X corner point recognition method and device

Abstract

The present invention provides an X corner point recognition method and device. It is applied to the fields of computer vision and image processing technology. The method comprises: determining a plurality of spatially symmetrical pixel points around a point to be detected; judging whether the plurality of pixel points have high contrast characteristics according to the pixel values ​​of the plurality of pixel points; if the plurality of pixel points have high contrast characteristics, judging whether the plurality of pixel points have symmetric characteristics according to the pixel values ​​of the plurality of pixel points; if the plurality of pixel points have symmetric characteristics, returning to the step of determining a plurality of spatially symmetrical pixel points around the point to be detected to re-determine a plurality of other spatially symmetrical pixel points, until all pixel points within a certain range around the point to be detected have high contrast characteristics and symmetric characteristics, then the point to be detected is judged to be an X corner point. The present invention effectively improves the accuracy of X corner point recognition, reduces the possibility of misjudgment through multiple verification steps, and improves the detection speed.

Description

Technical Field

[0001] The present invention relates to the technical field of computer vision and image processing, and in particular to an X-corner point recognition method and device. Background Art

[0002] Visual measurement is a key application in the field of computer vision and image processing. It relies on effective detection and precise identification of targets to obtain accurate measurement information. In order to simplify the detection process and improve the accuracy of detection, manual markers are often used in practical applications. Black and white X-shaped corner points have become a widely used choice in this field due to their significant contrast, easy detection characteristics and simple production process. The high contrast of black and white X-shaped corner points allows them to maintain good visibility under various lighting conditions, and the simple production process allows them to be quickly deployed in various experimental and industrial applications, greatly improving detection efficiency and accuracy.

[0003] In some specific application scenarios, the recognition of X-shaped corner points requires not only fast speed, but also avoids misidentifying other points as X-shaped corner points. This requires the corresponding detection algorithm to be able to quickly and accurately determine whether each pixel in the image belongs to an X-shaped corner point. In practical applications, there are a variety of algorithms that can be used to detect X-shaped corner points. For example, the Harris corner detection algorithm used in the OpenCV library performs corner detection based on the maximum curvature point of the image grayscale distribution. The Harris operator can detect the features of the corner points by calculating the gradient information of the local area of ​​the image. However, due to the large amount of calculations involved in its detection process, the amount of calculation is large and the complexity is high, resulting in low efficiency in processing large-scale image data. Summary of the invention

[0004] In view of this, one aspect of the present invention provides an X-corner point recognition method.

[0005] Determine a number of spatially symmetrical pixel points around the point to be detected;

[0006] Determining whether the plurality of pixel points have a high contrast characteristic according to the pixel values ​​of the plurality of pixel points;

[0007] If the plurality of pixel points have a high contrast characteristic, determining whether the plurality of pixel points have a symmetric characteristic according to the pixel values ​​of the plurality of pixel points;

[0008] If the plurality of pixel points have symmetric characteristics, the step of determining a plurality of spatially symmetric pixel points around the point to be detected is returned to re-determine a plurality of other spatially symmetric pixel points until all the pixel points within a certain range around the point to be detected have high contrast characteristics and symmetric characteristics, and the point to be detected is determined to be an X corner point.

[0009] Optionally, if the plurality of pixel points do not have a high contrast characteristic, or the plurality of pixel points do not have a symmetric characteristic, it is determined that the point to be detected is not an X corner point, and the point to be detected is re-determined.

[0010] Optionally, the plurality of pixel points are a plurality of pixel points evenly distributed in a ring area centered on the point to be detected.

[0011] Optionally, several spatially symmetrical pixel points are determined around the point to be detected, including:

[0012] Get the preset distance;

[0013] A number of pixel points are determined according to the position of the point to be detected and the preset distance.

[0014] Optionally, judging whether the plurality of pixel points have a high contrast characteristic according to the pixel values ​​of the plurality of pixel points includes:

[0015] Calculating a contrast characteristic value according to a pixel point having a maximum grayscale value and a pixel point having a minimum grayscale value among at least some of the plurality of pixel points;

[0016] Determining whether the contrast characteristic value exceeds a contrast threshold;

[0017] If the contrast characteristic value exceeds the contrast threshold, it is determined that the plurality of pixels have a high contrast characteristic.

[0018] Optionally, the contrast characteristic value is a difference between the maximum grayscale value and the minimum grayscale value.

[0019] Optionally, judging whether the plurality of pixel points have a symmetric characteristic according to the pixel values ​​of the plurality of pixel points includes:

[0020] Determine, from at least some of the plurality of pixels, a pixel point with a maximum grayscale value and a spatially symmetrical point thereof, and a pixel point with a minimum grayscale value and a spatially symmetrical point thereof;

[0021] Calculate a first symmetrical eigenvalue according to the maximum grayscale value and the grayscale value of its spatially symmetrical point, and calculate a second symmetrical eigenvalue according to the minimum grayscale value and the grayscale value of its spatially symmetrical point;

[0022] Whether the plurality of pixel points have a symmetric characteristic is determined according to the first symmetric eigenvalue and the second symmetric eigenvalue.

[0023] Optionally, the first symmetric eigenvalue is the absolute value of the difference between the maximum grayscale value and the grayscale value of its spatially symmetrical point; the second symmetric eigenvalue is the absolute value of the difference between the minimum grayscale value and the grayscale value of its spatially symmetrical point.

[0024] Optionally, judging whether the plurality of pixel points have a symmetric characteristic according to the first symmetric eigenvalue and the second symmetric eigenvalue includes:

[0025] Determining whether the first symmetric eigenvalue and the second symmetric eigenvalue are lower than a symmetry threshold;

[0026] If the first symmetric eigenvalue or the second symmetric eigenvalue is lower than a symmetry threshold, it is determined that the plurality of pixels have a symmetric characteristic.

[0027] Optionally, at least some of the pixel points are pixel points distributed in a half area of ​​the annular area.

[0028] A second aspect of the present invention provides an X corner point recognition device, the device comprising: a processor and a memory connected to the processor; wherein the memory stores instructions executable by the processor, and the instructions are executed by the processor so that the processor performs the above-mentioned X corner point recognition method.

[0029] The present invention ensures that the point to be detected is accurately identified as an X corner point by step-by-step screening and verification. First, several spatially symmetrical pixels are determined around the point to be detected, and it is determined whether these pixels have high contrast characteristics. Then, it is further verified whether these pixels have symmetric characteristics. If both characteristics are met, other spatially symmetrical pixels are reselected to perform the same judgment process. Only when all selected pixels have high contrast and symmetric characteristics, the point to be detected is finally confirmed to be an X corner point. This method effectively improves the accuracy of X corner point recognition and reduces the possibility of misjudgment through multiple verification steps. At the same time, most of the pixels in an entire image that are not X corner points can be excluded in the initial step through a small amount of detection, thereby greatly reducing the amount of calculation for detecting an entire image and improving the detection speed. BRIEF DESCRIPTION OF THE DRAWINGS

[0030] In order to more clearly illustrate the specific implementation methods of the present invention or the technical solutions in the prior art, the drawings required for use in the specific implementation methods or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are some implementation methods of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0031] Figure 1 is a flow chart of an X corner point identification method in an embodiment of the present invention;

[0032] Figure 2 It is a diagram of X corner points to be detected in an embodiment of the present invention. DETAILED DESCRIPTION

[0033] The technical solution of the present invention will be described clearly and completely below in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0034] In the description of the present invention, it should be noted that the terms "center", "upper", "lower", "left", "right", "vertical", "horizontal", "inner", "outer", etc., indicating the orientation or positional relationship, are based on the orientation or positional relationship shown in the drawings, and are only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore cannot be understood as limiting the present invention. In addition, the terms "first", "second", and "third" are used for descriptive purposes only, and cannot be understood as indicating or implying relative importance.

[0035] In the description of the present invention, it should be noted that, unless otherwise clearly specified and limited, the terms "installed", "connected", and "connected" should be understood in a broad sense, for example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be a direct connection, or it can be indirectly connected through an intermediate medium, it can also be the internal connection of two components, it can be a wireless connection, or it can be a wired connection. For ordinary technicians in this field, the specific meanings of the above terms in the present invention can be understood according to specific circumstances.

[0036] In addition, the technical features involved in the different embodiments of the present invention described below can be combined with each other as long as they do not conflict with each other.

[0037] like Figure 1 As shown, an embodiment of the present invention provides an X corner point recognition method, which is executed by an electronic device such as a computer or a server, and specifically includes:

[0038] S1, determine a number of spatially symmetrical pixels around the point to be detected. Figure 2 As shown, point O is the point to be detected, and the rectangular area is all pixel points. Taking points 1-40 as an example, several spatially symmetrical pixel points are determined, such as 8 pixel points, which can be pixel points 1, 6, 11, 16, 21, 26, 31, and 36.

[0039] S2, judging whether the plurality of pixels have high contrast characteristics according to the pixel values ​​of the plurality of pixels. If the plurality of pixels have high contrast characteristics, executing step S3, if the plurality of pixels do not have high contrast characteristics, executing step S5, i.e. judging that the point to be detected is not an X corner point.

[0040] S3, judging whether the plurality of pixels have symmetry characteristics according to the pixel values ​​of the plurality of pixels. If the plurality of pixels have symmetry characteristics, returning to step S1 to determine the plurality of spatially symmetrical pixels around the point to be detected, re-determining the plurality of spatially symmetrical pixels, and executing step S4; if the plurality of pixels do not have symmetry characteristics, executing step S5, i.e. determining that the point to be detected is not an X corner point.

[0041] If the eight pixel points 1, 6, 11, 16, 21, 26, 31, and 36 have high contrast characteristics, the symmetry characteristics can be further determined.

[0042] S4, determine whether all pixels within a certain range around the point to be detected have been tested for high contrast and symmetry characteristics. If so, execute step S6, that is, determine whether the point to be detected is an X corner point. If not, return to step S1 to continue testing other pixels. The range around the point to be detected can be defined as needed. Figure 2 For example, all the pixels within a certain range around the point to be detected are the pixels within the rectangular area around the pixel point O, or they may be some specific pixels within the rectangular area.

[0043] If pixel points 1, 6, 11, 16, 21, 26, 31, and 36 have both high contrast and symmetry characteristics, it is necessary to continue to measure based on other pixel points. For example, other spatially symmetrical pixel points are pixel points 2, 7, 12, 17, 22, 27, 32, and 37, pixel points 3, 8, 13, 18, 23, 28, 33, and 38, and pixel points 4, 9, 14, 19, 24, 29, 34, and 39. Pixel points 5, 10, 15, 20, 25, 30, 35, and 40 can be tested for S2-S3 respectively. If they all have contrast and symmetry characteristics, point O can be determined to be an X corner point.

[0044] This embodiment ensures that the point to be detected is accurately identified as an X corner point by step-by-step screening and verification. First, several spatially symmetrical pixels are determined around the point to be detected, and it is determined whether these pixels have high contrast characteristics. Then, it is further verified whether these pixels have symmetric characteristics. If both characteristics are met, other spatially symmetrical pixels are reselected to perform the same judgment process. Only when all selected pixels have high contrast and symmetric characteristics, the point to be detected is finally confirmed to be an X corner point. This method effectively improves the accuracy of X corner point recognition and reduces the possibility of misjudgment through multiple verification steps. At the same time, most of the pixels in an entire image that are not X corner points can be excluded in the initial step through a small amount of detection, thereby greatly reducing the amount of calculation for detecting an entire image and improving the detection speed.

[0045] Furthermore, when determining whether a plurality of pixel points have a high contrast characteristic according to the pixel values ​​of the plurality of pixel points in step S2, the method further includes:

[0046] If several pixels do not have a high contrast characteristic, or several pixels do not have a symmetric characteristic, it is determined that the point to be detected is not an X corner point, and the point to be detected is re-determined. When it is determined that the point to be detected is not an X corner point, X corner point detection will be performed again on other points to be detected, that is, the step of determining several spatially symmetrical pixels around the point to be detected is returned to re-determine several spatially symmetrical pixels.

[0047] In this embodiment, when judging whether a plurality of pixels have high contrast characteristics according to their pixel values, if it is found that these points do not have high contrast characteristics or symmetry characteristics, it is determined that the point to be detected is not an X corner point, and the point to be detected will be re-determined, and a plurality of spatially symmetrical pixel points will be re-selected around it. The function of this step is to ensure that the corner point finally detected not only meets the requirements in terms of contrast, but also must meet the symmetry conditions, thereby improving the accuracy of identifying the X corner point. Through each detection, a large number of pixel points that do not meet the X corner point characteristics in the entire image can be quickly eliminated, avoiding unnecessary subsequent judgments, thereby effectively reducing the image detection time. At the same time, only points that meet all the characteristics at the same time can be determined as X corner points, which significantly reduces the possibility of misjudging non-X corner points as X corner points. Therefore, while ensuring the accuracy of identifying X corner points, the present invention also effectively improves the timeliness of identifying X corner points.

[0048] In one embodiment, the plurality of pixels are uniformly distributed in an annular region centered on the point to be detected. The annular region may be a circular region or a rectangular region, etc., as long as it can reflect the local features of the point to be detected.

[0049] In this embodiment, a plurality of pixels are evenly distributed in a ring area with the point to be detected as the center, so that the local features around the point to be detected can be fully captured. By calculating the pixel values ​​of these pixels and deriving the contrast feature value, it is possible to effectively determine whether the point has high contrast characteristics and high symmetry, thereby accurately identifying the X corner point. This method improves the accuracy and robustness of corner point detection, and ensures that the contrast significance in the local area helps to accurately locate the feature point.

[0050] In one embodiment, determining a number of spatially symmetrical pixel points around the point to be detected in step S1 specifically includes:

[0051] S11, obtaining a preset distance. The preset distance is a radius R, that is, the distance between the point to be measured and the symmetrical pixel point.

[0052] S12, determine a number of pixel points according to the position of the point to be detected and the preset distance. Determine the area around the point to be detected according to the radius R, such as a circular area or a rectangular area, establish a rectangular coordinate system for the image, and select a number of evenly distributed pixel points in the determined area according to the coordinates (x, y) of the point to be detected and the radius R. For example, obtain 8 pixel points, whose coordinates are (xR, yR), (x, yR), (x+R, yR), (x+R, y), (x+R, y+R), (x, y+R), (xR, y+R), (xR, y+R), (xR, y, y+R), (xR, y).

[0053] For example, if R=10, we get a rectangular ring based on R, and then we get Figure 2 Pixel points 1-40 in the image are determined to be 8 evenly distributed pixels, namely pixel points 1, 6, 11, 16, 21, 26, 31, and 36.

[0054] In addition, in order to improve the detection accuracy and avoid misidentifying other points as X corner points, the radius of R can be reduced according to actual needs. Reducing the radius can make the detection area more limited to the vicinity of the point to be detected, thereby reducing the influence of background noise and other irrelevant points. Doing so can improve the accuracy of recognition, because the detection will only focus on local features closer to the point to be detected, reducing the risk of misidentification. Therefore, adjusting the size of the radius is an effective way to optimize the detection results, especially when there is a high requirement for detection accuracy.

[0055] In one embodiment, judging whether a plurality of pixels have a high contrast characteristic according to the pixel values ​​of the plurality of pixels in step S2 specifically includes:

[0056] S21, calculating contrast feature values ​​according to the pixel with the maximum grayscale value and the pixel with the minimum grayscale value among at least some of the plurality of pixels. For example, the pixel values ​​of all or some of the 8 pixels are obtained, and the maximum grayscale value gmax and the minimum grayscale value gmin can be obtained by sorting.

[0057] S22, determine whether the contrast feature value exceeds the contrast threshold. If the contrast feature value exceeds the contrast threshold, execute step S23, if the contrast feature value does not exceed the contrast threshold, determine that the point to be detected is not an X corner point, and perform detection of the next point to be detected.

[0058] S23, determining that a plurality of pixels have a high contrast characteristic.

[0059] The contrast characteristic value in step S21 is the difference between the maximum grayscale value and the minimum grayscale value.

[0060] For example, the difference gmax-gmin between the maximum grayscale value and the minimum grayscale value is compared with the contrast threshold L1. If gmax-gmin>L1, it is determined that several pixels have high contrast characteristics. If gmax-gmin<L1, it is determined that several pixels do not have high contrast characteristics, that is, it is determined that the point to be detected is not an X corner point. Return to step S1 to determine 8 pixels symmetrically around the point to be detected except for pixels 1, 6, 11, 16, 21, 26, 31, and 36, and loop through the method of steps S2-S4 until all pixels are traversed and all pixels determine that the point to be detected is a candidate X corner point, and finally determine that the point to be detected is an X corner point. Among them, when judging the high contrast characteristics, the judgment can be made based on 8 or 5.

[0061] This embodiment determines whether several pixel points have high contrast characteristics by calculating the contrast characteristic value of at least some pixel points, thereby improving the accuracy of X corner point detection. Specifically, step S21 calculates the difference between the maximum and minimum grayscale values ​​as the contrast characteristic value, and compares it with the set contrast threshold. If the characteristic value exceeds the threshold, it means that these points have high contrast and meet the characteristics of X corner points. Step S23 further confirms the high contrast characteristics of these points. Through this process, X corner points with significant contrast can be accurately identified, and the accuracy of corner point detection is improved. At the same time, by taking a small number of pixel points from several pixel points for high contrast characteristic judgment, the detection amount is reduced, and the detection speed can be improved.

[0062] In order to further reduce the amount of calculation and improve the recognition speed, at least part of the pixel points in step S21 are pixel points distributed in half of the annular area.

[0063] like Figure 2 As shown, for example, several pixels are 1, 6, 11, 16, 21, 26, 31, 36 pixels, and their gray values ​​are g1, g2, g3, g4, g5, g6, g7, g8 respectively. In this embodiment, some of the pixels, such as 5 pixels, are selected, and these pixels are distributed in a half circle in the annular area. If point O is an X corner point, the points in the half circle around it must include at least one pair of pixels with the largest contrast difference, such as point 1 and point 11, point 1 and point 31, etc., thereby reducing the amount of calculation and improving the recognition efficiency.

[0064] By taking at least some of the pixels among a number of pixels, including pixels with high contrast, it is not necessary to sort all of the pixels. Only at least some of the pixels can be used to determine whether they meet the high contrast characteristic, thereby improving the detection speed.

[0065] In one embodiment, judging whether the plurality of pixels have a symmetric characteristic according to the pixel values ​​of the plurality of pixels in step S3 specifically includes:

[0066] S31, at least some of the pixels among the plurality of pixels determine the pixel with the maximum gray value and its spatial symmetrical point, as well as the pixel with the minimum gray value and its spatial symmetrical point. When judging the high contrast characteristic, all of the plurality of pixels can be selected for judgment, or some of the plurality of pixels can be selected for judgment. The selection of some of the pixels can refer to step S2, which will not be repeated here.

[0067] For example, the pixel with the maximum gray value gmax is index_max (e.g. Figure 2 11 in the figure), the pixel with the minimum gray value gmin is index_min (e.g. Figure 2 1 in the example), among the 8 pixels, its spatial symmetric points are index_max+4R (31 in this example) and index_min+4R (21 in this example).

[0068] S32, calculating a first symmetrical eigenvalue according to the maximum grayscale value and the grayscale value of its spatially symmetrical point, and calculating a second symmetrical eigenvalue according to the minimum grayscale value and the grayscale value of its spatially symmetrical point.

[0069] S33, judging whether a plurality of pixel points have a symmetric characteristic according to the first symmetric eigenvalue and the second symmetric eigenvalue.

[0070] This embodiment determines whether these points have symmetry characteristics by analyzing the grayscale values ​​of several pixels in the image. Specifically, first identify the pixels with the maximum and minimum grayscale values ​​and their corresponding spatial symmetric points, then calculate the grayscale difference between these points to obtain the first and second symmetry eigenvalues, and finally, by comparing these eigenvalues ​​with the set threshold, determine whether the pixel points show symmetry. Through symmetry characteristic identification, some asymmetric interference points can be excluded and focus on points that truly show symmetry, thereby improving the accuracy of X corner point detection. This also means that if several pixels in the image show good symmetry, these points are more likely to constitute X corner points, thereby helping the machine vision system or image analysis algorithm to more reliably identify the feature points and corner points of the object.

[0071] The first symmetrical eigenvalue is the absolute value of the difference between the maximum grayscale value and the grayscale value of the spatially symmetrical point; the second symmetrical eigenvalue is the absolute value of the difference between the minimum grayscale value and the grayscale value of the spatially symmetrical point.

[0072] For example, the grayscale values ​​of two spatial symmetric points index_max+4R and index_min+4R are recorded as gmax' and gmin' respectively, the first symmetric eigenvalue is abs(gmax-gmax'), and the second symmetric eigenvalue is abs(gmin-gmin').

[0073] Furthermore, in step S33, judging whether a plurality of pixel points have a symmetric characteristic according to the first symmetric eigenvalue and the second symmetric eigenvalue specifically includes:

[0074] S331, determining whether the first symmetric eigenvalue and the second symmetric eigenvalue are lower than a symmetry threshold.

[0075] S332: If the first symmetric eigenvalue or the second symmetric eigenvalue is lower than the symmetry threshold, it is determined that a plurality of pixel points have a symmetric characteristic.

[0076] For example, the symmetry threshold is L2. If abs(gmax- gmax')<L2 or abs(gmin- gmin') <L2, the symmetry characteristic is satisfied, and it is determined that several pixel points have the symmetry characteristic. Otherwise, it is determined that several pixel points do not have the symmetry characteristic, and point O is directly determined not to be an X corner point.

[0077] This embodiment determines whether the pixels in the image have symmetry characteristics by comparing the symmetry eigenvalues ​​of the pixels with the symmetry threshold. If the difference between the eigenvalues ​​is lower than the threshold, the pixels are determined to be symmetric. This process plays an important role in corner point detection, and can effectively help distinguish areas with symmetry characteristics and improve the accuracy of X corner point analysis.

[0078] It will be appreciated by those skilled in the art that embodiments of the present invention may be provided as methods, systems, or computer program products. Therefore, the present invention may take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Furthermore, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0079] The present invention is described with reference to flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowchart and / or block diagram, as well as the combination of processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0080] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0081] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0082] Obviously, the above embodiments are merely examples for the purpose of clear explanation, and are not intended to limit the implementation methods. For those skilled in the art, other different forms of changes or modifications can be made based on the above description. It is not necessary and impossible to list all the implementation methods here. The obvious changes or modifications derived therefrom are still within the scope of protection of the invention.

Claims

1. A method for identifying X corner points, characterized in that: include: Determine a plurality of spatially symmetrical pixel points around the point to be detected, wherein the plurality of pixel points are a plurality of pixel points evenly distributed in a ring region centered on the point to be detected; Calculating a contrast characteristic value according to a pixel point having a maximum grayscale value and a pixel point having a minimum grayscale value among at least part of the plurality of pixel points, the contrast characteristic value being a difference between the maximum grayscale value and the minimum grayscale value; Determining whether the contrast characteristic value exceeds a contrast threshold; If the contrast characteristic value exceeds the contrast threshold, it is determined that the plurality of pixels have a high contrast characteristic; Determine, from at least some of the plurality of pixels, a pixel point with a maximum grayscale value and a spatially symmetrical point thereof, and a pixel point with a minimum grayscale value and a spatially symmetrical point thereof; A first symmetrical eigenvalue is calculated according to the maximum grayscale value and the grayscale value of its spatially symmetrical point, and a second symmetrical eigenvalue is calculated according to the minimum grayscale value and the grayscale value of its spatially symmetrical point, wherein the first symmetrical eigenvalue is the absolute value of the difference between the maximum grayscale value and the grayscale value of its spatially symmetrical point, and the second symmetrical eigenvalue is the absolute value of the difference between the minimum grayscale value and the grayscale value of its spatially symmetrical point; Determining whether the plurality of pixel points have a symmetric characteristic according to the first symmetric eigenvalue and the second symmetric eigenvalue; If the plurality of pixel points have symmetric characteristics, the step of determining a plurality of spatially symmetric pixel points around the point to be detected is returned to re-determine a plurality of other spatially symmetric pixel points until all the pixel points within a certain range around the point to be detected have high contrast characteristics and symmetric characteristics, and the point to be detected is determined to be an X corner point.

2. The method according to claim 1, characterized in that If the plurality of pixel points do not have a high contrast characteristic, or the plurality of pixel points do not have a symmetric characteristic, it is determined that the point to be detected is not an X corner point, and the point to be detected is re-determined.

3. The method according to claim 1, characterized in that Determine several spatially symmetrical pixel points around the point to be detected, including: Get the preset distance; A number of pixel points are determined according to the position of the point to be detected and the preset distance.

4. The method according to claim 1, characterized in that: Judging whether the plurality of pixel points have a symmetric characteristic according to the first symmetric eigenvalue and the second symmetric eigenvalue includes: Determining whether the first symmetric eigenvalue and the second symmetric eigenvalue are lower than a symmetry threshold; If the first symmetric eigenvalue or the second symmetric eigenvalue is lower than a symmetry threshold, it is determined that the plurality of pixels have a symmetric characteristic.

5. The method according to claim 1, characterized in that The at least part of the pixel points are pixel points distributed in a half area of ​​the annular area.

6. An X corner point recognition device, characterized in that: include: A processor and a memory connected to the processor; wherein the memory stores instructions executable by the processor, and the instructions are executed by the processor so that the processor executes the X corner point recognition method according to any one of claims 1 to 5.

Images