WO2022161172A1 - 图像中图形角点的识别方法、装置、介质及电子设备 - Google Patents

图像中图形角点的识别方法、装置、介质及电子设备 Download PDF

Info

Publication number
WO2022161172A1
WO2022161172A1 PCT/CN2022/071616 CN2022071616W WO2022161172A1 WO 2022161172 A1 WO2022161172 A1 WO 2022161172A1 CN 2022071616 W CN2022071616 W CN 2022071616W WO 2022161172 A1 WO2022161172 A1 WO 2022161172A1
Authority
WO
WIPO (PCT)
Prior art keywords
corner
points
point
boundary
processed
Prior art date
Application number
PCT/CN2022/071616
Other languages
English (en)
French (fr)
Inventor
贾坤
王霖
唐泽达
赵振宇
李屹
Original Assignee
深圳光峰科技股份有限公司
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by 深圳光峰科技股份有限公司 filed Critical 深圳光峰科技股份有限公司
Publication of WO2022161172A1 publication Critical patent/WO2022161172A1/zh

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/60—Analysis of geometric attributes
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/10—Segmentation; Edge detection
    • G06T7/13—Edge detection

Definitions

  • the present application relates to the technical field of image processing, and in particular, to a method, device, medium and electronic device for identifying corner points of graphics in an image.
  • the embodiments of the present application provide a method, device, medium and electronic device for identifying corner points of graphics in an image, so that the test graphics and the corner positions of the screen can be quickly identified at least to a certain extent to determine the positions of the test graphics and the screen , to ensure the accuracy of corner position recognition.
  • a method for identifying corner points of graphics in an image comprising:
  • the position information of the boundary point determine the angle formed by each of the three adjacent boundary points, wherein the boundary point at the middle position is the vertex of the angle;
  • the near-corner point to be processed is identified from the included angle vertices of each of the included angles;
  • a target near-corner point is identified from the near-corner points to be processed, so as to determine the corner point of the to-be-identified figure according to the target near-corner point.
  • the location information includes coordinate information and an adjacent relationship; image recognition is performed on an image containing a pattern to be recognized, and the location information of the boundary point corresponding to the pattern to be recognized is determined, Including: taking the center point of the image as the coordinate origin, identifying the boundary points of the to-be-recognized graphics in a clockwise or counterclockwise order, and determining the coordinate information of the boundary points corresponding to the to-be-identified graphics; according to the boundary points The sequence of identification is determined, and the adjacent relationship of the boundary points is determined.
  • identifying a target near-corner point from the to-be-processed near-corner points according to the number of adjacent near-corner points to be processed includes:
  • the near-corner point to be processed has no other adjacent near-corner points to be processed, compare the included angles of the two adjacent boundary points as the included angle vertices, and determine the near-corner point to be processed and its adjacent near-corner point.
  • the included angle vertex with the larger included angle is the target near-corner point.
  • identifying a target near-corner point from the to-be-processed near-corner points according to the number of adjacent near-corner points to be processed includes:
  • the to-be-processed near-corner point has other adjacent to-be-processed near-corner points, and the to-be-processed near-corner point exists continuously, wherein the number of the continuous to-be-processed near-corner points is two, the two consecutive near-corner points to be processed The near-corner point to be processed is identified as the target near-corner point.
  • identifying a target near-corner point from the to-be-processed near-corner points according to the number of adjacent near-corner points to be processed includes:
  • the to-be-processed near-corner point has other adjacent to-be-processed near-corner points, the to-be-processed near-corner points exist continuously, wherein the number of the continuous near-corner points to be processed is greater than or equal to three, then the continuous near-corner points to be processed.
  • the to-be-processed near-corner points other than both ends of the to-be-processed near-corner points are deleted, so as to perform target near-corner point identification according to the deleted boundary point information.
  • identifying the near-corner point to be processed from the included angle vertices of each of the included angles includes:
  • the included angle vertex of the included angle is identified as the near-corner point to be processed
  • the number of the target near-corner points is multiple, and the determining the corner points of the to-be-recognized graphic according to the target near-corner points includes:
  • the corner points of the to-be-recognized figure are determined.
  • determining the corner points of the to-be-recognized graphic includes:
  • each of the target near-corner point combinations and each of the target near-corner points in a clockwise or counterclockwise order, and determine the first number corresponding to each target near-corner point combination and the first number corresponding to each of the target near-corner points. number two;
  • image recognition is performed on an image containing a graphic to be identified, and boundary point information corresponding to the graphic to be identified is determined, including:
  • the boundary points of the to-be-recognized figure are identified in a clockwise or counterclockwise order to determine the boundary point information corresponding to the to-be-recognized figure.
  • the number of the target near-corner points is multiple, and the determining the corner points of the to-be-recognized pattern according to the target near-corner points includes: according to the target near-corner points The position information of the points is used to determine the near corner points of the target located on the same boundary; the corner points of the to-be-recognized graphic are determined according to the near corner points of the target located on the same boundary.
  • a device for identifying corner points of graphics in an image comprising:
  • the image recognition module is used to perform image recognition on the image containing the figure to be recognized, and determine the boundary point information corresponding to the figure to be recognized, and the boundary point information includes the coordinate information of the boundary point and the adjacent boundary points between the boundary points. relation;
  • a calculation module for determining the included angle formed by every adjacent three boundary points according to the coordinate information of the boundary point and the adjacent relationship, wherein, the boundary point positioned at the middle position is the angle vertex;
  • a near-corner point identification module used for identifying the near-corner point to be processed from the included angle vertices of each of the included angles according to the included angle formed by each of the three adjacent boundary points;
  • a corner point determination module is used to identify a target near corner point from the near corner points to be processed according to the number of adjacent near corner points to be processed, so as to determine the corner points of the to-be-recognized graphics according to the target near corner points .
  • a computer-readable medium on which a computer program is stored, and when the computer program is executed by a processor, realizes the recognition of the corner points of the graphics in the image as described in the above-mentioned embodiments method.
  • an electronic device including: one or more processors; and a storage device for storing one or more programs, when the one or more programs are stored by the one or more programs When executed by multiple processors, the one or more processors are made to implement the method for identifying corner points of graphics in an image as described in the above embodiments.
  • the near-corner points to be processed are identified in advance, and the near-corner points to be processed are screened according to the number of adjacent near-corner points to be processed, and the wrongly identified near-corner points to be processed are removed.
  • identify the target near-corner point from the near-corner points to be processed to ensure the accuracy of the target near-corner point recognition, and then determine the corner point of the to-be-recognized pattern according to the identified target near-corner point. Therefore, no complicated steps are required in the process of identifying the position of the corner point, which can not only improve the efficiency of identifying the position of the corner point, but also ensure the accuracy of identifying the position of the corner point.
  • Fig. 1 shows a schematic diagram of an exemplary system architecture to which the technical solutions of the embodiments of the present application can be applied.
  • FIG. 2 shows a schematic flowchart of a method for identifying corner points of graphics in an image according to an embodiment of the present application.
  • FIG. 3 shows a schematic flowchart of determining corner points in a method for identifying corner points of graphics in an image according to an embodiment of the present application.
  • FIG. 4 shows a schematic flowchart of step S320 in the method for identifying corner points of graphics in the image of FIG. 3 according to an embodiment of the present application.
  • Figures 5a to 5c show schematic flowcharts of a method for identifying corner points of graphics in an image according to an embodiment of the present application.
  • FIG. 6 shows a flowchart of determining a corner point in a method for identifying a corner point of a graphic in an image according to an embodiment of the present application.
  • FIG. 7 shows a block diagram of an apparatus for identifying corner points of graphics in an image according to an embodiment of the present application.
  • FIG. 8 shows a schematic structural diagram of a computer system suitable for implementing the electronic device according to the embodiment of the present application.
  • Example embodiments will now be described more fully with reference to the accompanying drawings.
  • Example embodiments can be embodied in various forms and should not be construed as limited to the examples set forth herein; rather, these embodiments are provided so that this application will be thorough and complete, and will fully convey the concept of example embodiments to those skilled in the art.
  • FIG. 1 shows a schematic diagram of an exemplary system architecture to which the technical solutions of the embodiments of the present application can be applied.
  • the system architecture may include terminal devices (one or more of a smart phone 101 , a tablet computer 102 and a portable computer 103 as shown in FIG. 1 , of course, a desktop computer, etc.), a network 104 and server 105.
  • the network 104 is the medium used to provide the communication link between the terminal device and the server 105 .
  • the network 104 may include various connection types, such as wired communication links, wireless communication links, and the like.
  • the numbers of terminal devices, networks and servers in FIG. 1 are merely illustrative. There can be any number of terminal devices, networks and servers according to implementation needs.
  • the server 105 may be a server cluster composed of multiple servers, or the like.
  • the user can use the terminal device to interact with the server 105 through the network 104 to receive or send messages and the like.
  • the server 105 may be a server that provides various services.
  • the user can use the image acquisition device (such as a camera, etc.) configured by the terminal device 103 (it may also be the terminal device 101 or 102) to acquire the image containing the graphic to be recognized, and upload it to the server 105.
  • Perform image recognition on the image of the recognized figure determine the position information of the boundary point corresponding to the figure to be recognized, and determine the included angle formed by each of the three adjacent boundary points according to the position information of the boundary point, wherein the middle position is located in the middle position.
  • the boundary point is the angle vertex.
  • the near-corner point to be processed is identified from the angle vertices of each included angle, and then according to the number of adjacent near-corner points to be processed, from The target near-corner point is identified from the near-corner points to be processed, so as to determine the corner point of the to-be-recognized figure according to the target near-corner point.
  • the method for recognizing the corner points of the graphics in the image provided by the embodiments of the present application is generally performed by the server 105 , and accordingly, the device for identifying the corner points of the graphics in the image is generally set in the server 105 .
  • the terminal device may also have a similar function to the server, so as to implement the solution of the method for identifying the corner points of a graph in an image provided by the embodiments of the present application.
  • FIG. 2 shows a schematic flowchart of a method for identifying corner points of graphics in an image according to an embodiment of the present application.
  • the method for identifying the corners of the graphics in the image includes at least steps S210 to S240, and the details are as follows:
  • step S210 image recognition is performed on the image containing the to-be-recognized figure, and position information of the boundary point corresponding to the to-be-recognized figure is determined.
  • the graphic to be recognized may be a graphic corresponding to the shape of the area to be recognized. For example, if the corner of the test chart is identified, the graphic to be identified is the shape of the test chart. If the corner of the screen is identified, the graphic to be identified is the shape of the test chart. That is, the shape of the screen, and so on.
  • the to-be-recognized figure may be any polygonal figure, such as a triangle, quadrilateral, pentagon or hexagon, and the to-be-recognized figure may also be any other irregular figure, which is not specifically limited in this application.
  • the position information may be information used to determine the position of the boundary point in the image, and the position information may include, but is not limited to, the coordinate information of the boundary point, the adjacent relationship, and the like.
  • the adjacent relationship may be information used to represent the relative positions of boundary points, so as to determine whether a certain boundary point is adjacent to other boundary points.
  • the boundary points may be identified in a certain direction (eg, clockwise or counterclockwise), and the previously identified boundary points and the later identified boundary points are adjacent to the current boundary point.
  • a certain direction eg, clockwise or counterclockwise
  • the positions corresponding to the front and rear identified boundary points are also located on both sides of the current boundary point, so the adjacent boundary points can be determined according to the identification sequence of the boundary points. relation.
  • Adjacent boundary points include boundary point B and boundary point D, and so on.
  • a user may acquire an image containing a to-be-recognized graphic through an image acquisition device configured in a terminal device (such as a smart phone, etc.), and upload the image to the server.
  • the server may perform image recognition on the image containing the to-be-recognized figure, thereby determining boundary point information corresponding to the to-be-recognized figure.
  • the server may also acquire a previously stored image containing the graphic to be recognized from its own storage location to complete subsequent recognition, which is not specifically limited in this application.
  • the server may use the center point of the image as the coordinate origin, identify the boundary points of the image to be recognized in a certain order (eg, clockwise or counterclockwise), and obtain the coordinate information of each boundary point.
  • the identified boundary points can be numbered in sequence, so as to obtain the adjacent relationship between the boundary points.
  • the server identifies and obtains boundary points 1, 2, 3, 4, 5, ..., N-1, N in turn. It should be understood that for boundary point 2, boundary point 1 and boundary point 3 are adjacent boundaries.
  • boundary point 3 the boundary point 2 and the boundary point 4 are adjacent boundary points, ..., and since the server identifies in the clockwise or counterclockwise direction, after one rotation, for the finally identified boundary point N, its adjacent boundary points should be boundary point (N-1) and boundary point 1, and the adjacent boundary points of boundary point 1 should be boundary point N and boundary point 2.
  • the identification of the boundary points of the graphics to be identified may be based on the difference between the gray value of the graphics to be identified and the gray value of the background. If the difference in degree value is too large, the point can be identified as the boundary point of the graphic to be identified. Specifically, the difference between the gray values on both sides of the point can be compared with a preset difference threshold, and if the difference between the gray values is greater than or equal to the difference threshold, it means the gray values on both sides of the point If the difference is too large, the point can be identified as a boundary point; if the difference between the gray values is less than the difference threshold, it means that the difference between the gray values on both sides of the point is small, and the point is not a boundary point.
  • the difference threshold may be preset by those skilled in the art based on prior experience, or may be separately set for the degree of difference in color between the image to be recognized and the background. For example, if the image to be recognized is white and the background is black, and the difference between the two gray values is large, the corresponding difference threshold should also be increased accordingly. If the image to be recognized is purple and the background is blue, the gray values of the two If the difference is small, the corresponding difference threshold should also be reduced accordingly. Therefore, the difference threshold can be adapted to the color of the image to be recognized and the color of the background, so as to ensure the accuracy of boundary point recognition. Those skilled in the art can select different difference threshold determination methods according to actual implementation needs, which are not specifically limited in this application.
  • step S220 an included angle formed by each of the three adjacent boundary points is determined according to the position information of the boundary points, wherein the boundary point at the middle position is the angle vertex.
  • the server may determine an included angle based on each group of adjacent three border points according to the adjacent relationship between the border points, wherein the border point at the middle position is the part of the included angle. Angle vertex.
  • boundary points 1, 2, 3, 4, 5, ... in sequence the boundary points 1, 2 and 3 can form a point.
  • Sets, boundary points 2, 3 and 4 can form a set of points
  • boundary points 3, 4 and 5 can form a set of points, and so on.
  • the three boundary points in each point set use the boundary point at the middle position as the angle vertex, which can form an included angle.
  • the point set A contains boundary points a, b and c, and the boundary point b at the middle position is the vertex of the angle, then the angle formed is ⁇ abc, and so on.
  • the angle vertices of the adjacent point sets are also adjacent, for example, the point set A includes boundary points [a, b, c], and the adjacent point set B includes boundary points [b, c, d] , then the angle vertices b and c of point set A and point set B are adjacent, and so on.
  • the server can calculate the size of the included angle formed by the three adjacent boundary points according to the coordinate information of each of the three adjacent boundary points, using the boundary point at the middle position as the angle vertex, wherein, according to the For the size of the included angle corresponding to the calculation of the coordinate information, reference may be made to the existing mathematical calculation method, which will not be repeated in this application.
  • step S230 according to the included angle formed by each of the three adjacent boundary points, the near-corner point to be processed is identified from the included angle vertices of each of the included angles.
  • the near-corner point can be the boundary point closest to the position of the corner point. It should be understood that since the angle formed by any three points on the same boundary on the polygon is close to 180°, the angle formed by any three points on the same boundary on the polygon can be The size of the angle formed determines whether the vertex of the angle is a near-corner point, that is, if the angle is close to 180°, it means that the vertex of the angle is not a near-corner point, and if the angle is far less than 180° , it means that the included angle vertex of the included angle is more likely to be the near-corner point, so the included angle vertex can be identified as the near-corner point to be processed.
  • identifying the near-corner point to be processed from the angle vertices of each of the included angles according to the included angle formed by each of the three adjacent boundary points includes:
  • the included angle vertex of the included angle is identified as the near-corner point to be processed.
  • those skilled in the art can preset a predetermined threshold for defining the upper limit of the included angle formed by the near-corner point.
  • the server may compare the included angle formed by each of the three adjacent boundary points with the predetermined threshold. If the included angle formed by a certain group of adjacent three boundary points is smaller than the predetermined threshold, it means that the The angle is too small. Therefore, the vertex of the included angle, that is, the boundary point at the middle position, is identified as the near-corner point to be processed. If the included angle formed by a certain group of adjacent three boundary points is greater than or equal to the predetermined threshold, it means that the included angle is relatively large, and the included angle vertex of the included angle is not a near-corner point, so no processing is required.
  • the predetermined threshold may be preset by those skilled in the art based on prior experience, or different predetermined thresholds may be set according to different shapes of the graphics to be identified, for example, for a quadrilateral, the predetermined threshold It can be set to 177°. If the figure to be identified is a regular octagon, since the interior angles of the figure are all obtuse angles, the predetermined threshold can be correspondingly increased to 179°, etc., to ensure the accuracy of the near-angle point identification to be processed. .
  • the above predetermined threshold is only an exemplary example, which is not specifically limited in the present application.
  • step S240 a target near-corner point is identified from the to-be-processed near-corner points according to the number of the adjacent near-corner points to be processed, so as to determine the corner point of the to-be-identified figure according to the target near-corner point.
  • the near corner points should generally appear in pairs, and the near corner points appearing in pairs should be adjacent boundary points.
  • the near-corner point to be processed is a correctly identified near-corner point according to the number of adjacent near-corner points to be processed, so that the target near-corner point can be identified from the near-corner points to be processed, so as to remove the wrongly identified near-corner point.
  • the near-corner point is processed to ensure the accuracy of the target near-corner point recognition, and then the accuracy of the subsequent corner points determined according to the target near-corner point.
  • identifying a target near-corner point from the to-be-processed near-corner points according to the number of adjacent near-corner points to be processed includes:
  • the near-corner point to be processed has no other adjacent near-corner points to be processed, compare the included angles of the two adjacent boundary points as the included angle vertices, and determine the near-corner point to be processed and its adjacent near-corner point.
  • the included angle vertex with the larger included angle is the target near-corner point.
  • the server can compare the included angle between the two adjacent boundary points as the included angle vertices, so as to identify the included angle vertex with the larger included angle and the near-corner point to be processed as the target near-corner point, so as to ensure the target near-angle Points can be paired to facilitate subsequent corner determination.
  • the angle vertex c is identified as the near-corner point to be processed, and the angle vertices of the adjacent point set A and point set B are not identified as the near-corner point to be processed. Therefore, the near-corner point c to be processed does not have other adjacent corners. The near corner point to be processed.
  • the server can compare the angles corresponding to point set A and point set C, that is, compare ⁇ abc and ⁇ cde, if ⁇ abc is greater than ⁇ cde, then compare the included angle vertex b in point set A with that in point set B If ⁇ abc is less than ⁇ cde, the included angle vertex c in point set B and the included angle vertex d in point set C are identified as the target near corner point; if ⁇ abc is equal to ⁇ cde, any one of the included angle vertices b or c and the included angle vertex c in the point set B can be identified as the target near-corner point, and so on.
  • identifying a target near-corner point from the to-be-processed near-corner points according to the number of adjacent near-corner points to be processed includes:
  • the to-be-processed near-corner point has other adjacent to-be-processed near-corner points, and the to-be-processed near-corner point exists continuously, wherein the number of the continuous to-be-processed near-corner points is two, the two consecutive near-corner points to be processed The near-corner point to be processed is identified as the target near-corner point.
  • the two near-corner points to be processed conform to the law that near-corner points appear in pairs. Therefore, the two near-corner points to be processed can both be identified as The target near-corner point for subsequent identification of the corner point.
  • identifying a target near-corner point from the to-be-processed near-corner points according to the number of adjacent near-corner points to be processed includes:
  • the to-be-processed near-corner point has other adjacent to-be-processed near-corner points, the to-be-processed near-corner points exist continuously, wherein the number of the continuous near-corner points to be processed is greater than or equal to three, then the continuous near-corner points to be processed.
  • the to-be-processed near-corner points other than both ends of the to-be-processed near-corner points are deleted, so as to perform target near-corner point identification according to the deleted boundary point information.
  • the server may retain the two ends of the three or more near-corner points to be processed, and delete the three or more near-corner points.
  • the above to-be-processed near-corner points at the intermediate positions of the to-be-processed near-corner points are obtained, thereby obtaining the deleted boundary point information.
  • the two ends of the four near-corner points to be processed are reserved, that is, the near-corner points a and d to be processed are reserved, and the middle
  • the near-corner points b and c to be processed are deleted to update the boundary point information of the figure to be recognized.
  • the server may re-divide the point set according to the deleted boundary point information, and perform the above-mentioned near-corner point identification process to be processed according to the re-divided point set to identify the deleted
  • the to-be-processed near-corner point corresponding to the boundary point information and finally the above-mentioned target near-corner point identification process is performed according to the newly obtained to-be-processed near-corner point.
  • the location information includes coordinate information; then, according to the location information of the boundary points, the angle formed by each of the three adjacent boundary points is determined, including:
  • an included angle formed by every three adjacent boundary points is determined.
  • the server may perform image recognition according to the image containing the graphic to be recognized, so as to determine the coordinate information of the boundary point corresponding to the graphic to be recognized. And the adjacent relationship of the boundary points can be determined according to the coordinate information of the boundary points. It should be understood that, in the coordinate information of the boundary points on the same boundary, at least the absolute value of the difference between the X coordinate or the Y coordinate is within a certain threshold range. Those skilled in the art can preset the difference threshold based on prior experience, and the server can calculate the absolute value of the difference between the X coordinate or Y coordinate between any two boundary points, and the absolute value of the difference can be compared with the preset difference. The set difference threshold is compared to determine whether two boundary points are on the same boundary.
  • the coordinate information of boundary point A is (6, 8), and the coordinate information of boundary point B is (9, 9), then the absolute value of the difference between the coordinate values of boundary point A and boundary point B is 3 respectively (the absolute value of the difference of the X coordinate) and 1 (the absolute value of the difference of the Y coordinate), if the preset difference threshold is 1, so after the comparison, the Y coordinate of the boundary point A and the boundary point B are determined.
  • the absolute value of the difference is less than or equal to the difference threshold, so it can be determined that the boundary point A and the boundary point B are boundary points on the same horizontal side, and so on.
  • two difference thresholds which are respectively used for the comparison of the X coordinate and the Y coordinate, and the absolute difference between the coordinate information of the two boundary points.
  • the values should be less than or equal to the difference threshold, so that the distance between the two determined boundary points can be guaranteed to be small, thereby ensuring the accuracy of the boundary points on the same boundary. Avoid identifying boundary points on two horizontal sides or two vertical sides as boundary points on the same boundary.
  • the server determines the boundary points pair by pair. After all boundary points are determined, the server can integrate the determination results in a chained manner. For example, boundary point A and Boundary point B is a boundary point on the same boundary, and boundary point B and boundary point C are boundary points on the same boundary, then boundary point A, boundary point B, and boundary point C should all be boundary points on the same boundary, etc. .
  • the server can also compare the coordinate information of the boundary points located on the same boundary to prevent the boundary points on adjacent boundaries from being identified as located on the same boundary, and ensure the accuracy of the division of boundary points located on the same boundary .
  • the server can calculate according to the coordinate information of the boundary points located on the same boundary, and the two boundary points that are located on both sides of a boundary point and have the smallest distance from the boundary point will be calculated.
  • the boundary point is identified as the adjacent boundary point of the boundary point, that is, the adjacent relationship between the boundary points is determined. It should be noted that, for a horizontal side, the two sides refer to the left and right sides, and if it is a vertical side, the two sides refer to the upper side and the lower side.
  • the server may determine the horizontal edge or the vertical edge based on the coordinate information of the boundary points on the same boundary.
  • the difference in the X coordinate is greater than the difference on the Y coordinate, and the boundary on the vertical edge
  • the difference of the X coordinate in the coordinate information of the point is smaller than the difference of the Y coordinate, and so on.
  • the server may identify the endpoints on each boundary as the adjacent boundary points of the adjacent boundary points, thereby ensuring that each boundary point has two adjacent boundary points , for subsequent calculations.
  • the angle formed by each of the three adjacent boundary points can be determined.
  • the specific determination method can refer to the above, which is not repeated in this application. Repeat.
  • the position information of the boundary point includes coordinate information and an adjacent relationship, then image recognition is performed on the image containing the graphic to be recognized, and the The position information of the boundary points corresponding to the graphics to be identified, including:
  • the adjacent relationship of the boundary points is determined.
  • the server may use the center point of the image as the origin of coordinates, identify the boundary points of the graphics to be identified in a clockwise or counterclockwise direction, thereby obtain the coordinate information of the boundary points of the graphics to be identified, and according to the boundary points of the graphics to be identified Identify the sequence and determine the adjacent relationship of the boundary point.
  • the server may take the origin of the coordinates as a starting point, and radiate a plurality of rays outward at equal intervals in a clockwise or counterclockwise order. It should be understood that when the ray extends to the boundary of the to-be-recognized figure, the boundary point corresponding to the to-be-recognized figure can be identified and the coordinate information corresponding to the boundary point can be obtained due to the large difference in the grayscale values on both sides of the boundary.
  • the interval can be preset by those skilled in the art. For example, if the server can radiate a ray every 10°, there should be 36 rays in a circle, that is, 360°, and each ray can determine a boundary point, then 36 boundary points can be obtained correspondingly, and so on. It should be noted that those skilled in the art can determine the corresponding interval size according to actual implementation needs. For example, the interval can be 12°, 18°, or 20°, etc. The above are only exemplary examples, which are not specifically limited in this application.
  • the server identifies the boundary points of the graphic to be recognized in a certain order (for example, clockwise or counterclockwise), and obtains coordinate information of each boundary point.
  • the identified boundary points can be numbered in sequence, so as to obtain the adjacent relationship between the boundary points.
  • the server identifies and obtains boundary points 1, 2, 3, 4, 5, ..., N-1, N in turn.
  • boundary point 1 and boundary point 3 are adjacent boundaries.
  • border point 2 and border point 4 are adjacent border points, . .
  • the server since the server identifies in a clockwise or counterclockwise direction. Therefore, after one rotation, for the finally recognized boundary point N, its adjacent boundary points should be boundary point (N-1) and boundary point 1, and the adjacent boundary points of boundary point 1 should be Boundary Point N and Boundary Point 2.
  • the adjacent relationship of the boundary points can be determined, which improves the subsequent identification efficiency.
  • FIG. 3 shows a schematic flowchart of determining a corner point in a method for identifying a corner point of a graphic in an image according to an embodiment of the present application.
  • determining the corner points at least includes steps S310 to S320, which are described in detail as follows:
  • step S310 a plurality of the target near-corner points are grouped according to the adjacent relationship to obtain a target near-corner point combination.
  • the target near-corner points all appear in pairs and each pair of target near-corner points are adjacent boundary points.
  • the server can group adjacent target near-corner points into a group, thereby obtaining multiple target near-corner point combinations. For example, there are eight target near-corner points A, B, C, D, E, F, G, and H, where A and B are adjacent, C and D are adjacent, E and F are adjacent, and G and H are adjacent. Then A and B can be divided into a target near-corner combination, C and D can be divided into a target near-corner combination, ..., and so on.
  • step S320 according to the positional relationship of each target near-corner point in each of the target near-corner point combinations, the corner point of the to-be-recognized figure is determined.
  • the corner points of the to-be-recognized pattern are determined according to the positional relationship of the target near-corner points in each target near-corner point combination. For example, if the figure to be recognized is a rectangle, it should be understood that in the combination of two adjacent target near-corner points, there are two target near-corner points located on the same boundary of the rectangle. By identifying one boundary of the figure, and so on, four boundaries of the figure to be recognized can be determined, and the intersection of the four boundaries is the corner point of the figure to be recognized, and so on.
  • the target near-corner points by grouping the target near-corner points, a combination of multiple target near-corner points is obtained, and according to the positional relationship between the target near-corner points among the multiple target near-corner points, the to-be-identified near-corner point is determined.
  • the corner positions of the graphics to ensure the accuracy of the corner positions of the graphics to be identified.
  • FIG. 4 shows a schematic flowchart of step S320 in the method for recognizing the corner points of the graphics in the image of FIG. 3 according to an embodiment of the present application.
  • step S320 includes at least steps S410 to S430, and the details are as follows:
  • step S410 number each of the target near-corner point combinations and each of the target near-corner points in a clockwise or counterclockwise order, and determine a first number corresponding to each of the target near-corner point combinations and each of the targets The second number corresponding to the near corner point.
  • the server may use a certain boundary as a starting point, such as the positive direction of the X-axis in the coordinate system, etc., and number each target near-corner point combination and each target near-corner point respectively in a clockwise or counterclockwise order, thereby A first number corresponding to each target near-corner point combination and a second number corresponding to each target near-corner point are obtained.
  • the server numbers the target near-corner point combinations as A, B, C, and D in clockwise order.
  • the two target near-corner points in the target near-corner point combination A can be numbered 1 and 2 in sequence
  • the two target near-corner points in point combination B can be numbered 3 and 4, . . . in sequence, so that the first number corresponding to each target near-corner point combination and the second number corresponding to each target near-corner point can be obtained.
  • the target near-corner point combination and the number of the target near-corner point correspond to the numbering direction of the server, that is, after the server is numbered, the target near-corner point combination and the number of the target near-corner point are the same as the server's numbering direction (for example, in the order of the numbering direction). Clockwise or counterclockwise) correspondingly distributed.
  • step S420 a straight line equation between the two target near-corner points is determined according to the two target near-corner points in the combination of the two adjacent target near-corner points with the first number, the two target near-corner points The point is the near-corner point of the target adjacent to the second number.
  • the two target near-corner points are targets adjacent to the second number. near corner.
  • the second numbers corresponding to the target near-corner points a, b, c and d are 1, 2, 3, and 4, then it can be determined that the adjacent target near corner points b and c of the second number (ie, 2 and 3) are located on the same boundary, and so on.
  • a boundary of the to-be-recognized pattern can be determined. Therefore, the straight line equation of the boundary can be determined according to the coordinate information of the near-corner points of the two targets.
  • the server can determine the equation of the straight line corresponding to each boundary in the graphic to be recognized.
  • step S430 the corner points of the to-be-recognized figure are determined according to the straight line equation determined by the combination of the target near-corner points adjacent to the first number.
  • the combination of two adjacent target near-corner points with the first number can determine a straight line equation
  • the combination of three consecutive target near-corner points with the first number can determine two straight lines Since the straight lines corresponding to the two straight line equations are the straight lines where the boundaries of the graphics to be identified are located, the corner points of the straight lines corresponding to the two straight line equations can be determined as the corner points of the graphics to be identified.
  • a and B can determine a straight line equation a
  • B and C can determine a straight line equation b
  • C and D can determine a straight line equation c
  • D and A can determine a straight line equation d
  • the straight line corresponding to the boundary of the graphic to be identified can be determined by using the two adjacent target near-corner points with the second number in the combination of the two adjacent target near-corner points with the first number. equation, and then determine the corner points of the figure to be identified through the determined straight line equation, which can ensure the accuracy of the determined corner point positions.
  • the server can also calculate the angle between each target near-corner point relative to the positive direction of the X-axis in the coordinate system, according to the angle from small to large If the difference between the angles corresponding to the two target near-corner points is large, it means that the two target near-corner points are located on the same boundary. Determine a line equation, and then determine the corner points of the figure to be recognized according to the intersection between the line equations.
  • the angle difference can be compared with a preset threshold, or the angle difference with other adjacent target near-corner points can be compared. If the included angle difference is significantly larger (for example, the included angle difference is more than three times the included angle difference between the adjacent target near-corner point), it can be determined that the included angle difference is larger.
  • a preset threshold for example, the included angle difference is more than three times the included angle difference between the adjacent target near-corner point
  • the determining the corner points of the to-be-recognized figure according to the target near-corner points include:
  • the corner points of the to-be-recognized figure are determined.
  • the server may determine the target near-corner points located on the same boundary according to the coordinate information and the adjacent relationship of the boundary points. For example, according to the previously determined division result of the boundary point located on the same boundary (the division result may include the coordinate information of the boundary point located on the boundary), the coordinate information of the target near-corner point can be compared with the division result, Thereby, it is determined to which boundary the target near-corner point belongs, and the target near-corner point located on the same boundary is determined. It should be understood that the target near-corner points should be located on a boundary in pairs, and at both ends of the boundary, so the server can establish the boundary line equation according to the coordinate information of the target near-corner points on the same boundary.
  • the server can determine the equations of the straight lines corresponding to the four boundaries, and identify the intersections between the equations of the straight lines as the corner points of the figure to be identified.
  • the server may also determine whether the two target near corners are located on the same boundary according to the coordinate information of the target near corners. Specifically, if the two target near-corner points are located on the same boundary, the absolute value of the difference between at least one coordinate value of the two target near-corner points is small (the comparison can be made according to the difference threshold described above), and The absolute value of the difference between the other coordinate values is neither the smallest nor the largest, so as to determine the target near-corner point located on the same boundary.
  • target near-corner points respectively A(3,8), B(5,10), H(2,3) and G(3,2).
  • B, H, G are all the same as A is closer, so it is possible to be on the same boundary as point A.
  • the point on the same boundary as point A must be between the other two points.
  • the point H is located between the point B and the point G, that is, the absolute value of the difference between the coordinate values in the Y-axis direction is neither the largest nor the smallest.
  • the server may determine, according to the coordinate information of point A, the target near-corner point with a smaller absolute value of the difference between a certain coordinate value and the coordinate value of A, that is, point B, point H, and point G, depending on the For the points B, G, and H, the candidate is the point where the point A is on the same boundary. Then compare the absolute value of the difference between the Y coordinate values of the three points and point A, at this time, it can be confirmed that the coordinate value of point H satisfies the condition, 2 ⁇ 5 ⁇ 6.
  • those skilled in the art can also set a difference threshold for determining the same boundary point. If the error is small, the threshold can be set small to exclude interference from other points.
  • Figures 5a to 5c show schematic flowcharts of a method for identifying corner points of graphics in an image according to an embodiment of the present application (the graphics to be identified are used as test charts as an example for description below).
  • the server can use the center point of the image as the coordinate origin, radiate multiple rays outward at equal intervals in a clockwise order, and determine the boundary points of the test chart according to the gray value changes along the path of each ray, so that The coordinate information of the boundary point is obtained, and the identified boundary points are numbered A , B, C, D, .
  • the server divides every three adjacent boundary points into a point set, for example, boundary points A, B and C form a point set, and boundary points B, C and D form a point set ...,and many more. And according to the coordinate information of the three boundary points in each point set, the boundary point at the middle position is used as the angle vertex to calculate the corresponding angle of each point set, so as to obtain ⁇ ABC, ⁇ BCD, ⁇ CDE, ..., ⁇ The size of F 1 G 1 A, ⁇ G 1 AB.
  • the server can obtain the included angles corresponding to the two point sets adjacent to the boundary point A to be processed, that is, ⁇ F 1 G 1 A and ⁇ ABC, and compare the two, so as to compare the included angle of the point set with the larger angle.
  • the corner vertex and the near-corner point A to be processed are identified as the target near-corner point. If ⁇ F 1 G 1 A is greater than ⁇ ABC, then point G 1 and point A can be determined as the target near-corner point.
  • the boundary points L, M , N, S, T, U, and V are all identified as near-corner points to be processed, so there will be three or more consecutive near-corner points to be processed.
  • the server may remove the near-corner points to be processed except for the two ends among the three or more consecutive near-corner points to be processed, that is, remove the boundary points M, T and U, so as to delete the wrongly identified boundary points. , to obtain the updated boundary point information as shown in Figure 6.
  • the server then identifies the near-corner points to be processed according to the updated boundary point information, and obtains new near-corner points to be processed as A, J, K, Q, R, Z, A 1 and G 1 (as shown in FIG. 7 ). Since the near-corner points to be processed all appear in pairs, the above-mentioned near-corner points to be processed can all be identified as target near-corner points.
  • the server may determine a straight line equation according to the coordinate information of the adjacent two adjacent target near corner points in the combination of two adjacent target near corner points, and the straight line equation corresponds to the boundary of the test chart.
  • the target near-corner point A and the target near-corner point J determine a line equation
  • the target near-corner point K and the target near-corner point Q determine a line equation, and so on.
  • the intersection point between the determined line equations is the corner point of the test chart. In this way, the corner points of the test map can be quickly identified, and the accuracy of the corner point position identification can be ensured.
  • FIG. 6 shows a flowchart of determining a corner point in a method for identifying a corner point of a graphic in an image according to an embodiment of the present application.
  • the server can identify the boundary points in a certain order. If a boundary point is determined to be a near-corner point to be processed, the number of consecutive near-corner points to be processed is set to 1, and then the next boundary point is identified. If the next boundary point is a near-corner point to be processed, add one to the number of continuous near-corner points to be processed, and continue to judge whether the next boundary point is a near-corner point to be processed. If not, the number of consecutive near-corner points to be processed is judged. If the number of continuous near-corner points to be processed is equal to 2, it means that the group of near-corner points to be processed is normal and can be identified as the target near-corner point.
  • processing mode 1 can be used for processing, that is, the above-mentioned situation where there are no adjacent near-corner points to be processed can be used to identify another near-corner point. A pending near corner point.
  • processing mode 2 can be adopted, that is, processing is performed with reference to the above-mentioned situation that the number of continuous near-corner points to be processed is greater than or equal to three, Delete the points to obtain the updated boundary point information, and then re-identify the near-corner points to be processed according to the updated boundary point information.
  • the corresponding straight line equation can be determined according to the identified target near corner points, and the intersection between the straight line equations can be calculated as the corner points of the figure to be identified.
  • the wrongly identified near-corner points can be removed from the to-be-processed near-corner points, thereby obtaining the target near-corner points, ensuring the accuracy of the recognition of the target near-corner points, and thus ensuring The accuracy of subsequent corner position determination.
  • FIG. 7 shows a block diagram of an apparatus for identifying corner points of graphics in an image according to an embodiment of the present application.
  • an apparatus for identifying corner points of graphics in an image includes:
  • the image recognition module 710 is configured to perform image recognition on the image containing the figure to be recognized, and determine the boundary point information corresponding to the figure to be recognized, and the boundary point information includes the coordinate information of the boundary point and the phase relationship between the boundary points. neighbor relationship;
  • the calculation module 720 is used to calculate according to the coordinate information of the boundary points and the adjacent relationship, and determine the angle formed by the boundary points in each point set, and the point set is composed of three adjacent boundary points. , and the boundary point at the middle position is the angle vertex;
  • a near-corner point identification module 730 configured to identify a near-corner point to be processed from the angle vertices of each of the point sets according to the angle corresponding to each of the point sets;
  • the corner determination module 740 is configured to identify a target near corner point from the near corner points to be processed according to the number of adjacent near corner points to be processed, so as to determine the corner of the to-be-recognized figure according to the target near corner point. point.
  • modules or units of the apparatus for action performance are mentioned in the above detailed description, this division is not mandatory. Indeed, according to embodiments of the present application, the features and functions of two or more modules or units described above may be embodied in one module or unit (eg, with image recognition, computation, near-corner point recognition, and corner point determination capable processors, etc.). Conversely, the features and functions of one module or unit described above may be further divided into multiple modules or units to be embodied.
  • FIG. 8 shows a schematic structural diagram of a computer system suitable for implementing the electronic device according to the embodiment of the present application.
  • the computer system includes a central processing unit (Central Processing Unit, CPU) 801, which can be loaded into random access according to a program stored in a read-only memory (Read-Only Memory, ROM) 802 or from a storage part 808
  • the program in the memory (Random Access Memory, RAM) 803 performs various appropriate actions and processes, such as performing the methods described in the above embodiments.
  • RAM 803 Random Access Memory
  • various programs and data required for system operation are also stored.
  • the CPU 801, the ROM 802, and the RAM 803 are connected to each other through a bus 804.
  • An Input/Output (I/O) interface 805 is also connected to the bus 804 .
  • the following components are connected to the I/O interface 805: an input section 806 including a keyboard, a mouse, etc.; an output section 807 including a cathode ray tube (CRT), a liquid crystal display (LCD), etc., and a speaker, etc. ; a storage part 808 including a hard disk, etc.; and a communication part 809 including a network interface card such as a LAN (Local Area Network) card, a modem, and the like.
  • the communication section 809 performs communication processing via a network such as the Internet.
  • a drive 810 is also connected to the I/O interface 805 as needed.
  • a removable medium 811 such as a magnetic disk, an optical disk, a magneto-optical disk, a semiconductor memory, etc., is mounted on the drive 810 as needed so that a computer program read therefrom is installed into the storage section 808 as needed.
  • embodiments of the present application include a computer program product comprising a computer program carried on a computer-readable medium, the computer program comprising a computer program for performing the method illustrated in the flowchart.
  • the computer program may be downloaded and installed from the network via the communication portion 809, and/or installed from the removable medium 811.
  • CPU central processing unit
  • the computer-readable medium shown in the embodiments of the present application may be a computer-readable signal medium or a computer-readable storage medium, or any combination of the above two.
  • the computer-readable storage medium can be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus or device, or a combination of any of the above.
  • Computer readable storage media may include, but are not limited to, electrical connections with one or more wires, portable computer disks, hard disks, random access memory (RAM), read only memory (ROM), erasable Erasable Programmable Read Only Memory (EPROM), flash memory, optical fiber, portable Compact Disc Read-Only Memory (CD-ROM), optical storage device, magnetic storage device, or any suitable of the above The combination.
  • a computer-readable storage medium can be any tangible medium that contains or stores a program that can be used by or in conjunction with an instruction execution system, apparatus, or device.
  • a computer-readable signal medium may include a data signal propagated in baseband or as part of a carrier wave, carrying a computer-readable computer program therein.
  • Such propagated data signals may take a variety of forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination of the foregoing.
  • a computer-readable signal medium can also be any computer-readable medium other than a computer-readable storage medium that can transmit, propagate, or transport the program for use by or in connection with the instruction execution system, apparatus, or device .
  • a computer program embodied on a computer-readable medium may be transmitted using any suitable medium, including but not limited to: wireless, wired, etc., or any suitable combination of the foregoing.
  • each block in the flowchart or block diagram may represent a module, program segment, or part of code, and the above-mentioned module, program segment, or part of code contains one or more executables for realizing the specified logical function instruction.
  • the functions noted in the blocks may occur out of the order noted in the figures. For example, two blocks shown in succession may, in fact, be executed substantially concurrently, or the blocks may sometimes be executed in the reverse order, depending upon the functionality involved.
  • the units involved in the embodiments of the present application may be implemented in software or hardware, and the described units may also be provided in a processor. Among them, the names of these units do not constitute a limitation on the unit itself under certain circumstances.
  • the present application also provides a computer-readable medium.
  • the computer-readable medium may be included in the electronic device described in the above embodiments; it may also exist alone without being assembled into the electronic device. middle.
  • the above-mentioned computer-readable medium carries one or more programs, and when the above-mentioned one or more programs are executed by an electronic device, enables the electronic device to implement the methods described in the above-mentioned embodiments.
  • the exemplary embodiments described herein may be implemented by software, or may be implemented by software combined with necessary hardware. Therefore, the technical solutions according to the embodiments of the present application may be embodied in the form of software products, and the software products may be stored in a non-volatile storage medium (which may be CD-ROM, U disk, mobile hard disk, etc.) or on the network , which includes several instructions to cause a computing device (which may be a personal computer, a server, a touch terminal, or a network device, etc.) to execute the method according to the embodiments of the present application.
  • a computing device which may be a personal computer, a server, a touch terminal, or a network device, etc.

Landscapes

  • Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • General Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • Geometry (AREA)
  • Image Analysis (AREA)

Abstract

一种图像中图形角点的识别方法、装置、介质及电子设备。该图像中图形角点的识别方法包括：对包含待识别图形的图像进行图像识别，确定所述待识别图形对应的边界点的位置信息（S210）；根据所述边界点的位置信息，确定每相邻的三个边界点形成的夹角（S220）；根据每相邻的三个边界点形成的夹角，从各所述夹角的夹角顶点中识别出待处理近角点（S230）；根据相邻的所述待处理近角点的数量，从所述待处理近角点中识别出目标近角点，以根据所述目标近角点确定所述待识别图形的角点（S240）。本申请实施例的技术方案可以快速识别待识别图形的角点位置，并保证角点位置识别的准确性。

Description

图像中图形角点的识别方法、装置、介质及电子设备 技术领域
本申请涉及图像处理技术领域，具体而言，涉及一种图像中图形角点的识别方法、装置、介质及电子设备。
背景技术
在安装激光电视时，需要专业人员对投影机机身位置进行校正，以使投影机的投影与屏幕相重合。在目前的技术方案中，投影机在屏幕上投影测试图形，通过检测测试图形的形状以校正投影机的机身位置。由此，如何快速识别测试图形以及屏幕的角点位置以确定测试图形与屏幕的位置，保证角点位置识别的准确性成为了亟待解决的技术问题。
发明内容
本申请的实施例提供了一种图像中图形角点的识别方法、装置、介质及电子设备，进而至少在一定程度上可以快速识别测试图形以及屏幕的角点位置以确定测试图形与屏幕的位置，保证角点位置识别的准确性。
本申请的其他特性和优点将通过下面的详细描述变得显然，或部分地通过本申请的实践而习得。
根据本申请实施例的一个方面，提供了一种图像中图形角点的识别方法，该方法包括：
对包含待识别图形的图像进行图像识别，确定所述待识别图形对应的边界点的位置信息；
根据所述边界点的位置信息，确定每相邻的三个边界点形成的夹角，其中，位于中间位置的所述边界点为夹角顶点；
根据每相邻的三个边界点形成的夹角，从各所述夹角的夹角顶点中识别出待处理近角点；
根据相邻的所述待处理近角点的数量，从所述待处理近角点中识别出目标近角点，以根据所述目标近角点确定所述待识别图形的角点。
在本申请的一些实施例中，基于上述方案，所述位置信息包括坐标信息；所述根据所述边界点的位置信息，确定每相邻的三个边界点形成的夹角，包括：根据所述边界点的坐标信息，确定所述边界点的相邻关系；根据所述边界点的坐标信息和所述相邻关系，确定每相邻的三个边界点形成的夹角。
在本申请的一些实施例中，基于上述方案，所述位置信息包括坐标信息和相邻关系；对包含待识别图形的图像进行图像识别，确定所述待识别图形对应的边界点的位置信息，包括；以所述图像的中心点为坐标原点，按照顺时针或者逆时针的顺序识别所述待识别图形的边界点，确定所述待识别图形对应的边界点的坐标信息；根据所述边界点的识别先后顺序，确定所述边界点的相邻关系。
在本申请的一些实施例中，基于上述方案，所述根据相邻的所述待处理近角点的数量，从所述待处理近角点中识别出目标近角点，包括：
若所述待处理近角点未存在相邻的其他待处理近角点，则将与其相邻的两个边界点作为夹角顶点的夹角进行比较，确定所述待处理近角点以及与其相邻的两个边界点作为夹角顶点的夹角中夹角较大的夹角顶点为目标近角点。
在本申请的一些实施例中，基于上述方案，所述根据相邻的所述待处理近角点的数量，从所述待处理近角点中识别出目标近角点，包括：
若所述待处理近角点存在相邻的其他待处理近角点，所述待处理近角点连续存在，其中，连续的所述待处理近角点的数量为两个，则将两个连续的所述待处理近角点识别为目标近角点。
在本申请的一些实施例中，基于上述方案，所述根据相邻的所述待处理近角点的数量，从所述待处理近角点中识别出目标近角点，包括：
若所述待处理近角点存在相邻的其他待处理近角点，所述待处理近角点连续存在，其中，连续的所述待处理近角点的数量大于或等于三个，则将连续的所述待处理近角点中除两端之外的待处理近角点进行删除，以根据删除后的所述边界点信息进行目标近角点识别。
在本申请的一些实施例中，基于上述方案，所述根据每相邻的三个边界点形成的夹角，从各所述夹角的夹角顶点中识别出待处理近角点，包括：
若相邻的三个边界点形成的夹角小于预定阈值，则将所述夹角的夹角顶点识别为待处理近角点
在本申请的一些实施例中，基于上述方案，所述目标近角点的数量为多个，则所述根据所述目标近角点确定所述待识别图形的角点，包括：
将多个所述目标近角点按照相邻关系进行分组，得到目标近角点组合；
根据各所述目标近角点组合中各目标近角点的位置关系，确定所述待识别图形的角点。
在本申请的一些实施例中，基于上述方案，根据各所述目标近角点组合中各目标近角点的位置关系，确定所述待识别图形的角点，包括：
按照顺时针或逆时针的顺序对各所述目标近角点组合以及各所述目标近角点分别进行编号，确定各所述目标近角点组合对应的第一编号以及各所述目标近角点对应的第二编号；
根据所述第一编号相邻的两个所述目标近角点组合中的两个目标近角点，确定所述两个目标近角点之间的直线方程，所述两个目标近角点为所述第二编号相邻的目标近角点；
将各所述直线方程对应的直线的交点识别为所述待识别图形的角点。
在本申请的一些实施例中，基于上述方案，对包含待识别图形的图像进行图像识别，确定所述待识别图形对应的边界点信息，包括；
以所述图像的中心点为坐标原点，按照顺时针或者逆时针的顺序识别所述待识别图形的边界点，以确定所述待识别图形对应的边界点信息。
在本申请的一些实施例中，基于上述方案，所述目标近角点的数量为多个，则所述根据所述目标近角点确定所述待识别图形的角点，包括：根据所述目标近角点的位置信息，确定位于同一边界上的目标近角点；根据位于同一边界上的目标近角点，确定所述待识别图形的角点。
根据本申请实施例的一个方面，提供了一种图像中图形角点的识别装置，该装置包括：
图像识别模块，用于对包含待识别图形的图像进行图像识别，确定所述待识别图形对应的边界点信息，所述边界点信息包括边界点的坐标信息以及所述边界点之间的相邻关系；
计算模块，用于根据所述边界点的坐标信息以及相邻关系，确定每相 邻的三个边界点形成的夹角，其中，位于中间位置的所述边界点为夹角顶点；
近角点识别模块，用于根据每相邻的三个边界点形成的夹角，从各所述夹角的夹角顶点中识别出待处理近角点；
角点确定模块，用于根据相邻的所述待处理近角点的数量，从所述待处理近角点中识别出目标近角点，以根据所述目标近角点确定所述待识别图形的角点。
根据本申请实施例的一个方面，提供了一种计算机可读介质，其上存储有计算机程序，所述计算机程序被处理器执行时实现如上述实施例中所述的图像中图形角点的识别方法。
根据本申请实施例的一个方面，提供了一种电子设备，包括：一个或多个处理器；存储装置，用于存储一个或多个程序，当所述一个或多个程序被所述一个或多个处理器执行时，使得所述一个或多个处理器实现如上述实施例中所述的图像中图形角点的识别方法。
在本申请的一些实施例所提供的技术方案中，通过预先识别出待处理近角点，并根据相邻的待处理近角点的数量，对待处理近角点进行筛选，去除识别错误的待处理近角点，以从待处理近角点中识别出目标近角点，保证了目标近角点识别的准确性，再根据识别得到的目标近角点确定待识别图形的角点。由此，在角点位置的识别过程中无需繁琐步骤，不仅可以提高角点位置识别的效率，同时也能够保证角点位置识别的准确性。
应当理解的是，以上的一般描述和后文的细节描述仅是示例性和解释性的，并不能限制本申请。
附图说明
此处的附图被并入说明书中并构成本说明书的一部分，示出了符合本申请的实施例，并与说明书一起用于解释本申请的原理。显而易见地，下面描述中的附图仅仅是本申请的一些实施例，对于本领域普通技术人员来讲，在不付出创造性劳动的前提下，还可以根据这些附图获得其他的附图。在附图中：
图1示出了可以应用本申请实施例的技术方案的示例性系统架构的示 意图。
图2示出了根据本申请的一个实施例的图像中图形角点的识别方法的流程示意图。
图3示出了根据本申请的一个实施例的图像中图形角点的识别方法中确定角点的流程示意图。
图4示出了根据本申请的一个实施例的图3的图像中图形角点的识别方法中步骤S320的流程示意图。
图5a至图5c示出了根据本申请的一个实施例的图像中图形角点的识别方法的流程示意图。
图6示出了根据本申请的一个实施例的图像中图形角点的识别方法中确定角点的流程框图。
图7示出了根据本申请的一个实施例的图像中图形角点的识别装置的框图。
图8示出了适于用来实现本申请实施例的电子设备的计算机系统的结构示意图。
具体实施方式
现在将参考附图更全面地描述示例实施方式。然而，示例实施方式能够以多种形式实施，且不应被理解为限于在此阐述的范例；相反，提供这些实施方式使得本申请将更加全面和完整，并将示例实施方式的构思全面地传达给本领域的技术人员。
此外，所描述的特征、结构或特性可以以任何合适的方式结合在一个或更多实施例中。在下面的描述中，提供许多具体细节从而给出对本申请的实施例的充分理解。然而，本领域技术人员将意识到，可以实践本申请的技术方案而没有特定细节中的一个或更多，或者可以采用其它的方法、组元、装置、步骤等。在其它情况下，不详细示出或描述公知方法、装置、实现或者操作以避免模糊本申请的各方面。
附图中所示的方框图仅仅是功能实体，不一定必须与物理上独立的实体相对应。即，可以采用软件形式来实现这些功能实体，或在一个或多个硬件模块或集成电路中实现这些功能实体，或在不同网络和/或处理器装置 和/或微控制器装置中实现这些功能实体。
附图中所示的流程图仅是示例性说明，不是必须包括所有的内容和操作/步骤，也不是必须按所描述的顺序执行。例如，有的操作/步骤还可以分解，而有的操作/步骤可以合并或部分合并，因此实际执行的顺序有可能根据实际情况改变。
图1示出了可以应用本申请实施例的技术方案的示例性系统架构的示意图。
如图1所示，系统架构可以包括终端设备(如图1中所示智能手机101、平板电脑102和便携式计算机103中的一种或多种，当然也可以是台式计算机等等)、网络104和服务器105。网络104用以在终端设备和服务器105之间提供通信链路的介质。网络104可以包括各种连接类型，例如有线通信链路、无线通信链路等等。
应该理解，图1中的终端设备、网络和服务器的数目仅仅是示意性的。根据实现需要，可以具有任意数目的终端设备、网络和服务器。比如服务器105可以是多个服务器组成的服务器集群等。
用户可以使用终端设备通过网络104与服务器105交互，以接收或发送消息等。服务器105可以是提供各种服务的服务器。例如用户利用终端设备103(也可以是终端设备101或102)配置的图像获取装置(例如摄像头等)可以获取包含待识别图形的图像，并将其向服务器105进行上传，服务器105可以对包含待识别图形的图像进行图像识别，确定待识别图形对应的边界点的位置信息，并根据所述边界点的位置信息，确定每相邻的三个边界点形成的夹角，其中，位于中间位置的边界点为夹角顶点，根据每相邻的三个边界点形成的夹角，从各夹角的夹角顶点中识别出待处理近角点，再根据相邻的待处理近角点的数量，从所述待处理近角点中识别出目标近角点，以根据目标近角点确定待识别图形的角点。
需要说明的是，本申请实施例所提供的图像中图形角点的识别方法一般由服务器105执行，相应地，图像中图形角点的识别装置一般设置于服务器105中。但是，在本申请的其它实施例中，终端设备也可以与服务器具有相似的功能，从而执行本申请实施例所提供的图像中图形角点的识别方法的方案。
以下对本申请实施例的技术方案的实现细节进行详细阐述：
图2示出了根据本申请的一个实施例的图像中图形角点的识别方法的流程示意图。参照图2所示，该图像中图形角点的识别方法至少包括步骤S210至步骤S240，详细介绍如下：
在步骤S210中，对包含待识别图形的图像进行图像识别，确定所述待识别图形对应的边界点的位置信息。
其中，待识别图形可以是与欲识别的区域的形状相对应的图形，例如若是识别测试图的角点，则待识别图形即为测试图的形状，若是识别屏幕的角点，则待识别图形即为屏幕的形状，等等。
需要说明的，该待识别图形可以是任意多边形的图形，例如三角形、四边形、五边形或者六边形等，待识别图形也可以是其他任意不规则的图形，本申请对此不作特殊限定。
位置信息可以是用于确定边界点在图像中的位置的信息，该位置信息可以包括但不限于边界点的坐标信息以及相邻关系等。其中，相邻关系可以是用于表示边界点的相对位置的信息，从而确定某一边界点与其他边界点是否相邻。
具体地，可以按照一定方向(例如顺时针方向或者逆时针方向)对边界点进行识别，在先识别的边界点以及在后识别的边界点则与当前的边界点为相邻。应该理解的，按照一定顺序对边界点进行识别时，前后所识别出的边界点相对应的位置也是位于当前边界点的两侧，所以可以根据边界点的识别顺序确定边界点之间的相邻关系。例如按照顺时针方向，依次识别出边界点A、边界点B、边界点C、边界点D…，则与边界点B相邻的边界点有边界点A和边界点C，与边界点C相邻的边界点有边界点B和边界点D，等等。
在本申请一示例性实施例中，用户可以通过终端设备(例如智能手机等)所配置的图像获取装置获取包含待识别图形的图像，并将该图像上传至服务器。服务器则可以对该包含待识别图形的图像进行图像识别，从而确定待识别图形对应的边界点信息。在一示例中，服务器也可以从自身的存储位置中获取在先存储的包含待识别图形的图像，以完成后续识别，本申请对此不作特殊限定。
具体地，服务器可以以图像的中心点为坐标原点，按照一定顺序(例如顺时针方向或者逆时针方向)对待识别图形的边界点进行识别，获取每一边界点的坐标信息。同时，可以依次对识别得到的边界点进行编号，从而得到边界点之间的相邻关系。例如服务器依次识别得到边界点1、2、3、4、5、……、N-1、N，应该理解的，对于边界点2而言，边界点1和边界点3为其相邻的边界点，对于边界点3而言，边界点2和边界点4为其相邻的边界点，……，而由于服务器是按照顺时针方向或者逆时针方向进行识别，因此，在旋转一周后，对于最后识别得到的边界点N而言，其相邻的边界点应为边界点(N-1)和边界点1，而边界点1的相邻的边界点应为边界点N和边界点2。
在一示例中，对待识别图形的边界点进行识别，可以是根据待识别图形的灰度值与背景的灰度值差异进行识别，若某一点一侧的灰度值与另一侧的灰度值差异过大，则可以将该点识别为待识别图形的边界点。具体地，可以将该点两侧的灰度值差值与预先设定的差值阈值进行比较，若灰度值差值大于或等于该差值阈值，则表示该点两侧的灰度值差异过大，因此可以将该点识别为边界点；若灰度值差值小于该差值阈值，则表示该点两侧的灰度值差异较小，该点并不是边界点。
需要说明的，差值阈值可以是由本领域技术人员根据在先经验所预先设定的，其也可以是针对待识别图形与背景之间颜色的差异度分别进行设定的。例如，待识别图形为白色，背景为黑色，二者灰度值差异较大，则相应的差值阈值也应该对应增大，若待识别图形为紫色，背景为蓝色，二者灰度值差异较小，则相应的差值阈值也应该对应减小。由此可以使得差值阈值与待识别图形的颜色以及背景的颜色相适配，保证边界点识别的准确性。本领域技术人员可以根据实际实现需要选用不同的差值阈值确定方法，本申请对此不作特殊限定。
请继续参考图2，在步骤S220中，根据所述边界点的位置信息，确定每相邻的三个边界点形成的夹角，其中，位于中间位置的所述边界点为夹角顶点。
在本申请一示例性实施例中，服务器可以根据边界点之间的相邻关系，基于每组相邻的三个边界点确定一个夹角，其中，位于中间位置的边界点为该夹角的夹角顶点。
以下以每相邻的三个边界点组成一个点集合为例进行说明，例如服务器依次识别得到边界点1、2、3、4、5、…，则边界点1、2和3可以组成一个点集合，边界点2、3和4可以组成一个点集合，边界点3、4和5可以形成一个点集合，等等。
每个点集合中的三个边界点以位于中间位置的边界点作为夹角顶点，可以形成一个夹角，例如点集合A中包含边界点a、b和c，以位于中间位置的边界点b为夹角顶点，则所形成的夹角为∠abc，等等。
同时，应该理解的，相邻点集合的夹角顶点也是相邻的，例如点集合A包括边界点[a,b,c]，相邻的点集合B包括边界点[b,c,d]，则点集合A和点集合B的夹角顶点b和c是相邻的，等等。
服务器可以根据每相邻的三个边界点的坐标信息，以位于中间位置的边界点为夹角顶点，计算相邻的三个边界点所形成的的夹角大小，其中，根据三个点的坐标信息计算对应的夹角大小可以参照现有的数学计算方法，本申请在此不再赘述。
在步骤S230中，根据每相邻的三个边界点形成的夹角，从各所述夹角的夹角顶点中识别出待处理近角点。
其中，近角点可以是距离角点位置最接近的边界点，应该理解的，由于多边形上同一条边界上任意三点所形成的角度接近180°，所以，可以根据各相邻的三个边界点形成的夹角大小，确定该夹角的夹角顶点是否为近角点，即若该夹角接近于180°，则表示该夹角的夹角顶点不是近角点，若该夹角远小于180°，则表示该夹角的夹角顶点具有较大可能性为近角点，因此可以将该夹角顶点识别为待处理近角点。
在本申请一示例性实施例中，所述根据每相邻的三个边界点形成的夹角，从各所述夹角的夹角顶点中识别出待处理近角点，包括：
若相邻的三个边界点形成的夹角小于预定阈值，则将所述夹角的夹角顶点识别为待处理近角点。
在该实施例中，本领域技术人员可以预先设定用于限定近角点所形成的夹角的上限的预定阈值。根据该预定阈值，服务器可以将各个相邻的三个边界点形成的夹角与预定阈值进行比较，若某一组相邻的三个边界点形成的夹角小于该预定阈值，则表示该夹角过小，因此，将该夹角的夹角顶点即位于中间位置的边界点识别为待处理近角点。若某一组相邻的三个边 界点形成的夹角大于或等于该预定阈值，则表示该夹角较大，该夹角的夹角顶点并不是近角点，可以不作处理。
需要说明的是，预定阈值可以是由本领域技术人员根据在先经验预先设定的，也可以是根据所欲识别的图形的不同形状设定不同的预定阈值，例如针对于四边形而言，预定阈值可以设定为177°，若所欲识别的图形为正八边形，则由于该图形的内角均为钝角，因此可以相应提高该预定阈值为179°等，以保证待处理近角点识别的准确性。以上预定阈值仅为示例性举例，本申请对此不作特殊限定。
在步骤S240中，根据相邻的所述待处理近角点的数量，从所述待处理近角点中识别出目标近角点，以根据所述目标近角点确定所述待识别图形的角点。
在本申请一示例性实施例中，应该理解的，近角点一般情况下应该是成对出现的，而且成对出现的近角点，应为相邻的边界点。由此，可以根据相邻的待处理近角点的数量，确定该待处理近角点是否为正确识别的近角点，从而从各待处理近角点中识别出目标近角点，以去除掉识别错误的待处理近角点，保证目标近角点识别的准确性，进而保证后续根据目标近角点确定的角点的准确性。
在本申请一示例性实施例中，所述根据相邻的所述待处理近角点的数量，从所述待处理近角点中识别出目标近角点，包括：
若所述待处理近角点未存在相邻的其他待处理近角点，则将与其相邻的两个边界点作为夹角顶点的夹角进行比较，确定所述待处理近角点以及与其相邻的两个边界点作为夹角顶点的夹角中夹角较大的夹角顶点为目标近角点。
在该实施例中，若某一待处理近角点未存在相邻的其他待处理近角点，则表示该待处理近角点较大可能性位于图形的角点上。因此，服务器可以将与其相邻的两个边界点作为夹角顶点的夹角进行比较，以将较大的夹角的夹角顶点以及该待处理近角点识别为目标近角点，从而保证目标近角点可以成对出现，以便于后续确定角点。
例如存在三个相邻的点集合分别为点集合A[a,b,c]、点集合B[b,c,d]以及点集合C[c,d,e]，其中，点集合B的夹角顶点c被识别为待处理近角点，而相邻的点集合A和点集合B的夹角顶点未被识别为待处理近角点，因 此，待处理近角点c未存在与其相邻的其他待处理近角点。服务器则可以将点集合A和点集合C对应的夹角进行比较，即将∠abc和∠cde进行比较，若∠abc大于∠cde，则将点集合A中的夹角顶点b和点集合B中的夹角顶点c识别为目标近角点；若∠abc小于∠cde，则将点集合B中的夹角顶点c和点集合C中的夹角顶点d识别为目标近角点；若∠abc等于∠cde，则可以将夹角顶点b或c中的任意一个以及点集合B中的夹角顶点c识别为目标近角点，等等。
在本申请一示例性实施例中，所述根据相邻的所述待处理近角点的数量，从所述待处理近角点中识别出目标近角点，包括：
若所述待处理近角点存在相邻的其他待处理近角点，所述待处理近角点连续存在，其中，连续的所述待处理近角点的数量为两个，则将两个连续的所述待处理近角点识别为目标近角点。
在该实施例中，若连续的待处理近角点的数量为两个，则该两个待处理近角点符合近角点成对出现的规律，因此，可以将该两个待处理近角点均识别为目标近角点，以备后续识别角点。
在本申请一示例性实施例中，所述根据相邻的所述待处理近角点的数量，从所述待处理近角点中识别出目标近角点，包括：
若所述待处理近角点存在相邻的其他待处理近角点，所述待处理近角点连续存在，其中，连续的所述待处理近角点的数量大于或等于三个，则将连续的所述待处理近角点中除两端之外的待处理近角点进行删除，以根据删除后的所述边界点信息进行目标近角点识别。
在该实施例中，若在边界点识别中，边界点识别错误即可能导致在待处理近角点识别时出现连续三个或者三个以上的待处理近角点。针对于此，若连续的待处理近角点的数量大于或等于三个，服务器可以将该三个或三个以上的待处理近角点中的两端点进行保留，并删除掉位于三个或三个以上的待处理近角点中间位置的待处理近角点，从而得到删除后的边界点信息。
例如存在四个连续的待处理近角点a、b、c和d，则将该四个待处理近角点的两端进行保留，即保留待处理近角点a和d，并将该位于中间位置的待处理近角点b和c进行删除，以更新待识别图形的边界点信息。
而在得到删除后的边界点信息之后，服务器可以根据删除后的边界点 信息，重新划分点集合，并根据重新划分后的点集合进行上述的待处理近角点识别过程，以识别出删除后的边界点信息所对应的待处理近角点，最后再根据新得到的待处理近角点进行上述的目标近角点识别过程。由此，可以防止因为边界点识别错误，从而导致三个或三个以上的待处理近角点连续出现的情况发生，保证了待处理近角点识别的准确性。
在本申请一示例性实施例中，所述位置信息包括坐标信息；则根据所述边界点的位置信息，确定每相邻的三个边界点形成的夹角，包括：
根据所述边界点的坐标信息，确定所述边界点的相邻关系；
根据所述边界点的坐标信息和所述相邻关系，确定每相邻的三个边界点形成的夹角。
在该实施例中，服务器可以根据包含待识别图形的图像进行图像识别，从而确定该待识别图形对应的边界点的坐标信息。且可以根据边界点的坐标信息，确定边界点的相邻关系。应该理解的，处于同一边界上的边界点的坐标信息中至少X坐标或者Y坐标的之间差值的绝对值处于一定的阈值范围内。本领域技术人员可以根据在先经验预先设定差值阈值，服务器可以计算任意两个边界点之间的X坐标或者Y坐标之间的差值的绝对值，将该差值的绝对值与预先设定的差值阈值进行比较，从而确定两个边界点是否处于同一边界上。
例如，边界点A的坐标信息为(6,8)，边界点B的坐标信息为(9，9)，则边界点A与边界点B之间的坐标值的差值的绝对值分别为3(X坐标的差值的绝对值)和1(Y坐标的差值的绝对值)，若预先设定的差值阈值为1，所以进行比较后，确定边界点A和边界点B的Y坐标的差值的绝对值小于或等于该差值阈值，所以，可以确定边界点A和边界点B为同一横边上的边界点，等等。
需要说明的，为了保证识别的准确度，本领域技术人员可以预先设定两个差值阈值，分别用作X坐标和Y坐标的比较，两个边界点的坐标信息之间的差值的绝对值均应小于或等于该差值阈值，由此可以保证所确定的两个边界点之间的距离较小，进而保证同一边界上的边界点的准确性。避免将分别位于两条横边上的边界点或者分别位于两条纵边上的边界点识别为同一边界上的边界点。
基于上述说明，应该理解的，服务器在确定边界点时是逐对进行确定 的，在将所有边界点确定完成之后，服务器可以通过链式的方式，将确定结果进行整合，例如，边界点A和边界点B为同一边界上的边界点，边界点B和边界点C为同一边界上的边界点，则边界点A、边界点B和边界点C均应为同一边界上的边界点，等等。
在整合之后，服务器还可以根据位于同一边界上的边界点的坐标信息进行比较，防止将相邻边界上的边界点识别为位于同一边界上，保证位于同一边界上的边界点的划分的准确性。
在确定位于同一边界上的边界点之后，服务器可以根据位于同一边界上的边界点的坐标信息进行计算，将分别位于某一边界点的两侧，且与该边界点之间距离最小的两个边界点，识别为该边界点的相邻边界点，即确定边界点之间的相邻关系。需要说明的，对于横边，则两侧指的是左侧和右侧，若为纵边，则两侧指的是上侧和下侧。服务器可以基于位于同一边界上的边界点的坐标信息确定为横边或者纵边，如横边上的边界点的坐标信息中X坐标的差值大于Y坐标上的差值，纵边上的边界点的坐标信息中X坐标的差值小于Y坐标上的差值，等等。
另外，针对于不同边界上的边界点，服务器可以将各边界上的端点识别为与其相邻的边界的端点的相邻边界点，由此可以保证每一边界点均具有两个相邻边界点，以备后续进行计算。
由此，根据所确定的边界点的坐标信息和相邻关系，可以确定每相邻的三个边界点所形成的夹角，具体的确定方式可以参照上文所述，本申请在此不再赘述。
基于图2所示的实施例，在本申请的一个示例性实施例中，所述边界点的位置信息包括坐标信息和相邻关系，则对包含待识别图形的图像进行图像识别，确定所述待识别图形对应的边界点的位置信息，包括：
以所述图像的中心点为坐标原点，按照顺时针或者逆时针的顺序识别所述待识别图形的边界点，确定所述待识别图形对应的边界点的坐标信息；
根据所述边界点的识别先后顺序，确定所述边界点的相邻关系。
在该实施例中，服务器可以以图像的中心点为坐标原点，按照顺时针或者逆时针的方向识别待识别图形的边界点，从而获取待识别图形的边界 点的坐标信息，并根据边界点的识别先后顺序，确定该边界点的相邻关系。
具体地，服务器可以以坐标原点为起点，按照顺时针或者逆时针的顺序，等间隔向外辐射多条射线。应该理解的，当射线延伸到待识别图形的边界时，由于边界两侧的灰度值差异较大，即可识别出待识别图形对应的边界点，并获取该边界点对应的坐标信息。
其中，该间隔可以是由本领域技术人员预先设定的，例如，服务器可以每间隔10°向外辐射一条射线，则一周应有36条射线即360°，每一条射线可以确定一个边界点，则可以对应获得36个边界点，等等。需要说明的，本领域技术人员可以根据实际实现需要确定对应的间隔大小，例如该间隔可以是12°、18°或者20°等等，以上仅为示例性举例，本申请对此不作特殊限定。
服务器按照一定顺序(例如顺时针方向或者逆时针方向)对待识别图形的边界点进行识别，获取每一边界点的坐标信息。同时，可以依次对识别得到的边界点进行编号，从而得到边界点之间的相邻关系。例如服务器依次识别得到边界点1、2、3、4、5、……、N-1、N，应该理解的，对于边界点2而言，边界点1和边界点3为其相邻的边界点，对于边界点3而言，边界点2和边界点4为其相邻的边界点，……，而由于服务器是按照顺时针方向或者逆时针方向进行识别。因此，在旋转一周后，对于最后识别得到的边界点N而言，其相邻的边界点应为边界点(N-1)和边界点1，而边界点1的相邻的边界点应为边界点N和边界点2。
由此，在识别边界点的坐标信息时，即可确定边界点的相邻关系，提高了后续的识别效率。
基于图2所示的实施例，图3示出了根据本申请的一个实施例的图像中图形角点的识别方法中确定角点的流程示意图。参照图3所示，目标近角点的数量为多个，则确定角点至少包括步骤S310至步骤S320，详细介绍如下：
在步骤S310中，将多个所述目标近角点按照相邻关系进行分组，得到目标近角点组合。
在该实施例中，应该理解的，根据上述的目标近角点识别过程，目标 近角点均是成对出现的且每一对目标近角点均是相邻的边界点。由此，服务器可以将相邻的目标近角点分为一组，从而得到多个目标近角点组合。例如存在八个目标近角点分别为A、B、C、D、E、F、G和H，其中A和B相邻、C和D相邻、E和F相邻，G和H相邻，则可以将A和B划分为一个目标近角点组合，C和D划分为一个目标近角点组合，……，等等。
在步骤S320中，根据各所述目标近角点组合中各目标近角点的位置关系，确定所述待识别图形的角点。
在该实施例中，由于每一目标近角点组合中的目标近角点分别位于角点的两侧，因此，根据各目标近角点组合中目标近角点的位置关系，确定待识别图形的角点。例如若待识别图形为矩形，则应该理解的，在相邻的两个目标近角点组合中存在两个目标近角点位于矩形的同一边界上，由此，可以根据该两个目标近角点确定待识别图形的一条边界，依次类推，可以确定待识别图形的四条边界，而四条边界的交点即为待识别图形的角点，等等。
由此，在图3所示的实施例中，通过对目标近角点进行分组，得到多个目标近角点组合，并根据多个目标近角点中各目标近角点之间的位置关系，确定待识别图形的角点位置，以保证待识别图形的角点位置识别的准确性。
基于图2和图3所示的实施例，图4示出了根据本申请的一个实施例的图3的图像中图形角点的识别方法中步骤S320的流程示意图。参照图4所示，步骤S320至少包括步骤S410至步骤S430，详细介绍如下：
在步骤S410中，按照顺时针或逆时针的顺序对各所述目标近角点组合以及各所述目标近角点分别进行编号，确定各所述目标近角点组合对应的第一编号以及各所述目标近角点对应的第二编号。
在该实施例中，服务器可以某一边界作为起点，例如坐标系中的X轴的正方向等，按照顺时针或者逆时针的顺序对各目标近角点组合以及各目标近角点分别进行编号，从而得到各目标近角点组合对应的第一编号以及各目标近角点对应的第二编号。
例如服务器按照顺时针顺序对目标近角点组合进行编号为A、B、C和 D，对应的，目标近角点组合A中的两个目标近角点可以按照顺序进行编号为1和2，而目标近角点组合B中的两个目标近角点可以按照顺序编号为3和4，……，以此可以得到各目标近角点组合对应的第一编号和各目标近角点对应的第二编号。
应该理解的，目标近角点组合以及目标近角点的编号是与服务器的编号方向相对应的，即当服务器进行编号后，目标近角点组合以及目标近角点的编号是与服务器的编号方向(例如顺时针方向或者逆时针方向)对应分布的。
在步骤S420中，根据所述第一编号相邻的两个所述目标近角点组合中的两个目标近角点，确定所述两个目标近角点之间的直线方程，所述两个目标近角点为所述第二编号相邻的目标近角点。
在该实施例中，应该理解的，在第一编号相邻的两个目标近角点组合中存在两个目标近角点位于同一边界上，且该两个目标近角点为第二编号相邻的目标近角点。例如在第一编号相邻的目标近角点组合A[a,b]和目标近角点组合B[c,d]中，目标近角点a、b、c和d对应的第二编号分别为1、2、3和4，则可以确定第二编号(即2和3)相邻的目标近角点b和c位于同一边界上，等等。
由此，根据第一编号相邻的两个目标近角点组合中的第二编号相邻的两个目标近角点，可以确定待识别图形的一条边界。因此，可以根据该两个目标近角点的坐标信息，确定所在边界的直线方程。其中，建立直线方可以参照现有的数学方法，本申请不再赘述。由此，基于上述过程，服务器可以确定待识别图形中的每条边界所对应的直线方程。
在步骤S430中，根据所述第一编号相邻的所述目标近角点组合所确定的直线方程，确定所述待识别图形的角点。
在本申请一示例性实施例中，应该理解的，第一编号相邻的两个目标近角点组合可以确定一个直线方程，而第一编号连续的三个目标近角点组合则可以确定两个直线方程，由于该两个直线方程所对应直线即为待识别图形的边界所在的直线，所以，可以将该两个直线方程对应的直线的角点确定为待识别图形的角点。
例如存在相邻的四个目标近角点组合为A、B、C和D，A和B可以确定一个直线方程a，B和C可以确定一个直线方程b，C和D可以确定一个直线方程c，D和A可以确定一个直线方程d，则直线方程a和直线方程b之间的交点、直线方程b和直线方程c之间的交点…即为待识别图形的角点。
由此，在图4所示的实施例中，通过第一编号相邻的两个目标近角点组合中第二编号相邻的两个目标近角点，可以确定待识别图形的边界所对应的直线方程，再通过所确定的直线方程进而确定待识别图形的角点，可以保证所确定的角点位置的准确性。
在本申请的其他实施例中，在以图像的中心点为坐标原点之后，服务器也可以通过计算每个目标近角点相对于坐标系中X轴的正方向的夹角，按照夹角从小到大的顺序，若两个目标近角点所对应的夹角差值较大，则表示该两个目标近角点位于同一边界上，因此可以根据夹角差值较大的两个目标近角点的坐标信息确定一个直线方程，再根据直线方程之间的交点确定待识别图形的角点。
其中，判断两个目标近角点的夹角差值是否较大，可以将该夹角差值与预先设定的阈值进行比较，也可以是将其与其他相邻的目标近角点的夹角差值进行比较，若明显大于(例如夹角差值为与其相邻的目标近角点的夹角差值的三倍以上)则可以确定该夹角差值较大。本领域技术人员可以根据实际实现需要采用对应的判断方法，本申请对此不作特殊限定。
基于图2所示的实施例，在本申请另一示例性实施例中，所述目标近角点的数量为多个，则所述根据所述目标近角点确定所述待识别图形的角点，包括：
根据所述目标近角点的位置信息，确定位于同一边界上的目标近角点；
根据位于同一边界上的目标近角点，确定所述待识别图形的角点。
在该实施例中，服务器可以根据边界点的坐标信息和相邻关系，确定位于同一边界上的目标近角点。例如可以根据在先确定的位于同一边界上的边界点的划分结果(该划分结果可以包含位于该边界上的边界点的坐标 信息)，将目标近角点的坐标信息与该划分结果进行比对，从而确定目标近角点属于哪一边界，并确定位于同一边界上的目标近角点。应该理解的，目标近角点应成对位于一个边界上，且分别位于该边界的两端位置，所以服务器可以根据位于同一边界上的目标近角点的坐标信息，建立该边界的直线方程。
由此，服务器可以确定四条边界所对应的直线方程，并将直线方程之间的交点识别为该待识别图形的角点。
此外，服务器也可以根据目标近角点的坐标信息，确定两个目标近角点是否位于同一边界上。具体地，若两个目标近角点位于同一边界上，则该两个目标近角点至少在一个坐标值的差值的绝对值较小(可以根据上文所述的差值阈值进行比较)，而另一个坐标值差值的绝对值不为最小也不为最大，从而确定位于同一边界上的目标近角点。
例如存在四个目标近角点分别为A(3,8)、B(5,10)、H(2,3)和G(3,2)，在X轴方向上，B、H、G都与A较近，因此均有可能与A点位于同一边界上。但是在Y轴方向上，与A点处于同一边界的点，肯定是在其他两点之间的。而H点位于B点和G两点之间，即在Y轴方向上的坐标值的差值的绝对值不为最大也不为最小。
基于上述举例，服务器可以根据A点的坐标信息确定在某一坐标值上与A的坐标值之间的差值的绝对值较小的目标近角点，即点B、点H和点G，视为点B、G、H备选为A点同边界的点。再比较该三点与A点的Y坐标值差值绝对值，此时可以确认H点的坐标值满足条件，2＜5＜6。此外，本领域技术人员也可以设置确定同一边界点的差值阈值，如果误差较小，则该阈值可以设置较小，以排除其他点的干扰。
基于上述实施例的技术方案，以下介绍本申请实施例的一个具体应用场景：
图5a至图5c示出了根据本申请的一个实施例的图像中图形角点的识别方法的流程示意图(以下以待识别图形为测试图为例进行说明)。
请参考图5a所示，服务器可以以图像的中心点作为坐标原点，按照顺时针顺序，等间隔向外辐射多条射线，根据每一射线沿路的灰度值变化确 定测试图的边界点，从而获取该边界点坐标信息，并依次对识别得到的边界点进行编号A、B、C、D、…、F 1和G 1，从而得到边界点之间的相邻关系。
根据边界点之间的相邻关系，服务器将每三个相邻的边界点划分成一个点集合，例如边界点A、B和C组成一个点集合，边界点B、C和D组成一个点集合…，等等。并根据每一点集合中的三个边界点的坐标信息，以位于中间位置的边界点作为夹角顶点，计算每一点集合对应的夹角，从而得到∠ABC、∠BCD、∠CDE、…、∠F 1G 1A、∠G 1AB的大小。
将每一点集合对应的夹角与预定阈值进行比较，将夹角小于预定阈值的点集合的夹角顶点识别为待处理近角点，由此可以得到待处理近角点为A、J、K、L、M、N、S、T、U、V、Z、A 1。
再计算连续的待处理近角点的数量(即相邻的待处理近角点数量)，并根据连续的待处理近角点数量确定目标近角点。如图5中510所示，由于待处理近角点A位于图形的角点位置上，所以其未存在相邻的待处理近角点。因此，服务器可以获取与待处理边界点A相邻的两个点集合对应的夹角即∠F 1G 1A和∠ABC，将二者进行比较，从而将夹角较大的点集合的夹角顶点与待处理近角点A识别为目标近角点，如∠F 1G 1A大于∠ABC，则可以确定点G 1和点A为目标近角点。
如图5中520和530所示，由于边界点M、T和U出现识别错误，与测试图的边界存在较大偏离，则在待处理近角点的识别过程中，会将边界点L、M、N、S、T、U和V均识别为待处理近角点，因此会出现三个或三个以上连续的待处理近角点。
对此，服务器可以将三个或三个以上连续的待处理近角点中除两端以外的待处理近角点进行去除，即去除边界点M、T和U，从而将识别错误的边界点进行删除，得到如图6所示的更新后的边界点信息。
服务器再根据更新后的边界点信息进行待处理近角点识别，得到新的待处理近角点为A、J、K、Q、R、Z、A 1和G 1(如图7所示)。由于待处理近角点均为成对出现，所以上述待处理近角点均可以识别为目标近角点。
服务器可以根据相邻的两个目标近角点组合中相邻的两个目标近角点的坐标信息确定一个直线方程，该直线方程即与测试图的边界相对应。例如目标近角点A和目标近角点J确定一个直线方程、目标近角点K和目标近角点Q确定给一个直线方程等。所确定的直线方程之间的交点即为测试图的角点。由此可以快速识别出测试图的角点，并保证角点位置识别的准确性。
图6示出了根据本申请的一个实施例的图像中图形角点的识别方法中确定角点的流程框图。
如图6所示，服务器可以按照一定顺序对边界点进行识别，若确定某一边界点为待处理近角点，则将连续待处理近角点数量置为1，再对下一边界点进行识别，若下一边界点为待处理近角点，则将连续待处理近角点数量累计加一，并继续判断下一边界点是否为待处理近角点。若不是，则对连续待处理近角点数量进行判断，若连续待处理近角点数量等于2，则表示该组待处理近角点正常，可以将其识别为目标近角点。若连续待处理近角点数量等于1，则表示该待处理近角点可能位于角点上，可以采用处理模式1进行处理，即可以采用上述未存在相邻的待处理近角点的情况，识别出另一个待处理近角点。
若连续待处理近角点数量大于或等于三，则可以采用处理模式2，即参照上述连续的待处理近角点的数量大于或等于三个的情况进行处理，将除两端之外的待处理近角点进行删除，得到更新后的边界点信息，再根据更新后的边界点信息重新进行待处理近角点识别。
当遍历完成且待处理近角点均是成对出现后，则可以根据所识别得到的目标近角点，确定对应的直线方程，并计算直线方程之间的交点以作为待识别图形的角点。
由此，根据对连续的待处理近角点的数量，可以从待处理近角点中去除识别错误的待处理近角点，从而得到目标近角点，保证了目标近角点的识别的准确性，进而保证了后续角点位置确定的准确性。
以下介绍本申请的装置实施例，可以用于执行本申请上述实施例中的图像中图形角点的识别方法。对于本申请装置实施例中未披露的细节，请 参照本申请上述的图像中图形角点的识别方法的实施例。
图7示出了根据本申请的一个实施例的图像中图形角点的识别装置的框图。
参照图7所示，根据本申请的一个实施例的图像中图形角点的识别装置，包括：
图像识别模块710，用于对包含待识别图形的图像进行图像识别，确定所述待识别图形对应的边界点信息，所述边界点信息包括边界点的坐标信息以及所述边界点之间的相邻关系；
计算模块720，用于根据所述边界点的坐标信息以及相邻关系进行计算，确定各点集合中的边界点形成的夹角，所述点集合由相邻的三个所述边界点所组成，且位于中间位置的所述边界点为夹角顶点；
近角点识别模块730，用于根据各所述点集合对应的夹角，从各所述点集合的夹角顶点中识别出待处理近角点；
角点确定模块740，用于根据相邻的所述待处理近角点的数量，从所述待处理近角点中识别出目标近角点，以根据所述目标近角点确定所述待识别图形的角点。
上述图像中图形角点的识别装置中各模块的具体细节已经在对应的图像中图形角点的识别方法中进行了详细的描述，因此此处不再赘述。
应当注意，尽管在上文详细描述中提及了用于动作执行的设备的若干模块或者单元，但是这种划分并非强制性的。实际上，根据本申请的实施方式，上文描述的两个或更多模块或者单元的特征和功能可以在一个模块或者单元中具体化(例如具有图像识别、计算、近角点识别和角点确定功能的处理器等)。反之，上文描述的一个模块或者单元的特征和功能可以进一步划分为由多个模块或者单元来具体化。
图8示出了适于用来实现本申请实施例的电子设备的计算机系统的结构示意图。
需要说明的是，图8示出的电子设备的计算机系统仅是一个示例，不应对本申请实施例的功能和使用范围带来任何限制。
如图8所示，计算机系统包括中央处理单元(Central Processing Unit，CPU)801，其可以根据存储在只读存储器(Read-Only Memory，ROM) 802中的程序或者从储存部分808加载到随机访问存储器(Random Access Memory，RAM)803中的程序而执行各种适当的动作和处理，例如执行上述实施例中所述的方法。在RAM 803中，还存储有系统操作所需的各种程序和数据。CPU 801、ROM 802以及RAM 803通过总线804彼此相连。输入/输出(Input/Output，I/O)接口805也连接至总线804。
以下部件连接至I/O接口805：包括键盘、鼠标等的输入部分806；包括诸如阴极射线管(Cathode Ray Tube，CRT)、液晶显示器(Liquid Crystal Display，LCD)等以及扬声器等的输出部分807；包括硬盘等的储存部分808；以及包括诸如LAN(Local Area Network，局域网)卡、调制解调器等的网络接口卡的通信部分809。通信部分809经由诸如因特网的网络执行通信处理。驱动器810也根据需要连接至I/O接口805。可拆卸介质811，诸如磁盘、光盘、磁光盘、半导体存储器等等，根据需要安装在驱动器810上，以便于从其上读出的计算机程序根据需要被安装入储存部分808。
特别地，根据本申请的实施例，上文参考流程图描述的过程可以被实现为计算机软件程序。例如，本申请的实施例包括一种计算机程序产品，其包括承载在计算机可读介质上的计算机程序，该计算机程序包含用于执行流程图所示的方法的计算机程序。在这样的实施例中，该计算机程序可以通过通信部分809从网络上被下载和安装，和/或从可拆卸介质811被安装。在该计算机程序被中央处理单元(CPU)801执行时，执行本申请的系统中限定的各种功能。
需要说明的是，本申请实施例所示的计算机可读介质可以是计算机可读信号介质或者计算机可读存储介质或者是上述两者的任意组合。计算机可读存储介质例如可以是——但不限于——电、磁、光、电磁、红外线、或半导体的系统、装置或器件，或者任意以上的组合。计算机可读存储介质的更具体的例子可以包括但不限于：具有一个或多个导线的电连接、便携式计算机磁盘、硬盘、随机访问存储器(RAM)、只读存储器(ROM)、可擦式可编程只读存储器(Erasable Programmable Read Only Memory，EPROM)、闪存、光纤、便携式紧凑磁盘只读存储器(Compact Disc Read- Only Memory，CD-ROM)、光存储器件、磁存储器件、或者上述的任意合适的组合。在本申请中，计算机可读存储介质可以是任何包含或存储程序的有形介质，该程序可以被指令执行系统、装置或者器件使用或者与其结合使用。而在本申请中，计算机可读的信号介质可以包括在基带中或者作为载波一部分传播的数据信号，其中承载了计算机可读的计算机程序。这种传播的数据信号可以采用多种形式，包括但不限于电磁信号、光信号或上述的任意合适的组合。计算机可读的信号介质还可以是计算机可读存储介质以外的任何计算机可读介质，该计算机可读介质可以发送、传播或者传输用于由指令执行系统、装置或者器件使用或者与其结合使用的程序。计算机可读介质上包含的计算机程序可以用任何适当的介质传输，包括但不限于：无线、有线等等，或者上述的任意合适的组合。
附图中的流程图和框图，图示了按照本申请各种实施例的系统、方法和计算机程序产品的可能实现的体系架构、功能和操作。其中，流程图或框图中的每个方框可以代表一个模块、程序段、或代码的一部分，上述模块、程序段、或代码的一部分包含一个或多个用于实现规定的逻辑功能的可执行指令。也应当注意，在有些作为替换的实现中，方框中所标注的功能也可以以不同于附图中所标注的顺序发生。例如，两个接连地表示的方框实际上可以基本并行地执行，它们有时也可以按相反的顺序执行，这依所涉及的功能而定。也要注意的是，框图或流程图中的每个方框、以及框图或流程图中的方框的组合，可以用执行规定的功能或操作的专用的基于硬件的系统来实现，或者可以用专用硬件与计算机指令的组合来实现。
描述于本申请实施例中所涉及到的单元可以通过软件的方式实现，也可以通过硬件的方式来实现，所描述的单元也可以设置在处理器中。其中，这些单元的名称在某种情况下并不构成对该单元本身的限定。
作为另一方面，本申请还提供了一种计算机可读介质，该计算机可读介质可以是上述实施例中描述的电子设备中所包含的；也可以是单独存在，而未装配入该电子设备中。上述计算机可读介质承载有一个或者多个程序，当上述一个或者多个程序被一个该电子设备执行时，使得该电子设备实现上述实施例中所述的方法。
应当注意，尽管在上文详细描述中提及了用于动作执行的设备的若干模块或者单元，但是这种划分并非强制性的。实际上，根据本申请的实施方式，上文描述的两个或更多模块或者单元的特征和功能可以在一个模块或者单元中具体化。反之，上文描述的一个模块或者单元的特征和功能可以进一步划分为由多个模块或者单元来具体化。
通过以上的实施方式的描述，本领域的技术人员易于理解，这里描述的示例实施方式可以通过软件实现，也可以通过软件结合必要的硬件的方式来实现。因此，根据本申请实施方式的技术方案可以以软件产品的形式体现出来，该软件产品可以存储在一个非易失性存储介质(可以是CD-ROM，U盘，移动硬盘等)中或网络上，包括若干指令以使得一台计算设备(可以是个人计算机、服务器、触控终端、或者网络设备等)执行根据本申请实施方式的方法。
本领域技术人员在考虑说明书及实践这里公开的实施方式后，将容易想到本申请的其它实施方案。本申请旨在涵盖本申请的任何变型、用途或者适应性变化，这些变型、用途或者适应性变化遵循本申请的一般性原理并包括本申请未公开的本技术领域中的公知常识或惯用技术手段。
应当理解的是，本申请并不局限于上面已经描述并在附图中示出的精确结构，并且可以在不脱离其范围进行各种修改和改变。本申请的范围仅由所附的权利要求来限制。

Claims (11)

  1. 一种图像中图形角点的识别方法，其特征在于，包括：
    对包含待识别图形的图像进行图像识别，确定所述待识别图形对应的边界点的位置信息；
    根据所述边界点的位置信息，确定每相邻的三个边界点形成的夹角，其中，位于中间位置的所述边界点为夹角顶点；
    根据每相邻的三个边界点形成的夹角，从各所述夹角的夹角顶点中识别出待处理近角点；
    根据相邻的所述待处理近角点的数量，从所述待处理近角点中识别出目标近角点，以根据所述目标近角点确定所述待识别图形的角点。
  2. 根据权利要求1所述的识别方法，其特征在于，所述位置信息包括坐标信息；
    所述根据所述边界点的位置信息，确定每相邻的三个边界点形成的夹角，包括：
    根据所述边界点的坐标信息，确定所述边界点的相邻关系；
    根据所述边界点的坐标信息和所述相邻关系，确定每相邻的三个边界点形成的夹角。
  3. 根据权利要求1所述的识别方法，其特征在于，所述位置信息包括坐标信息和相邻关系；
    对包含待识别图形的图像进行图像识别，确定所述待识别图形对应的边界点的位置信息，包括；
    以所述图像的中心点为坐标原点，按照顺时针或者逆时针的顺序识别所述待识别图形的边界点，确定所述待识别图形对应的边界点的坐标信息；
    根据所述边界点的识别先后顺序，确定所述边界点的相邻关系。
  4. 根据权利要求1所述的识别方法，其特征在于，所述根据相邻的所述待处理近角点的数量，从所述待处理近角点中识别出目标近角点，包括：
    若所述待处理近角点未存在相邻的其他待处理近角点，则将与其相邻的两个边界点作为夹角顶点的夹角进行比较，确定所述待处理近角点以及 与其相邻的两个边界点作为夹角顶点的夹角中夹角较大的夹角顶点为目标近角点。
  5. 根据权利要求1所述的识别方法，其特征在于，所述根据相邻的所述待处理近角点的数量，从所述待处理近角点中识别出目标近角点，包括：
    若所述待处理近角点存在相邻的其他待处理近角点，所述待处理近角点连续存在，其中，连续的所述待处理近角点的数量为两个，则将两个连续的所述待处理近角点识别为目标近角点。
  6. 根据权利要求1所述的识别方法，其特征在于，所述根据相邻的所述待处理近角点的数量，从所述待处理近角点中识别出目标近角点，包括：
    若所述待处理近角点存在相邻的其他待处理近角点，所述待处理近角点连续存在，其中，连续的所述待处理近角点的数量大于或等于三个，则将连续的所述待处理近角点中除两端之外的待处理近角点进行删除，以根据删除后的所述边界点信息进行目标近角点识别。
  7. 根据权利要求1所述的识别方法，其特征在于，所述根据每相邻的三个边界点形成的夹角，从各所述夹角的夹角顶点中识别出待处理近角点，包括：
    若相邻的三个边界点形成的夹角小于预定阈值，则将所述夹角的夹角顶点识别为待处理近角点。
  8. 根据权利要求1所述的识别方法，其特征在于，所述目标近角点的数量为多个，则所述根据所述目标近角点确定所述待识别图形的角点，包括：
    根据所述目标近角点的位置信息，确定位于同一边界上的目标近角点；
    根据位于同一边界上的目标近角点，确定所述待识别图形的角点。
  9. 一种图像中图形角点的识别装置，其特征在于，包括：
    图像识别模块，用于对包含待识别图形的图像进行图像识别，确定所述待识别图形对应的边界点的位置信息；
    计算模块，用于根据根据所述边界点的位置信息，确定每相邻的三个边界点形成的夹角，其中，位于中间位置的所述边界点为夹角顶点；
    近角点识别模块，用于根据每相邻的三个边界点形成的夹角，从各所述夹角的夹角顶点中识别出待处理近角点；
    角点确定模块，用于根据相邻的所述待处理近角点的数量，从所述待处理近角点中识别出目标近角点，以根据所述目标近角点确定所述待识别图形的角点。
  10. 一种计算机可读介质，其上存储有计算机程序，其特征在于，所述计算机程序被处理器执行时实现如权利要求1至8中任一项所述的图像中图形角点的识别方法。
  11. 一种电子设备，其特征在于，包括：
    一个或多个处理器；
    存储装置，用于存储一个或多个程序，当所述一个或多个程序被所述一个或多个处理器执行时，使得所述一个或多个处理器实现如权利要求1至8中任一项所述的图像中图形角点的识别方法。
PCT/CN2022/071616 2021-01-29 2022-01-12 图像中图形角点的识别方法、装置、介质及电子设备 WO2022161172A1 (zh)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN202110126172.3A CN114913216A (zh) 2021-01-29 2021-01-29 图像中图形角点的识别方法、装置、介质及电子设备
CN202110126172.3 2021-01-29

Publications (1)

Publication Number Publication Date
WO2022161172A1 true WO2022161172A1 (zh) 2022-08-04

Family

ID=82653046

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2022/071616 WO2022161172A1 (zh) 2021-01-29 2022-01-12 图像中图形角点的识别方法、装置、介质及电子设备

Country Status (2)

Country Link
CN (1) CN114913216A (zh)
WO (1) WO2022161172A1 (zh)

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN115294078A (zh) * 2022-08-11 2022-11-04 创新奇智(合肥)科技有限公司 玻璃非对称切角识别方法、装置、设备及存储介质
CN116862908A (zh) * 2023-09-01 2023-10-10 菏泽学院 一种基于图像处理的金属齿轮啮合缺陷检测方法

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US5933523A (en) * 1997-03-18 1999-08-03 Cognex Corporation Machine vision method and apparatus for determining the position of generally rectangular devices using boundary extracting features
US20140232893A1 (en) * 2012-11-26 2014-08-21 Pixart Imaging, Inc. Image sensor and operating method thereof
CN108898147A (zh) * 2018-06-27 2018-11-27 清华大学 一种基于角点检测的二维图像边缘平直化方法、装置
CN110490887A (zh) * 2019-06-26 2019-11-22 河南埃尔森智能科技有限公司 一种基于3d视觉的对矩形包裹边缘快速识别定位方法
CN110753213A (zh) * 2018-11-07 2020-02-04 成都极米科技股份有限公司 一种梯形校正方法、系统、投影设备及存储介质

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US5933523A (en) * 1997-03-18 1999-08-03 Cognex Corporation Machine vision method and apparatus for determining the position of generally rectangular devices using boundary extracting features
US20140232893A1 (en) * 2012-11-26 2014-08-21 Pixart Imaging, Inc. Image sensor and operating method thereof
CN108898147A (zh) * 2018-06-27 2018-11-27 清华大学 一种基于角点检测的二维图像边缘平直化方法、装置
CN110753213A (zh) * 2018-11-07 2020-02-04 成都极米科技股份有限公司 一种梯形校正方法、系统、投影设备及存储介质
CN110490887A (zh) * 2019-06-26 2019-11-22 河南埃尔森智能科技有限公司 一种基于3d视觉的对矩形包裹边缘快速识别定位方法

Cited By (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN115294078A (zh) * 2022-08-11 2022-11-04 创新奇智(合肥)科技有限公司 玻璃非对称切角识别方法、装置、设备及存储介质
CN115294078B (zh) * 2022-08-11 2024-01-30 创新奇智(合肥)科技有限公司 玻璃非对称切角识别方法、装置、设备及存储介质
CN116862908A (zh) * 2023-09-01 2023-10-10 菏泽学院 一种基于图像处理的金属齿轮啮合缺陷检测方法
CN116862908B (zh) * 2023-09-01 2023-11-24 菏泽学院 一种基于图像处理的金属齿轮啮合缺陷检测方法

Also Published As

Publication number Publication date
CN114913216A (zh) 2022-08-16

Similar Documents

Publication Publication Date Title
WO2022161172A1 (zh) 图像中图形角点的识别方法、装置、介质及电子设备
CN112348815B (zh) 图像处理方法、图像处理装置以及非瞬时性存储介质
CN113869293B (zh) 车道线识别方法、装置、电子设备和计算机可读介质
CN107169489B (zh) 倾斜图像校正的方法和装置
CN113194004B (zh) 网络拓扑构建方法与装置、网络变更处理方法与装置
CN105069453A (zh) 一种图像校正方法及装置
CN110390327B (zh) 前景提取方法、装置、计算机设备及存储介质
WO2020107326A1 (zh) 车道线检测方法、设备、计算机可读存储介质
CN115540894B (zh) 车辆轨迹规划方法、装置、电子设备和计算机可读介质
WO2019128495A1 (zh) 图像清晰度检测方法、装置、存储介质及电子设备
WO2019080702A1 (zh) 图像处理方法和装置
WO2023040113A1 (zh) 图模的智能复核方法及装置
CN103837135B (zh) 工件检测方法及其系统
US9613288B2 (en) Automatically identifying and healing spots in images
CN111783777B (zh) 图像处理方法、装置、电子设备和计算机可读介质
WO2022095318A1 (zh) 字符检测方法、装置、电子设备、存储介质及程序
CN113516739B (zh) 一种动画处理方法、装置、存储介质及电子设备
CN107239776B (zh) 倾斜图像校正的方法和装置
CN111583376B (zh) 光照贴图中黑边的消除方法及装置、存储介质及电子设备
CN111401229B (zh) 视觉小目标自动标注方法、装置及电子设备
CN112465692A (zh) 图像处理方法、装置、设备及存储介质
CN113628208B (zh) 船舶检测方法、装置、电子设备和计算机可读介质
CN114219946B (zh) 文本图像的二值化方法及装置、电子设备和介质
CN117952928A (zh) 图像处理方法及装置
CN117975772A (zh) 基于物体拼摆的互动方法、装置、设备及存储介质

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 22745050

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 22745050

Country of ref document: EP

Kind code of ref document: A1