance differences to perform image feature matching. It not only uses color as a judgment criterion, but also uses the distance from the edge to the pixel to describe the position of the pixel in the image, increasing the information available for image feature matching judgment. This can avoid judgment errors due to similar colors and effectively improve the accuracy of image feature matching. Attached Figure Description
[0032] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only embodiments of this application. For those skilled in the art, other drawings can be obtained based on the provided drawings without creative effort.
[0033] Figure 1 This is a flowchart of the image feature matching method provided in the embodiments of this application.
[0034] Figure 2 This is an example diagram of the first feature point and the target corresponding point provided in the embodiments of this application.
[0035] Figure 3 This is an example diagram illustrating the calculation of the first distance parameter along the "up" direction provided in the embodiments of this application.
[0036] Figure 4 This is an example diagram illustrating the calculation of the first distance parameter along the "left" direction provided in the embodiments of this application.
[0037] Figure 5 This is an example diagram of the first feature points A and C and the target corresponding points B and D provided in the embodiments of this application.
[0038] Figure 6This is an architectural diagram of the computer device provided in the embodiments of this application.
[0039] Explanation of main component symbols
[0040] Computer devices 3 processor 32 memory 31 Image feature matching system 30
[0041] The following detailed description, in conjunction with the accompanying drawings, will further illustrate this application. Detailed Implementation
[0042] To better understand the above-mentioned objectives, features, and advantages of this application, the application will be described in detail below with reference to the accompanying drawings and specific embodiments. It should be noted that, unless otherwise specified, the embodiments and features described in these embodiments can be combined with each other.
[0043] Numerous specific details are set forth in the following description to provide a thorough understanding of this application. The described embodiments are merely some, not all, of the embodiments described herein. All other embodiments obtained by those skilled in the art based on the embodiments of this application without inventive effort are within the scope of protection of this application.
[0044] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs. The terminology used herein in the specification of this application is for the purpose of describing particular embodiments only and is not intended to be limiting of this application.
[0045] See Figure 1 The diagram shown is a flowchart of an image feature matching method according to a preferred embodiment of this application.
[0046] In this embodiment, the image feature matching method can be applied to a computer device (e.g., Figure 6 For computer devices that require image feature matching, the image feature matching function provided by the method of this application can be directly integrated into the computer device, or it can be run on the computer device in the form of a software development kit (SDK).
[0047] like Figure 1 As shown, the image feature matching method specifically includes the following steps. Depending on different needs, the order of the steps in this flowchart can be changed, and some steps can be omitted.
[0048] Step S1: The computer device acquires the first image and the second image.
[0049] In one embodiment, the computer device can acquire the first image and the second image in response to user input. Alternatively, the computer device can be directly connected to one or more imaging devices to directly acquire multiple images, or it can pre-store the first image and the second image in the computer device's memory, or pre-store them in another device communicatively connected to the computer device. The first image and the second image can be two photographs of the same object or scene taken from two different perspectives.
[0050] In one embodiment, the computer device may also acquire an initial image, perform scale and resolution transformations on the initial image, and obtain two images at different scales and resolutions as the first image and the second image. For example, the computer device may perform scale transformation on the initial image according to a pre-set scale factor based on the scale-invariant feature transform (SIFT) technique, and then perform resolution transformation on the initial image based on the Gaussian Blur algorithm.
[0051] In the embodiments of this application, the first image and the second image have the same image size.
[0052] Step S2: The computer device determines the first weak texture region of the first image and the second weak texture region of the second image based on the edge detection algorithm.
[0053] In one embodiment, determining the first weak texture region of the first image and the second weak texture region of the second image based on the edge detection algorithm includes:
[0054] The edge detection algorithm is used to determine the first object edge of the first image, the first object edge is marked, and the region within the first object edge is taken as the first weak texture region. The edge detection algorithm includes the Canny algorithm.
[0055] The edge detection algorithm is used to determine the edge of the second object in the second image, the edge of the second object is marked, and the region within the edge of the second object is taken as the second weak texture region.
[0056] In one embodiment, determining the first object edge of the first image using the Canny algorithm includes:
[0057] The gray values of the pixels in the first image are obtained, a filtering operator is generated based on the Gaussian formula, and the first image is denoised according to the gray values of the pixels in the first image and the filtering operator to obtain the denoised first image.
[0058] Gradient calculation is performed on the denoised first image using a preset Gaussian filter to obtain the gradient of any pixel in the first image, thereby obtaining the gradient magnitude image of the denoised first image.
[0059] Non-maximum suppression processing is performed on the gradient magnitude image, and the pixel corresponding to the local maximum value of the gradient magnitude in the gradient magnitude image is taken as the initial object edge point. If the gradient magnitude at any pixel is greater than the gradient magnitude of the two adjacent pixels along the gradient direction, then the pixel is taken as the pixel corresponding to the local maximum value of the gradient magnitude.
[0060] The initial object edge points are filtered based on a dual threshold algorithm to obtain target edge points, wherein the filtering includes removing false edge points that are below a preset threshold.
[0061] Connect the target edge points to obtain the edge of the first object.
[0062] In one embodiment, the computer device may also perform edge detection on the first image using the Roberts operator, Prewitt operator, Sobel operator, or Laplacian operator.
[0063] In one embodiment, a computer device can mark the edge of the first object by using a pre-written MATLAB code program to set the edge of the first object to an arbitrary color.
[0064] In one embodiment, the method for determining the edge of the second object is the same as the method for determining the edge of the first object, and the method for marking the edge of the second object is the same as the method for marking the edge of the first object.
[0065] In one embodiment, a weak texture region is a relative concept; the same object or scene may be a weak texture region in a higher resolution image, but have richer details in a lower resolution image.
[0066] Step S3: The computer device detects the first feature point of the first weak texture region and the second feature point of the second weak texture region.
[0067] In one embodiment, extracting the first feature points of the first weak texture region and the second feature points of the second weak texture region includes:
[0068] The Harris corner points of the first image are detected as the first feature points;
[0069] The Harris corner points of the second image are detected as the second feature points.
[0070] In one embodiment, a computer device can use the cornerHairrs() function in OpenCV to detect Harris corners in the first image and the second image. The principle of the Harris corner detection algorithm includes: creating a local window centered on any pixel in the image; if a small movement of the local window in any direction results in a significant change in grayscale value (e.g., the change in grayscale value is greater than a preset change threshold), then the pixel is considered a Harris corner.
[0071] Step S4: The computer device matches the first feature point and the second feature point based on the color difference value, and determines the corresponding point with the smallest color difference value with the first feature point from the second feature point as the target corresponding point.
[0072] In one embodiment, the step of matching the first feature point and the second feature point based on the color difference value, and determining the corresponding point with the smallest color difference value to the first feature point from the second feature points as the target corresponding point, includes:
[0073] The first image is set as the reference image, and the corresponding point of any first feature point and the number of the corresponding points are determined from the second feature points according to the epipolar constraint.
[0074] When the number of corresponding points is greater than a preset value (e.g., 1), the color difference value between any first feature point and each corresponding point is calculated, and the corresponding point with the smallest color difference value is taken as the target corresponding point.
[0075] In one embodiment, the epipolar constraint describes the constraints formed by the image point and the camera optical center under the projection model when the same point is projected onto images from two different viewpoints. Based on the epipolar constraint, the search range for feature point matching can be narrowed.
[0076] For example Figure 2 The image shown is an example of a first feature point and a target corresponding point provided in an embodiment of this application. The left image represents the first image, the right image represents the second image, and the shaded areas represent the first weak texture region and the second weak texture region, respectively. Each small square represents a pixel (including the first feature point or the second feature point). As indicated by the dashed arrow, when the color difference value of any first feature point (the pixel where the dashed arrow in the left image starts) is the same as that of the two adjacent corresponding points of any first feature point (the pixel where the dashed arrow in the right image starts) is the same, two target corresponding points will appear. Therefore, color cannot be used as the sole criterion for matching.
[0077] Step S5: The computer device determines the positional difference between the target corresponding point and the first feature point.
[0078] In one embodiment, determining the positional difference between the target corresponding point and the first feature point includes:
[0079] Calculate the first distance parameter between the first feature point and the edge of the first object;
[0080] Calculate the second distance parameter between the target point and the edge of the second object;
[0081] The positional difference is calculated based on the first distance parameter and the second distance parameter.
[0082] In one embodiment, the first distance parameter and the second distance parameter each include a preset number of directional distance values (for example, calculated based on each pixel occupying a length of 1).
[0083] The stopping criteria for calculating the distance value along any of the preset number of directions include: stopping the calculation when the calculation reaches the edge of the object (including the edge of the first object or the edge of the second object) or the edge of the image (including the edge of the first image or the edge of the second image).
[0084] For example, the preset quantity can be 4, and the direction of the preset quantity can include the four directions of "up, down, left, and right". The first distance parameter can include the distance between the first feature point and the edge of the first object in the four directions of "up, down, left, and right".
[0085] For example Figure 3 The figure shown is an example diagram for calculating the first distance parameter along the "up" direction according to an embodiment of this application. Figure 3 The image represents the first image, where the shaded area represents the edge of the first object, each small square represents a pixel (including the first feature point), and the value of each small square represents the parameter in the "up" direction of the first distance parameter between each pixel and the edge of the first object. The pixel at the starting point of the solid arrow indicates the starting position for calculating the parameter in the "up" direction of the first distance parameter, and the pixel at the arrowhead indicates the ending position (including the edge of the first image or the edge of the first image) when the calculation stops according to the stopping criterion.
[0086] For example Figure 4 The figure shown is an example diagram for calculating the first distance parameter along the "left" direction according to an embodiment of this application. Figure 4The image represents the first image, where the shaded area represents the edge of the first object, each small square represents a pixel (including the first feature point), and the value of each small square represents the parameter in the "left" direction of the first distance parameter between each pixel and the edge of the first object. The pixel at the starting point of the solid arrow indicates the starting position for calculating the parameter in the "left" direction of the first distance parameter, and the pixel at the arrowhead indicates the ending position (including the edge of the first image or the edge of the first image) when the calculation stops according to the stopping criterion.
[0087] In one embodiment, the method for calculating the parameters in the "right" and "down" directions of the first distance parameter for the edge of the first object is similar to the example described above. The method for calculating the second distance parameter is the same as the method for calculating the first distance parameter.
[0088] For example Figure 5 As shown, following the above method, the first distance parameter of the first feature point A in the shaded area of the first image on the left is (1,1,1,1) (corresponding to the distance parameters in the four directions of "up, down, left, and right"), where the shaded area of the first image represents the first weak texture region. Similarly, following the above method, the second distance parameter of the target corresponding point B in the shaded area of the second image on the right is (2,1,0,1) (corresponding to the distance parameters in the four directions of "up, down, left, and right"), where the shaded area of the second image represents the second weak texture region.
[0089] Similarly, following the above method, the first distance parameter of the first feature point C in the shaded area of the first image on the left is (2,1,1,1) (corresponding to the distance parameters in the four directions of "up, down, left, and right" respectively); the second distance parameter of the target point D in the upper left corner of the shaded area of the second image on the right is (1,0,1,1) (corresponding to the distance parameters in the four directions of "up, down, left, and right" respectively).
[0090] In one embodiment, calculating the position difference based on the first distance parameter and the second distance parameter includes:
[0091] Calculate the position difference for each direction between the first distance parameter and the second distance parameter;
[0092] The position difference is obtained by weighting the position difference corresponding to each direction according to the preset weights.
[0093] For example, for instance Figure 5As shown, the first distance parameter of the first feature point A is (1,1,1,1), and the second distance parameter of the target corresponding point B is (2,1,0,1). Therefore, when calculating the position difference for each direction, these two distance parameters can be treated as vectors and subtracted accordingly: (1,1,1,1) - (2,1,0,1) = (-1,0,1,0). This yields the position differences between the first feature point A and the target corresponding point B in the four directions (up, down, left, right) as -1, 0, 1, and 0, respectively. It should be noted that in other embodiments, the position difference can also be taken as the absolute value.
[0094] Similarly, the positional differences between the first feature point C and the target corresponding point D in the four directions of "up, down, left, and right" can be obtained as 1, 1, 0, and 0, respectively.
[0095] In one embodiment, the preset weights can be set based on the same items or scenes in the first image and the second image. For example, when the same item is a vertical fence, more emphasis should be placed on its left-right calibration, and the weights for the left and right directions can be set to 2, while the weights for the up and down directions can be set to 1.
[0096] For example, the positional differences between the first feature point A and the target corresponding point B in the four directions of "up, down, left, and right" are -1, 0, 1, and 0, respectively. The preset weights for the left and right directions are 2, and the preset weights for the up and down directions are 1. Then, the weighted calculation yields: -1×1+0×1+1×2+0×2=1, that is, the positional difference between the first feature point A and the target corresponding point B is 1.
[0097] Similarly, the positional difference between the first feature point C and the target corresponding point D is: 1×1+1×1+0×2+0×2=2.
[0098] Step S6: The computer device determines the matching point of the first feature point based on the positional difference.
[0099] In one embodiment, determining the matching point of the first feature point based on the positional difference includes:
[0100] Determine whether the positional difference is within a preset threshold range (e.g., 1), and use the target corresponding point in the second feature point that is within the threshold range as the matching point of the first feature point.
[0101] For example, if the positional difference between the first feature point A and the target corresponding point B is 1, which is within the preset threshold range of 1, then the target corresponding point B is taken as the matching point of the first feature point A. If the positional difference between the first feature point C and the target corresponding point D is 2, which is not within the preset threshold range of 1, then the target corresponding point D is not the matching point of the first feature point C.
[0102] The above Figure 1 The image feature matching method of this application is described in detail below. Figure 6 The functional modules of the software system for implementing the image feature matching method and the hardware device architecture for implementing the image feature matching method are introduced.
[0103] It should be understood that the embodiments described are for illustrative purposes only and are not limited to this structure in the scope of the patent application.
[0104] See Figure 6 The diagram shown is a structural schematic of a computer device provided in a preferred embodiment of this application.
[0105] In a preferred embodiment of this application, the computer device 3 includes a memory 31 and at least one processor 32. Those skilled in the art should understand that... Figure 6 The structure of the computer device shown does not constitute a limitation of the embodiments of this application. It can be a bus-type structure or a star-type structure. The computer device 3 may also include more or fewer other hardware or software than shown, or different component arrangements.
[0106] In some embodiments, the computer device 3 includes a terminal capable of automatically performing numerical calculations and / or information processing according to pre-set or stored instructions, the hardware of which includes, but is not limited to, microprocessors, application-specific integrated circuits, programmable gate arrays, digital processors, and embedded devices.
[0107] It should be noted that the computer device 3 described is merely an example. Other existing or future electronic products that are suitable for this application should also be included within the scope of protection of this application and are incorporated herein by reference.
[0108] In some embodiments, the memory 31 is used to store program code and various data. For example, the memory 31 can be used to store an image feature matching system 30 installed in the computer device 3, and to enable high-speed, automatic access to programs or data during the operation of the computer device 3. The memory 31 includes read-only memory (ROM), programmable read-only memory (PROM), erasable programmable read-only memory (EPROM), one-time programmable read-only memory (OTPROM), electrically erasable programmable read-only memory (EEPROM), compact disc read-only memory (CD-ROM) or other optical disc storage, disk storage, magnetic tape storage, or any other computer-readable storage medium capable of carrying or storing data.
[0109] In some embodiments, the at least one processor 32 may be composed of integrated circuits, such as a single-packaged integrated circuit or multiple integrated circuits packaged with the same or different functions, including combinations of one or more central processing units (CPUs), microprocessors, digital processing chips, graphics processors, and various control chips. The at least one processor 32 is the control unit of the computer device 3, connecting various components of the computer device 3 via various interfaces and lines. It executes programs or modules stored in the memory 31 and calls data stored in the memory 31 to perform various functions of the computer device 3 and process data, such as executing... Figure 1 The image feature matching function shown.
[0110] In some embodiments, the image feature matching system 30 operates in a computer device 3. The image feature matching system 30 may include multiple functional modules composed of program code segments. The program code of each program segment in the image feature matching system 30 may be stored in the memory 31 of the computer device 3 and executed by at least one processor 32 to achieve... Figure 1 The image feature matching function shown.
[0111] In this embodiment, the image feature matching system 30 can be divided into multiple functional modules according to the functions it performs. A module, as referred to in this application, is a series of computer program segments that can be executed by at least one processor and perform a fixed function, and is stored in memory.
[0112] Although not shown, the computer device 3 may also include a power supply (such as a battery) to power the various components. Preferably, the power supply can be logically connected to the at least one processor 32 via a power management device, thereby enabling functions such as charging, discharging, and power consumption management. The power supply may also include one or more DC or AC power supplies, recharging devices, power failure testing circuits, power converters or inverters, power status indicators, and other arbitrary components. The computer device 3 may also include various sensors, Bluetooth modules, Wi-Fi modules, etc., which will not be described in detail here.
[0113] It should be understood that the embodiments described are for illustrative purposes only and are not limited to this structure in the scope of the patent application.
[0114] The integrated unit implemented as a software functional module described above can be stored in a computer-readable storage medium. This software functional module, stored in a storage medium, includes several instructions to cause a computer device (which may be a server, personal computer, etc.) or processor to execute portions of the methods described in the various embodiments of this application.
[0115] The memory 31 stores program code, and the at least one processor 32 can call the program code stored in the memory 31 to execute related functions. The program code stored in the memory 31 can be executed by the at least one processor 32 to realize the functions of each module to achieve the purpose of image feature matching.
[0116] In the several embodiments provided in this application, it should be understood that the disclosed apparatus and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of modules is only a logical functional division, and other division methods may be used in actual implementation.
[0117] The modules described as separate components may or may not be physically separate. The components shown as modules may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs.
[0118] Furthermore, the functional modules in the various embodiments of this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or in the form of hardware plus software functional modules.
[0119] It will be apparent to those skilled in the art that this application is not limited to the details of the exemplary embodiments described above, and that this application can be implemented in other specific forms without departing from the spirit or essential characteristics of this application. Therefore, the embodiments should be considered exemplary and non-limiting in all respects, and the scope of this application is defined by the appended claims rather than the foregoing description. Thus, all variations falling within the meaning and scope of equivalents of the claims are intended to be embraced within this application. No reference numerals in the claims should be construed as limiting the scope of the claims. Furthermore, it is clear that the word "comprising" does not exclude other elements or, and the singular does not exclude the plural. Multiple elements or devices recited in the apparatus claims may also be implemented by a single element or device in software or hardware. The terms "first," "second," etc., are used to indicate names and do not indicate any particular order.
[0120] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of this application and are not intended to limit it. Although this application has been described in detail with reference to the above preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of this application without departing from the spirit and scope of the technical solutions of this application.
Claims
1. An image feature matching method, applied to a computer device, characterized in that, The method includes: Acquire the first image and the second image; The first weak texture region of the first image and the second weak texture region of the second image are determined based on the edge detection algorithm. Detect the first feature point of the first weak texture region and the second feature point of the second weak texture region; Matching the first feature point and the second feature point based on color difference values, and determining the corresponding point with the smallest color difference value from the first feature point among the second feature points as the target corresponding point, includes: setting the first image as a reference image, determining the corresponding point of any first feature point and the number of corresponding points from the second feature points according to epipolar constraints; when the number of corresponding points is greater than a preset value, calculating the color difference value between any first feature point and each corresponding point, and taking the corresponding point with the smallest color difference value as the target corresponding point; Determining the positional difference between the target corresponding point and the first feature point includes: calculating a first distance parameter between the first feature point and a first object edge in the first image; calculating a second distance parameter between the target corresponding point and a second object edge in the second image; and calculating the positional difference based on the first distance parameter and the second distance parameter. The first distance parameter and the second distance parameter each include a preset number of directional distance values. Calculating the positional difference based on the first distance parameter and the second distance parameter includes: calculating the positional difference corresponding to each direction in the first distance parameter and the second distance parameter; and performing a weighted calculation on the positional differences corresponding to each direction according to preset weights to obtain the positional difference. The matching point of the first feature point is determined based on the positional difference.
2. The image feature matching method according to claim 1, characterized in that, The step of determining the first weak texture region of the first image and the second weak texture region of the second image based on the edge detection algorithm includes: The edge detection algorithm is used to determine the first object edge of the first image, the first object edge is marked, and the region within the first object edge is taken as the first weak texture region. The edge detection algorithm includes the Canny algorithm. The edge detection algorithm is used to determine the edge of the second object in the second image, the edge of the second object is marked, and the region within the edge of the second object is taken as the second weak texture region.
3. The image feature matching method according to claim 1, characterized in that, The detection of the first feature point in the first weak texture region and the second feature point in the second weak texture region includes: The Harris corner points of the first image are detected as the first feature points; The Harris corner points of the second image are detected as the second feature points.
4. The image feature matching method according to claim 1, characterized in that, Determining the matching point of the first feature point based on the positional difference includes: Determine whether the positional difference is within a preset threshold range, and use the target corresponding point in the second feature point that is within the threshold range as the matching point of the first feature point.
5. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores at least one instruction, which, when executed by a processor, implements the image feature matching method as described in any one of claims 1 to 4.
6. A computer device, characterized in that, The computer device includes a memory and at least one processor, the memory storing at least one instruction which, when executed by the at least one processor, implements the image feature matching method as described in any one of claims 1 to 4.