Marking positioning method and system for marking circle

Abstract

The present application provides a method and system for positioning a marker circle. The method can construct a circular segment based on a marker circle image and a marker circle finder control, and expand the circular segment containing the marker into a straight line segment image. Based on the first straight line segment image corresponding to the first marker circle image, the marker feature data and the first marker angle are obtained, and then the second straight line segment image corresponding to the second marker circle image is matched with the feature data to obtain the second marker angle. The difference between the first marker angle and the second marker angle is calculated to obtain the relative angle, thereby achieving marker positioning of the marker circle on the second marker circle image. The method is based on a marker circle positioning method for image processing, and adopts an algorithm for calculating the sum of the pixel values ​​of each column of the image to solve the problem of interference of strong light spots, background, etc. on image processing, thereby improving the marker positioning accuracy of the marker circle and having a wide range of applications.

Description

Technical Field

[0001] The present application relates to the field of image processing technology, and in particular to a marking positioning method and system for a marking circle. Background Art

[0002] With the rapid development of image sensors and vision algorithms, research on real-time positioning of products in industrial automation using image processing technology is increasing. In industries such as industry and manufacturing, the introduction of automatic positioning methods using machine vision, which select appropriate image processing methods based on the product's shape characteristics or the image's grayscale information, allows for positioning and alignment of products, achieving high-precision manufacturing processes and effectively ensuring product qualification rates.

[0003] For the marking positioning of products with marking circle shapes, such as flat wafers, grooved wafers, etc. A marking circle refers to a circle with a notch on it. This notch is called a mark and can be of various shapes. The positioning methods of the marking circle include contact and non-contact. Among them, the contact type is a mechanical clamping type. Since the clamping force is difficult to control, there is a risk of damaging the wafer. The non-contact type includes two methods. One is to use devices such as photoelectric sensors, linear array CCDs and laser displacement sensors to obtain wafer edge information, but the wafer needs to rotate one circle to detect the edge. Therefore, this method takes a long time to position. The second is to use visual imaging technology, use an industrial camera to obtain wafer images, and obtain the wafer position information after image processing. This method has a fast positioning time and high stability and reliability.

[0004] However, when using visual image technology for positioning, the collected image will be disturbed by factors such as strong light spots around the marker circle and interference of the image background on the marker circle outline, which will affect subsequent image processing, resulting in large deviations in the positioning results and reducing the accuracy of marker circle positioning. Summary of the Invention

[0005] The present application provides a marking positioning method and system for a marking circle to solve the problem of low marking positioning accuracy of the marking circle.

[0006] In a first aspect, the present application provides a marking positioning method for a marking circle, comprising:

[0007] Acquire a marker circle image, where the marker circle image includes a first marker circle image and a second marker circle image, where the first marker circle image and the second marker circle image are images of the marker circle to be located at different viewing angles;

[0008] Obtaining shape parameters of the marker circle to be located in the marker circle image based on a preset marker circle finding control;

[0009] Constructing a circular segment according to the shape parameters of the marker circle finding control and the shape parameters of the marker circle to be located in the marker circle image;

[0010] Expanding the circular segment image corresponding to the circular segment into a straight segment image, wherein the straight segment image includes a first straight segment image and a second straight segment image;

[0011] Acquire marker feature data and a first marker angle based on the first straight line segment image, where the first marker angle is the angle between the line connecting the center of the marker circle to be located and the marker center in the first marker circle image and the coordinate axis;

[0012] Extracting a data segment region having a similarity with the marker feature data greater than a similarity threshold from the second straight line segment image, and obtaining a second marker angle based on the data segment region, where the second marker angle is the angle between a line connecting the center of the marker circle to be located and the marker center in the second marker circle image and the coordinate axis;

[0013] The relative angle of the marker circle to be positioned in the marker circle image is calculated, where the relative angle is the difference between the first marker angle and the second marker angle.

[0014] In a second aspect, the present application provides a marker positioning system for a marker circle, the system being configured to:

[0015] Acquire a marker circle image, where the marker circle image includes a first marker circle image and a second marker circle image, where the first marker circle image and the second marker circle image are images of the marker circle to be located at different viewing angles;

[0016] Obtaining shape parameters of the marker circle to be located in the marker circle image based on a preset marker circle finding control;

[0017] Constructing a circular segment according to the shape parameters of the marker circle finding control and the shape parameters of the marker circle to be located in the marker circle image;

[0018] Expanding the circular segment image corresponding to the circular segment into a straight segment image, wherein the straight segment image includes a first straight segment image and a second straight segment image;

[0019] Acquire marker feature data and a first marker angle based on the first straight line segment image, where the first marker angle is the angle between the line connecting the center of the marker circle to be located and the marker center in the first marker circle image and the coordinate axis;

[0020] Extracting a data segment region having a similarity with the marker feature data greater than a similarity threshold from the second straight line segment image, and obtaining a second marker angle based on the data segment region, where the second marker angle is the angle between a line connecting the center of the marker circle to be located and the marker center in the second marker circle image and the coordinate axis;

[0021] The relative angle of the marker circle to be positioned in the marker circle image is calculated, where the relative angle is the difference between the first marker angle and the second marker angle.

[0022] As can be seen from the above technical solutions, the present application provides a marker positioning method and system for a marker circle, which can construct a circular segment based on a marker circle image and a marker circle finder control, and expand the circular segment containing the marker into a straight line segment image. Based on the first straight line segment image corresponding to the first marker circle image, the marker feature data and the first marker angle are obtained, and then the second straight line segment image corresponding to the second marker circle image is matched with the feature data to obtain the second marker angle, and the difference between the first marker angle and the second marker angle is calculated to obtain the relative angle, thereby realizing the marker positioning of the marker circle on the second marker circle image. The method is based on a marker circle positioning method of image processing, and adopts an algorithm for calculating the sum of the pixel values ​​of each column of the image to solve the problem of interference of strong light spots, background, etc. on image processing, thereby improving the marker positioning accuracy of the marker circle, and has a wide range of applications. BRIEF DESCRIPTION OF THE DRAWINGS

[0023] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without any creative work.

[0024] Figure 1 A schematic diagram of a marked circle provided in an embodiment of the present application;

[0025] Figure 2 A schematic flow chart of a marking positioning method for a marking circle provided in an embodiment of the present application;

[0026] Figure 3 A schematic diagram of marking angles provided in an embodiment of the present application;

[0027] Figure 4 A flowchart of the training part provided in the embodiment of the present application;

[0028] Figure 5 A schematic diagram of a find-marker-circle control provided in an embodiment of the present application on a marker-circle image;

[0029] Figure 6 A schematic diagram of the marking positions provided in an embodiment of the present application;

[0030] Figure 7 A schematic diagram of the position of the first circular ring segment provided in an embodiment of the present application;

[0031] Figure 8A schematic diagram of a process for expanding a first circular segment image corresponding to a first circular segment provided in an embodiment of the present application into a first straight line segment image;

[0032] Figure 9 A schematic diagram of a first circular ring segment provided in an embodiment of the present application;

[0033] Figure 10 Another schematic diagram of the first circular ring segment provided in an embodiment of the present application;

[0034] Figure 11 A schematic diagram of a first straight line segment image provided in an embodiment of the present application;

[0035] Figure 12 A schematic diagram of a process for obtaining marking feature data and a first marking angle provided in an embodiment of the present application;

[0036] Figure 13 A schematic diagram of first projection data provided in an embodiment of the present application;

[0037] Figure 14 A schematic diagram of the search process provided in the embodiment of the present application;

[0038] Figure 15 The second circular segment provided in the embodiment of the present application does not completely include the schematic diagram of the mark. DETAILED DESCRIPTION

[0039] The following embodiments are described in detail, with examples illustrated in the accompanying drawings. When the following description refers to the drawings, identical numbers in different figures represent identical or similar elements unless otherwise indicated. The embodiments described in the following embodiments are not intended to represent all possible implementations consistent with the present application. They are merely examples of systems and methods consistent with certain aspects of the present application, as detailed in the claims.

[0040] The embodiment of the present application proposes a marking circle positioning method based on image processing for products with a marking circle shape. Among them, the marking circle refers to a circle with a gap in the circle. This gap is called a mark and can be of various shapes. For example, Figure 1 The following are schematic diagrams of the marking circles provided in the embodiments of the present application, respectively for a flat wafer and a grooved wafer. The flat wafer has a flat marking, and the grooved wafer has a grooved marking. Using the marking circle positioning method, the flat marking and grooved marking are located on the flat wafer and grooved wafer, respectively.

[0041] Marker positioning of the marking circle involves two steps: center positioning and mark positioning. Positioning methods are categorized as contact and non-contact. The contact method involves mechanical clamping, but due to the difficulty in controlling the clamping force, there is a risk of wafer damage. The non-contact method includes two methods. One uses devices such as photoelectric sensors, linear array CCDs, and laser displacement sensors to obtain wafer edge information. However, this method requires the wafer to rotate one full rotation to detect the edge, resulting in a longer positioning time. The other method utilizes visual imaging technology, using an industrial camera to capture wafer images and then performing image processing to determine the wafer's position. This method offers fast positioning times and high stability and reliability.

[0042] In some embodiments, the arc segment containing the mark can be expanded into a straight line segment image, and then a caliper tool can be used to determine the longitudinal position of the mark. However, this method may have a large deviation in the mark positioning result when there is strong light spot interference around the mark or there are many transition pixels at the mark position.

[0043] In some embodiments, a deep learning semantic segmentation positioning method may also be used, but this method is only applicable to the positioning of marked circles with arc-shaped marks, and has a narrow scope of application.

[0044] When using visual image technology to locate the marker circle, the positioning results can be highly biased due to interference from strong light spots and a large number of transition pixels at the marker position. Furthermore, the method has a narrow scope of application and can only locate marker circles containing markers of a certain shape.

[0045] In order to improve the positioning accuracy, some embodiments of the present application also provide a marking positioning method for a marking circle, such as Figure 2 As shown, Figure 2 The present invention provides a flowchart of a method for positioning a marker circle according to an embodiment of the present invention. The method comprises the following steps:

[0046] Acquire the marked circle image. The marked circle image includes a first marked circle image and a second marked circle image, and the first marked circle image and the second marked circle image are images of the marked circle to be located at different viewing angles. For example, the marked circle to be located is Figure 1 The flat wafer shown can be photographed by a camera to obtain a first marked circle image, and then the flat wafer is rotated by a preset angle and photographed to obtain a second marked circle image.

[0047] It is understood that the positioning of the marker circle includes the positioning of the circle center and the positioning of the marker. The positioning result of the circle center is represented by the position of the circle center of the marker circle in the image coordinates, such as Figure 3As shown, the marker positioning result is expressed as the angle between the line connecting the marker center and the circle center and the horizontal coordinate axis, i.e., the marker angle. By performing marker positioning on the first and second marker circle images, the first marker angle of the marker in the marker circle to be positioned on the first marker circle image and the second marker angle on the second marker circle image are obtained. The difference between the two angles yields the relative angle of the marker on the second marker circle image relative to the first marker circle image. This is then combined with different viewing angles to achieve marker positioning of the marker circle.

[0048] The preset Find Marker Circle control is used to obtain the shape parameters of the marker circle to be located in the marker circle image. A circular segment is constructed based on the shape parameters of the Find Marker Circle control and the shape parameters of the marker circle to be located in the marker circle image. The corresponding circular segment image is then expanded into a straight segment image.

[0049] The straight line segment image includes a first straight line segment image and a second straight line segment image. It is understood that the first straight line segment image is a straight line segment image obtained based on the first marked circle image according to the above steps, and the second straight line segment image is a straight line segment image obtained based on the second marked circle image according to the above steps.

[0050] The marker positioning method for a marker circle provided in this application includes two parts: training and searching. The purpose of the training part is to obtain the marker feature data and marker angle of the marker circle to be located on the first marker circle image, and provide them to the search part for matching. The purpose of the search part is to search for the area with the highest similarity to the marker feature data on the second marker circle image, and then obtain the relative angle and marker angle of the marker circle to be located on the second marker circle image, where the relative angle is the difference between the first marker angle on the second marker circle image and the second marker angle on the first marker circle image.

[0051] like Figure 4 As shown, Figure 4 This is a flow chart of the training portion provided in an embodiment of the present application. The shape parameters of the marker circle to be located in the first marker circle image can be obtained based on a preset marker circle finder control. The shape parameters include the coordinates of the circle center and the radius.

[0052] like Figure 5 、 Figure 6 As shown, Figure 5 This is a schematic diagram of the find marker circle control provided in an embodiment of the present application on the marker circle image. Figure 6A schematic diagram of the marking position provided for an embodiment of the present application. The marking circle finding control includes a marking caliper and a marking area. By starting the marking circle finding control, the marking circle finding control is dragged to the position of the marking circle to be located in the first marking circle image, and the marking area is dragged to the marking position of the marking circle to be located in the first marking circle image. And the polarity of the caliper, the search length and other information are set according to the first marking circle image. Thus, the position data of the marking circle finding control on the first marking circle image is obtained. Then, based on the preset circle finding tool, the marking circle to be located in the first marking circle image is fitted to obtain the shape parameters of the marking circle to be located in the first marking circle image, that is, the center coordinate of the marking circle to be located in the first marking circle image is obtained as (x c1 ,y c1 ), the radius is denoted as r1.

[0053] Then, the first circular segment is constructed based on the shape parameters of the marker circle finder control and the shape parameters of the marker circle to be located in the first marker circle image. The starting angle and span of the first circular segment are the same as those of the marker circle finder control, and are denoted as Start and Span respectively. The center of the first circular segment is the center of the marker circle to be located, that is, the center coordinates of the marker circle to be located are set to the center coordinates of the first circular segment (x c1 ,y c1 ).

[0054] The ring width of the first ring segment is an integer obtained by rounding up the search length of the mark caliper in the mark circle finder control or twice the preset mark depth. That is to say, the ring width of the first ring segment includes two setting methods. The first setting method is to set the ring width according to the search length of the caliper in the mark circle finder control. In order to facilitate the subsequent calculation of the number of sampling steps, the integer obtained by rounding up the search length can be set as the ring segment width, that is, the ring segment width is d′=ceil(d), where d is the search length and ceil(d) represents a function rounded up to an integer. The second setting method is to pre-set the mark depth d w , where d w is a positive integer. Then the width of the ring is d′=2d w .

[0055] The outer radius of the first annular segment is the sum of the radius of the mark circle to be located and half the annular width. The radial scaling ratio of the first annular segment is calculated based on the annular width and the outer radius. The radial scaling ratio can be calculated by calculating a difference, where the difference is the difference between the radius of the mark circle to be located and half the annular width. If the difference is less than or equal to zero, the radial scaling ratio is zero. If the difference is greater than zero, the radial scaling ratio is the ratio of the difference to the outer radius.

[0056] For example, the outer circle radius and radial scaling ratio can be calculated as follows:

[0057] r 外 =r1+d′ / 2;

[0058]

[0059] Wherein, r1 is the radius of the marker circle to be located in the first marker circle image, d′ is the ring width, rout is the outer circle radius, and s is the radial scaling ratio of the outer circle radius.

[0060] like Figure 7 As shown, Figure 7 A schematic diagram of the position of the first circular segment provided in an embodiment of the present application. According to the constructed first circular segment, the circular segment image corresponding to the first circular segment on the first marked circle image is expanded into a first straight line segment image.

[0061] like Figure 8 As shown, Figure 8 A flow chart of expanding a first circular segment image corresponding to a first circular segment provided in an embodiment of the present application into a first straight segment image. First, the circular segment parameters of the first circular segment are obtained, and the circular segment parameters include circular width, outer circle radius, radial scaling ratio, starting angle, and opening angle. Then, the first straight segment parameters of the first straight segment image are calculated. Among them, the first straight segment parameters include straight line height and straight line width. The straight line height is calculated based on the outer circle radius, circular width, and radial scaling ratio, and the straight line width is calculated based on the radius of the mark circle to be positioned and the opening angle of the first circular segment.

[0062] For the straight line height, the radial scaling ratio of the first circular segment can be obtained. If the radial scaling ratio is greater than zero, the corresponding first circular segment is as follows Figure 9 As shown, Figure 9 The dotted line in the figure represents the arc where the circle to be positioned is located. The height of the straight line is L y1 is the ring width d′. If the radial scaling factor is zero, the corresponding first ring segment is as follows Figure 10 As shown, Figure 10 The dotted line in the figure represents the arc where the circle to be positioned is located. The height of the straight line is the integer obtained by rounding up the radius of the outer circle, that is, L y1 =ceil(r 外 )=ceil(r1+d′ / 2).

[0063] For the straight line width, when the first circular segment is expanded into a straight line segment, it is expanded based on the length of the dotted arc. That is, the length of the straight line segment is approximately equal to the length of the arc where the marker circle to be positioned in the first circular segment is located. Then the straight line width of the straight line segment image is L x1 =ceil(r1·|Span|).

[0064] When the first circular segment image is expanded into the first straight segment image, the mapping relationship between the first circular segment image and the first straight segment image is calculated based on the first circular segment parameters and the first straight segment parameters. Based on the mapping relationship, the pixels of the first circular segment image are assigned to the first straight segment image. For example, the coordinates of the first straight segment image can be traversed and the pixels of the first straight segment image can be assigned based on the bilinear interpolation method. In this way, Figure 7 The first circular segment image corresponding to the first circular segment shown is expanded into Figure 11 The first straight line segment image is shown.

[0065] After acquiring the first straight line segment image, the marker feature data and the first marker angle can be acquired based on the first straight line segment image. The first marker angle is the angle between the line connecting the center of the marker circle to be located and the marker center in the first marker circle image and the coordinate axis.

[0066] like Figure 12 As shown, Figure 12 This is a schematic diagram of a process for obtaining marker feature data and a first marker angle according to an embodiment of the present application. The marker feature data and the marker center coordinates of the marker circle to be located in the first marker circle image can be extracted from the first straight line segment image, and the marker center coordinates can be converted to the first circular segment image to obtain the first marker angle.

[0067] For the marking feature data, the first projection number data can be calculated. The first projection number data includes the sum of the pixel values ​​in the column direction of the first straight line segment image, that is, the sum of the pixel values ​​in each column. By calculating the sum of the pixel values ​​in each column of the first straight line segment image, that is, the projection number, the marking feature data corresponding to the marking area (the mark of the marking circle to be located) can be clearly distinguished, that is, Figure 13 The data within the dotted line portion of the first projection data is shown.

[0068] By calculating the starting position and the ending position of the marking area in the first projection digital data, the marking feature data is intercepted from the first projection digital data, wherein the marking feature data is the coordinate data between the starting position and the ending position.

[0069] For example, Figure 13 Taking the first projection data as an example, since the input marking area data are all relative arcs relative to the starting angle of the circular segment, this data is converted to the vertical position x on the first straight line segment image. s1 and x E1 , i.e. the starting position and the ending position. In the first projection data, a segment [x s1 , x g1 ] is the marking feature data P of the marking circle to be located.

[0070] For the marker center coordinates, the marker width of the marker feature data can be calculated, and the marker center coordinates can be calculated based on the marker width. The marker width is the sum of the absolute value of the difference between the start position and the end position and a preset value, and the marker center coordinates are half the marker width.

[0071] For example, in the above example, the number of data of the marking feature data P is m=|x S1 -x E1 |+1, that is, the mark width is m. Then the X-axis coordinate of the mark center is m c =m / 2, which is the mark center coordinate.

[0072] In some embodiments, the marking angle can also be corrected. According to the marking feature data, the peak point of the marking feature data is relatively close to the center of the marking area. Therefore, the center position of the marking feature data can be obtained. The center position is the coordinate value corresponding to the maximum or minimum value in the marking feature data. If there are multiple center positions, the average coordinate value of the center position is calculated to obtain the marking width. The maximum or minimum value in the marking feature data can be determined based on the caliper polarity of the marking circle control. For example, taking the above example, if there are multiple maximum or minimum values ​​in the marking feature data P, the average value of the subscripts of the multiple maximum or minimum values ​​is calculated to obtain m. c .

[0073] That is, if the marking angle is not corrected, the marking center coordinate is half the marking width; if the marking angle is corrected, the marking center coordinate is the coordinate value corresponding to the maximum or minimum value in the marking feature data.

[0074] After obtaining the marker center coordinates, the marker center coordinates are converted to the first circular segment image according to the mapping relationship between the first circular segment image and the first straight line segment image to obtain the first marker angle. For example, the X-axis coordinates of the marker center are converted to the first circular segment image according to the following formula to obtain the first marker angle θ S1 :

[0075]

[0076] Among them, Start is the starting angle of the first circular segment, Span is the span angle of the first circular segment, x S1 is the starting position, m c is the coordinate of the marker center, L x1 The line width of the line segment image.

[0077] like Figure 14 As shown, Figure 14A schematic diagram of the search process provided in an embodiment of the present application. The search process can obtain shape parameters of the marker circle to be located in the second marker circle image based on a preset find marker circle control, construct a second circular segment based on the shape parameters of the find marker circle control and the shape parameters of the marker circle to be located in the second marker circle image, and expand the second circular segment image corresponding to the second circular segment into a second straight line segment image.

[0078] It is understood that the first circular segment is constructed based on the shape parameters of the Find Marker Circle control and the shape parameters of the marker circle to be located in the first marker circle image. The second circular segment is constructed based on the shape parameters of the Find Marker Circle control and the shape parameters of the marker circle to be located in the second marker circle image. The steps for expanding the second circular segment into the second straight line segment image can be referred to the steps for expanding the first circular segment into the first straight line segment image and are not further described here.

[0079] In some embodiments, if the second circular segment is constructed according to the steps of constructing the first circular segment in the above training part, the starting angle and opening angle of the second circular segment are adjusted according to the starting angle and opening angle of the marker circle control in the above training part, the following situations may occur, such as Figure 15 As shown, the second circular segment fails to completely contain the mark, and the expanded second straight line segment image may have two incomplete marks, which may cause deviation in the search results.

[0080] Therefore, in order to ensure that the mark is completely included when the second circular segment is unfolded, the starting angle and opening angle of the second circular segment need to be adjusted. The marking arc of the mark width in the above training part can be calculated. Get the starting angle and opening angle of the first circular segment, and adjust the starting angle and opening angle of the second circular segment according to the marking arc, the starting angle and opening angle of the first circular segment. Among them, the starting angle of the second circular segment is the difference between the starting angle of the first circular segment and the marking arc, and the opening angle of the second circular segment is the sum of the opening angle of the first circular segment and twice the marking arc.

[0081] For example, the start angle start2 and span angle span2 of the adjusted second circular segment can be calculated according to the following formula:

[0082] Start2=Start-rad

[0083] Span2=Span+2*rad

[0084] Among them, rad represents the radian corresponding to the mark width m of the above training part. The starting angle of the adjusted second circular segment is start2, the span angle is span2, and the width of the second straight line segment image after the second circular segment is expanded is L x2 , height is L y2 .

[0085] After acquiring the second line segment image, data segments with a similarity greater than a similarity threshold to the marker feature data can be extracted from the second line segment image. A second marker angle can be obtained based on the data segment regions. Finally, the relative angle of the marker circle to be located in the marker circle image is calculated. The second marker angle is the angle between the line connecting the center of the marker circle to be located and the marker center in the second marker circle image and the coordinate axis. The relative angle is the difference between the first marker angle and the second marker angle.

[0086] For the second marker angle, the second projection data can be calculated using the same steps as those used to calculate the first projection data in the training section. The second projection data includes the sum of the pixel values ​​along the column direction of the second straight segment image. The data segment area and similarity score of the second projection data with the most similarity to the marker feature data are then calculated. A parabola equation is constructed based on the data segment area and similarity score. The abscissa at the maximum value of the parabola equation is calculated. This abscissa is then converted to the second circular segment image to obtain the second marker angle.

[0087] For example, the ZNCC algorithm can be used to calculate the similarity score between the second projection data Q and the marker feature data P:

[0088]

[0089] Wherein, j represents the start from the jth element in the second projection data Q, and 0≤j <n-m+1;S j is the j-th similarity score, ranging from [0, 1]; is the average value of array P; Represents the average value of the elements from j to j+m-1 in the array Q.

[0090] By traversing the elements in the second projection data Q, the similarity score between each data segment in the second projection data Q and the marker feature data P is calculated to obtain the similarity score S with the highest similarity. max And the starting position j of the corresponding data segment area.

[0091] Based on the parabolic interpolation method, the parabolic equation is constructed as:

[0092] S=a·j 2 +b·j+c

[0093] Select the point with the highest similarity score, the previous point, and the next point, and calculate the horizontal coordinate corresponding to the maximum value of this equation according to the following formula: This is the point obtained by interpolation.

[0094] Then, the horizontal coordinate is converted to the second circular segment image according to the following formula to obtain the second marking angle θ j′ :

[0095]

[0096] Then the relative angle of the search mark is θ = θ j′ -θ S1 . Among them, θ i′ is the second marking angle, θ S1 is the first marking angle.

[0097] In this embodiment, a marker positioning method based on image processing adopts a method of calculating the sum of pixel values ​​of each column of the image to train the markers of the training image, and matches the trained markers on the real-time image through the ZNCC method to achieve precise positioning of the markers of the marker circle on the real-time image, solve the problem of the influence of strong light spot interference or poor imaging effect of the marker position on image processing, improve positioning accuracy, and have a wide range of applications.

[0098] Based on the above-mentioned marker positioning method of the marker circle, some embodiments of the present application also provide a marker positioning system for a marker circle, wherein the system is configured as follows:

[0099] Acquire a marker circle image, where the marker circle image includes a first marker circle image and a second marker circle image, where the first marker circle image and the second marker circle image are images of the marker circle to be located at different viewing angles;

[0100] Obtaining shape parameters of the marker circle to be located in the marker circle image based on a preset marker circle finding control;

[0101] Constructing a circular segment according to the shape parameters of the marker circle finding control and the shape parameters of the marker circle to be located in the marker circle image;

[0102] Expanding the circular segment image corresponding to the circular segment into a straight segment image, wherein the straight segment image includes a first straight segment image and a second straight segment image;

[0103] Acquire marker feature data and a first marker angle based on the first straight line segment image, where the first marker angle is the angle between the line connecting the center of the marker circle to be located and the marker center in the first marker circle image and the coordinate axis;

[0104] Extracting a data segment region having a similarity with the marker feature data greater than a similarity threshold from the second straight line segment image, and obtaining a second marker angle based on the data segment region, where the second marker angle is the angle between a line connecting the center of the marker circle to be located and the marker center in the second marker circle image and the coordinate axis;

[0105] The relative angle of the marker circle to be positioned in the marker circle image is calculated, where the relative angle is the difference between the first marker angle and the second marker angle.

[0106] Similar parts between the embodiments provided in this application can be referenced to each other. The specific implementation methods provided above are only a few examples under the overall concept of this application and do not constitute a limitation on the scope of protection of this application. For those skilled in the art, any other implementation methods expanded based on the scheme of this application without expending creative work shall fall within the scope of protection of this application.

Claims

1. A marking positioning method for a marking circle, characterized in that: include: Acquire a marker circle image, where the marker circle image includes a first marker circle image and a second marker circle image, where the first marker circle image and the second marker circle image are images of the marker circle to be located at different viewing angles; Obtaining shape parameters of the marker circle to be located in the marker circle image based on a preset marker circle finding control; Constructing a circular segment according to the shape parameters of the marker circle finding control and the shape parameters of the marker circle to be located in the marker circle image; Expanding the circular segment image corresponding to the circular segment into a straight segment image, wherein the straight segment image includes a first straight segment image and a second straight segment image; Acquire marker feature data and a first marker angle based on the first straight line segment image, where the first marker angle is the angle between the line connecting the center of the marker circle to be located and the marker center in the first marker circle image and the coordinate axis; Extracting a data segment region having a similarity with the marker feature data greater than a similarity threshold from the second straight line segment image, and obtaining a second marker angle based on the data segment region, where the second marker angle is the angle between a line connecting the center of the marker circle to be located and the marker center in the second marker circle image and the coordinate axis; The relative angle of the marker circle to be positioned in the marker circle image is calculated, where the relative angle is the difference between the first marker angle and the second marker angle.

2. The marking positioning method of the marking circle according to claim 1, characterized in that: The method further comprises: Start a mark circle finder control, wherein the mark circle finder control includes a mark caliper and a mark area; Drag the find mark circle control to the position of the mark circle to be located in the first mark circle image, and drag the mark area to the mark position of the mark circle to be located in the first mark circle image; Fitting the marker circle to be located in the first marker circle image to obtain shape parameters of the marker circle to be located in the first marker circle image, wherein the shape parameters include the coordinates of the center of the circle and the radius; Constructing a first circular segment according to shape parameters of the marker circle finding control and shape parameters of the marker circle to be located in the first marker circle image; Expanding the first circular segment image corresponding to the first circular segment into a first straight line segment image; Extracting marker feature data and marker center coordinates of the marker circle to be located in the first marker circle image from the first straight line segment image; The marker center coordinates are converted to the first circular segment image to obtain a first marker angle.

3. The marking positioning method of the marking circle according to claim 2, characterized in that: The step of expanding the first circular segment image corresponding to the first circular segment into a first straight line segment image includes: Obtain first circular segment parameters of the first circular segment, where the first circular segment parameters include a circular width, an outer radius, a radial scaling ratio, a starting angle, and an opening angle. The circular width is an integer obtained by rounding up the search length of the marking caliper or twice the preset marking depth. The outer radius is the sum of the radius of the mark circle to be located and half the circular width. The radial scaling ratio is calculated based on the circular width and the outer radius. The starting angle and the opening angle are the same as the starting angle and the opening angle of the find mark circle control. Calculating first straight line segment parameters of the first straight line segment image, where the first straight line segment parameters include a straight line height and a straight line width, where the straight line height is calculated based on the outer circle radius, the annular width, and the radial scaling ratio, and the straight line width is calculated based on the radius of the mark circle to be positioned and the opening angle; Calculating a mapping relationship between the first circular segment image and the first straight segment image according to the first circular segment parameter and the first straight segment parameter; The pixels of the first circular segment image are assigned to the first straight line segment image based on the mapping relationship.

4. The marking positioning method of the marking circle according to claim 3, characterized in that: The method further comprises: Calculating a difference, where the difference is the difference between the radius of the circle of the mark to be positioned and half the width of the circle; If the difference is less than or equal to zero, the radial scaling factor is zero; If the difference is greater than zero, the radial scaling ratio is the ratio of the difference to the outer circle radius.

5. The marking positioning method of a marking circle according to claim 3, characterized in that: The method further comprises: Obtaining the radial scaling ratio; If the radial scaling ratio is greater than zero, the straight line height is the ring width; If the radial scaling factor is zero, the straight line height is an integer obtained by rounding up the outer circle radius.

6. The marking and positioning method of a marking circle according to claim 2, characterized in that: The step of extracting the marker feature data and the marker center coordinates of the marker circle to be located in the first marker circle image from the first straight line segment image comprises: Calculating first projection number data, where the first projection number data includes the sum of pixel values ​​in a column direction of the first straight line segment image; Calculating a starting position and an ending position of the marked area in the first projection data; intercepting the marking feature data from the first projection data, wherein the marking feature data is coordinate data between the starting position and the ending position; Calculating a mark width of the mark feature data, where the mark width is the sum of an absolute value of a difference between the start position and the end position and a preset value; The mark center coordinates are calculated according to the mark width, and the mark center coordinates are half of the mark width.

7. The marking positioning method of a marking circle according to claim 6, characterized in that: The method further comprises: Obtaining the center position of the mark feature data, where the center position is the coordinate value corresponding to the maximum value or the minimum value in the mark feature data; The average value of the coordinates of the center position is calculated to obtain the center coordinates of the marker.

8. The marking positioning method of a marking circle according to claim 6, characterized in that: The method further comprises: Calculating the marking arc of the marking width; Obtaining a starting angle and an opening angle of a first circular segment, where the first circular segment is a circular segment constructed according to shape parameters of the marker circle finder control and shape parameters of the marker circle to be located in the first marker circle image; The starting angle and the opening angle of the second circular segment are adjusted according to the marking arc, the starting angle and the opening angle of the first circular segment. The second circular segment is a circular segment constructed according to the shape parameters of the marking circle finding control and the shape parameters of the marking circle to be located in the second marking circle image. The starting angle of the second circular segment is the difference between the starting angle of the first circular segment and the marking arc, and the opening angle of the second circular segment is the sum of the opening angle of the first circular segment and twice the marking arc.

9. The marking positioning method of a marking circle according to claim 8, characterized in that: The method further comprises: Calculating second projection number data, where the second projection number data includes the sum of pixel values ​​in a column direction of the second straight line segment image; Calculating a data segment area and a similarity score having the highest similarity with the marker feature data in the second projection data; constructing a parabola equation based on the data segment area and the similarity score; Calculate the abscissa of the maximum value of the parabola equation; The horizontal coordinate is converted to the second circular segment image to obtain a second marking angle.

10. A marking positioning system for a marking circle, characterized in that: The system is configured to: Acquire a marker circle image, where the marker circle image includes a first marker circle image and a second marker circle image, where the first marker circle image and the second marker circle image are images of the marker circle to be located at different viewing angles; Obtaining shape parameters of the marker circle to be located in the marker circle image based on a preset marker circle finding control; Constructing a circular segment according to the shape parameters of the marker circle finding control and the shape parameters of the marker circle to be located in the marker circle image; Expanding the circular segment image corresponding to the circular segment into a straight segment image, wherein the straight segment image includes a first straight segment image and a second straight segment image; Acquire marker feature data and a first marker angle based on the first straight line segment image, where the first marker angle is the angle between the line connecting the center of the marker circle to be located and the marker center in the first marker circle image and the coordinate axis; Extracting a data segment region having a similarity with the marker feature data greater than a similarity threshold from the second straight line segment image, and obtaining a second marker angle based on the data segment region, where the second marker angle is the angle between a line connecting the center of the marker circle to be located and the marker center in the second marker circle image and the coordinate axis; The relative angle of the marker circle to be positioned in the marker circle image is calculated, where the relative angle is the difference between the first marker angle and the second marker angle.

Images