Motorcycle cylinder cover hole position deformation side detection device and method

By analyzing the edge profile characteristics and Hough circle matching in the motorcycle cylinder head hole position side image, identifying whether the cylinder head hole position has deformed, solving the problem of identifying errors in traditional methods and significantly improving the accuracy of detection.

CN119090891AActive Publication Date: 2024-12-06YUEQING FEIYA MASCH CO LTD
View PDF 7 Cites 0 Cited by

Patent Information

Application Number
CN202411587736.3
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-11-08
Publication Date
2024-12-06
Estimated Expiration
2044-11-08

AI Technical Summary

Technical Problem

When traditional image recognition methods detect the deformation of the motorcycle cylinder head hole position due to the small deformation, it is often impossible to accurately distinguish the deformation hole position and the undeformed hole position, resulting in identification errors.

Method used

By collecting the image of the hole position side of the cylinder head, the edge contour is constructed using the shape characteristics of the edge contour, and the deformation discrimination value is constructed based on the overlap difference between the hole edge contour and the matching Hough circle, and the hole edge contour is identified to identify the hole edge contour Whether deformation has occurred.

Benefits of technology

The accuracy of the detection of the hole position deformation of the motorcycle cylinder head is improved, and the deformation and undeformed hole position edge profile can be accurately distinguished.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN119090891A_ABST
    Figure CN119090891A_ABST
Patent Text Reader

Abstract

The invention relates to the technical field of image recognition, in particular to a motorcycle cylinder cover hole site deformation side detection device and method, and the method comprises the steps: collecting a motorcycle cylinder cover hole site side image; based on the shape features of the edge contour detected in the motorcycle cylinder cover hole site side image, constructing the edge contour quasi-circle value of the edge contour, and screening out the hole site edge contour by using the edge contour quasi-circle value; according to the center distance and radius difference between the hole site edge contour and the Hough circle, constructing an edge contour Hough matching degree, and screening out a matched Hough circle of each hole site edge contour by using the edge contour Hough matching degree; and according to the overlapping difference between the hole site edge contour and the edge and surrounding area between the hole site edge contour and the matched Hough circle, constructing a hole site deformation discrimination value of the hole site edge contour, and identifying whether the hole site edge contour is deformed or not. The invention aims to improve the accuracy of motorcycle cylinder cover hole side deformation identification.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present application relates to the field of image recognition technology, and in particular to a device and method for detecting the deformation side of a motorcycle cylinder head hole. Background Art

[0002] The function of a motorcycle cylinder head is to seal the cylinder and form a closed combustion space with the piston. There are many holes on the cylinder head, such as the intake hole, exhaust hole, spark plug installation hole, etc., and each of these holes has an important function.

[0003] During the production process of motorcycle cylinder heads, the holes of the cylinder heads may be deformed due to factors such as extrusion and collision. When the holes of the cylinder heads are deformed, it may cause problems such as cylinder leakage, oil leakage, or reduced engine performance. In order to prevent the deformation of the cylinder head holes from having an adverse effect on the performance of the motorcycle, it is necessary to identify the deformed holes during the production process. When detecting the deformation of the cylinder head holes, traditional image recognition methods are often unable to accurately distinguish between deformed holes and undeformed holes because the deformation of the holes is very small, resulting in recognition errors. Summary of the invention

[0004] In order to solve the above technical problems, the present application provides a motorcycle cylinder head hole position deformation side detection device and method, and the technical solutions adopted are as follows: In a first aspect, an embodiment of the present application provides a method for detecting the deformation side of a motorcycle cylinder head hole, the method comprising the following steps: Collect the side image of the motorcycle cylinder head hole; The edge contour features in the side image of the motorcycle cylinder head hole are used to identify whether the hole edge contour is deformed. Specifically: A1, based on the shape features of the edge contour detected in the side image of the motorcycle cylinder head hole, the edge contour circle-like value of the edge contour is constructed, and the edge contour of the hole is screened out using the edge contour circle-like value; A2: Based on the center distance and radius difference between the hole edge contour and the Hough circle, the edge contour Hough matching degree is constructed, and the matching Hough circle of each hole edge contour is screened out using the edge contour Hough matching degree; A3, based on the edge between the hole edge contour and its matching Hough circle, and the overlapping difference between the hole edge contour and the enclosed area of ​​the matching Hough circle, a hole deformation discrimination value of the hole edge contour is constructed to identify whether the hole edge contour is deformed.

[0005] Preferably, the shape features include the curvature of edge points on the edge contour and the overall distribution of gradient directions.

[0006] Preferably, the method for screening the hole edge profile is: Get the segmentation threshold value in the edge contour circle-like value of all edge contours; The edge contour corresponding to the circle-like value of the edge contour greater than the segmentation threshold is taken as the hole edge contour.

[0007] Preferably, the center coordinates of the hole edge contour used in calculating the center distance are determined by the center of the minimum circumscribed rectangle of the hole edge contour.

[0008] Preferably, the radius of the hole edge profile used in calculating the radius difference is determined by half of the average value of the distance between the two edge points that are farthest apart in the longitudinal direction and the distance between the two edge points that are farthest apart in the transverse direction on the hole edge profile.

[0009] Preferably, the Hough circle is obtained by using a Hough circle detection algorithm from a binary image of a motorcycle cylinder head hole side image.

[0010] Preferably, the matching Hough circle of each hole edge contour is: a Hough circle corresponding to the maximum Hough matching degree of the edge contour calculated with the hole edge contour.

[0011] Preferably, the construction expression of the hole position deformation discrimination value is: ; In the formula, represents the hole position deformation discrimination value of the jth hole position edge contour; It represents the ratio of the number of non-overlapping pixels to the number of overlapping pixels of the edge contour of the j-th hole position and all edge pixels of its matching Hough circle; represents the area of ​​the region enclosed by the matching Hough circle of the jth hole edge contour pixel point, It represents the area of ​​the region enclosed by the edge contour of the jth hole; exp is an exponential function with a natural constant as the base; iou is the intersection-union ratio function.

[0012] Preferably, the identifying whether the hole edge contour is deformed comprises: when a hole deformation discrimination value of the hole edge contour is greater than a preset threshold, identifying the hole edge contour as a deformed hole; otherwise, identifying the hole edge contour as a non-deformed hole.

[0013] In the second aspect, another embodiment of the present application provides a motorcycle cylinder head hole deformation side detection device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the steps of the above-mentioned motorcycle cylinder head hole deformation side detection method when executing the computer program.

[0014] The embodiments of the present application have at least the following beneficial effects: The present application constructs an edge contour circle-like value based on the shape feature of the edge contour, screens out the hole edge contour, and is used to accurately distinguish the hole edge contour from the non-hole edge contour; constructs the edge contour Hough matching degree based on the center distance and radius difference between the hole edge contour and the Hough circle, and is used to characterize the matching degree between a hole edge contour and a Hough circle, thereby making the Hough circle matching result more accurate; constructs a hole deformation discrimination value based on the overlapping difference between the edge between the hole edge contour and its matching Hough circle, and the enclosed area between the hole edge contour and its matching Hough circle, and is used to analyze the shape difference before and after the hole deformation, and judge whether a hole is deformed, thereby greatly improving the accuracy of the hole deformation detection of the motorcycle cylinder head. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] In order to more clearly illustrate the technical solutions and advantages in the embodiments of the present application or the prior art, the drawings required for use in the embodiments or the prior art descriptions are briefly introduced below. 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 creative work.

[0016] Figure 1 A flowchart of a method for detecting the deformation side of a motorcycle cylinder head hole provided by an embodiment of the present application; Figure 2 A flowchart of a process for determining whether a hole edge profile is deformed according to an embodiment of the present application; Figure 3 A schematic diagram of a grayscale image of a side image of a motorcycle cylinder head hole at a first angle provided by an embodiment of the present application; Figure 4 A schematic diagram of a grayscale image of a side image of a motorcycle cylinder head hole at a second angle provided by an embodiment of the present application; Figure 5 A schematic diagram of a grayscale image of a motorcycle cylinder head hole side image at a third angle provided by an embodiment of the present application; Figure 6 A schematic grayscale image of a side image of a motorcycle cylinder head hole at a fourth angle provided in an embodiment of the present application. DETAILED DESCRIPTION

[0017] In order to further explain the technical means and effects adopted by the present application to achieve the predetermined invention purpose, the following is a detailed description of a motorcycle cylinder head hole position deformation side detection device and method proposed in the present application, its specific implementation, structure, features and effects, in combination with the accompanying drawings and preferred embodiments. In the following description, different "one embodiment" or "another embodiment" does not necessarily refer to the same embodiment. In addition, specific features, structures or characteristics in one or more embodiments may be combined in any suitable form.

[0018] Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs.

[0019] The specific scheme of a motorcycle cylinder head hole position deformation side detection device and method provided by the present application is described in detail below with reference to the accompanying drawings.

[0020] See also Figure 1 , which shows a flowchart of a method for detecting the deformation side of a motorcycle cylinder head hole provided by an embodiment of the present application, the method comprising the following steps: The first step is to capture the side image of the motorcycle cylinder head hole.

[0021] First, the motorcycle cylinder heads to be inspected are placed on the conveyor belt one by one, and the high-resolution camera fixed above the conveyor belt is used to collect the side image of the hole position of the motorcycle cylinder head. During the collection process, sufficient ambient light is ensured, and the camera lens is aimed at the side with the hole position in the cylinder head. The collected image is an RGB image.

[0022] Since the camera may cause image deformation when shooting, such as barrel distortion and pincushion distortion, a distortion correction algorithm is used to correct the image to eliminate the image deformation. The distortion correction algorithm is a well-known technology and the specific process will not be repeated here.

[0023] At this point, the side image of the motorcycle cylinder head hole position can be acquired.

[0024] The second step is to use the edge contour features in the side image of the motorcycle cylinder head hole to identify whether the hole edge contour is deformed.

[0025] In the production process of motorcycle cylinder heads, the causes of hole deformation include excessive extrusion during the die pressing process, irregular operation process, etc. The edge contour of a normal hole is a standard circular edge contour. If the hole is deformed, the deformed hole edge contour is similar to an elliptical edge contour or an irregular circular edge contour. In order to detect the deformed hole, this embodiment compares the difference between the hole edge contour and the Hough circle in the hole side image of the motorcycle cylinder head to construct a hole deformation discrimination value for detecting the deformed hole. The flowchart of the judgment process of whether the hole edge contour is deformed is shown in the attached figure. Figure 2 As shown, the details are as follows: A1, based on the shape features of the edge contour detected in the side image of the motorcycle cylinder head hole, the edge contour circle-like value of the edge contour is constructed, and the edge contour of the hole is screened out using the edge contour circle-like value.

[0026] The collected motorcycle cylinder head hole side image is converted into a grayscale image, and the grayscale image is used as the input of the edge detection algorithm to obtain a binary image of the grayscale image. The edge points of the binary image are marked in white. At the same time, the gradient direction and gradient amplitude of each edge point can be obtained. Each gradient direction is rounded to an integer, so the gradient direction of the edge point is 360 directions within [1°, 360°].

[0027] In the embodiment of the present application, the grayscale images of the side images of the motorcycle cylinder head hole at four different angles are respectively as shown in the attached Figure 3 , Attachment Figure 4 , Attachment Figure 5 and attached Figure 6 shown.

[0028] This embodiment uses the Canny edge detection algorithm, which is a well-known technology and the specific process is not repeated. In other implementations of the embodiment of the present application, the Sobel operator and the Prewitt operator can also be used to obtain the edge points and their gradients in the grayscale image.

[0029] For the edge points obtained by the edge detection algorithm, these edge points are used as the input of the edge tracking algorithm to obtain a continuous edge contour. The edge tracking algorithm is a well-known technology, and the specific process is not repeated here. These edge contours include both hole edge contours and non-hole edge contours. Next, it is necessary to distinguish the hole edge contours from the non-hole edge contours.

[0030] In the cylinder head, each hole is circular and independently distributed, and the edge contour of the hole should be closed, the curvature of each edge point in the edge contour is relatively close, and the gradient direction of each edge point in the hole edge contour is different. Most of the non-hole edge contours are not closed. These edge contours are straight or curved. The average curvature of the edge points on the edge contour will be smaller. On the same non-hole edge contour, the difference in the gradient direction of each edge point is small.

[0031] In the embodiment of the present application, for any edge point of any edge contour in the hole side image of the motorcycle cylinder head, M edge points on the left and right sides of the any edge point are obtained respectively; the any edge point and the 2M edge points on both sides thereof are subjected to curve fitting to obtain the curvature of the any edge point on the curve where it is located. The calculation process of the curvature on the curve is a well-known technology and will not be described in detail.

[0032] In particular, when the number of edge points on the left or right of edge point C is less than M, the curvature of edge point C is taken to be equal to the curvature of the edge point whose number of left and right edge points is greater than or equal to M. Thus, the curvature of each edge point in each edge profile can be obtained. In this embodiment, M is set to 5, which can be set by the implementer according to the actual situation.

[0033] (1) Furthermore, based on the shape features of the edge contour detected in the side image of the motorcycle cylinder head hole, the edge contour circle-like value of the edge contour is constructed, specifically: Preferably, in the embodiment of the present application, the shape feature includes the curvature of edge points on the edge contour and the overall distribution of the gradient direction.

[0034] In other embodiments of the present application, based on the curvature of edge points on the edge contour and the overall distribution of gradient directions as shape features, the shape features also include the overall distribution of tangent slopes and gradient amplitudes of edge points on the edge contour.

[0035] Specifically, in one implementation of the embodiment of the present application, taking the ith edge contour as an example, the edge contour circle-likeness value of the ith edge contour is calculated, specifically: The gradient direction sequence of all edge points in the i-th edge contour is recorded as When the i-th edge contour is a hole edge contour, The data in is diverse. The information entropy of will be larger. When the i-th edge contour is a non-hole edge contour, the edge contour may be a straight line or a curve; if the edge contour is a straight line edge contour, the gradient direction of each edge point in the edge contour is close, then The information entropy and variance of the edge contour are small; if the edge contour is curved, the information entropy of the curved edge contour is smaller than that of the circular edge contour. The information entropy is small and the range is small.

[0036] Based on the above features, the edge contour circle-like value is constructed to characterize the possibility that an edge contour belongs to the hole edge contour. The edge contour circle-like value of the i-th edge contour is The expression is: ; In the formula, represents the edge contour circularity value of the i-th edge contour; Represents the sequence of curvatures of all edge points in the i-th edge contour; express The average value of all data in; D represents the variance function, which is used to calculate the variance of the input sequence; represents a sequence composed of the gradient directions of all edge points in the i-th edge profile; S represents an information entropy function, which is used to calculate the information entropy of the input sequence. The calculation of information entropy is a well-known technology and the specific process will not be repeated here; B represents a range function, which is used to calculate the range of the input sequence.

[0037] It should be understood that when When the value of is large, it means that the average curvature of the i-th edge profile is large, which means that the more curved the edge profile is, the more likely it is that the edge profile is the hole edge profile; The smaller the value of is, the closer the curvature value of each point on the i-th edge contour is, which means that the edge contour is more likely to be a circular hole edge contour; The larger the value of is, the more diverse the gradient directions of the edge points on the i-th edge contour are, which means that the edge contour is more likely to be a hole edge contour. The larger the value is, the greater the difference in the gradient direction of each edge point on the i-th edge contour is, which means that the edge contour is more likely to be a hole edge contour; The larger the value of , the larger the span of the gradient direction of the edge points on the i-th edge contour is, which further indicates that the edge contour is more likely to be a hole edge contour.

[0038] That is, the larger the circle-like value of the edge contour is, the more likely the edge contour is to be a hole edge contour. Accordingly, the hole edge contour is selected according to the circle-like value of the edge contour.

[0039] (2) Furthermore, the edge contour circle-like value is used to filter out the edge contour of the hole, specifically: Preferably, in the embodiment of the present application, a segmentation threshold in the edge contour circle-likeness values ​​of all edge contours is obtained; and the edge contour corresponding to the edge contour circle-likeness value greater than the segmentation threshold is used as the hole edge contour.

[0040] In other embodiments of the present application, all edge contour circle-likeness values ​​may be sorted from large to small, and edge contours corresponding to the first preset values ​​in the sorting result may be used as the hole edge contours.

[0041] In one implementation of the embodiments of the present application, the circle-likeness values ​​of edge contours of all edge contours are calculated, and all edge contour circle-likeness values ​​are used as input of the Otsu threshold segmentation algorithm to obtain the segmentation threshold z, and the edge contours are divided into hole edge contours and non-hole edge contours according to the size of z.

[0042] when When , the i-th edge contour is recorded as the hole edge contour; when When , the i-th edge contour is recorded as the non-hole edge contour. Among them, the Otsu threshold segmentation algorithm is a well-known technology, and the specific process is not repeated here.

[0043] In other implementations of the embodiments of the present application, a clustering algorithm may be used to segment the circle-like values ​​of all edge contours according to numerical values, which is specifically set by the implementer.

[0044] At this point, several hole edge contours in the hole side image of the motorcycle cylinder head are obtained.

[0045] A2: Based on the center distance and radius difference between the hole edge contour and the Hough circle, the edge contour Hough matching degree is constructed, and the matching Hough circle of each hole edge contour is screened out using the edge contour Hough matching degree.

[0046] The binary image obtained in step A1 is used as the input of the Hough circle detection algorithm to obtain the center coordinates and radius of each Hough circle in the binary image. These Hough circles may contain both normal hole positions and deformed hole positions. The Hough circle detection algorithm is a well-known technology, and the specific process will not be repeated here.

[0047] In the side image of the hole of the motorcycle cylinder head, due to the different thickness and size of the hole, a hole may contain multiple edge contours of different sizes, and these edge contours share the same center. In order to determine whether the hole contour is deformed, it is necessary to compare whether there is a shape difference between the hole contour and the Hough circle that matches the hole contour, so it is necessary to find the Hough circle that matches the hole contour. Therefore, it is necessary to specifically analyze the center distance and radius difference between the hole edge contour and the Hough circle, and match a suitable Hough circle for each hole edge contour, so that a hole contour and a Hough circle represent the same contour of the same hole.

[0048] (1) Furthermore, the method for obtaining the center coordinates and radius of the hole edge contour used in calculating the center distance and radius difference is as follows: Preferably, in the embodiment of the present application, the center coordinate of the hole edge contour is determined by the center of the minimum circumscribed rectangle of the hole edge contour. The approximate radius of the hole edge contour is determined by half of the average value of the distance between the two edge points that are farthest apart in the longitudinal direction and the distance between the two edge points that are farthest apart in the transverse direction on the hole edge contour.

[0049] In other embodiments of the present application, the coordinates of the geometric center of the hole edge contour may be used as its center coordinates, and the radius of the hole edge contour after Hough circle fitting may be used as its approximate radius.

[0050] In one implementation of the embodiment of the present application, the edge profile of the j-th hole position is taken as an example for analysis, specifically: For the jth hole edge contour, calculate the distance between the two edge points on the edge contour that are the farthest apart in the horizontal direction, and record it as ; Calculate the distance between the two pixels that are the farthest apart in the vertical direction on the edge contour, recorded as ; The radius of the edge contour of the j-th hole position is recorded as the approximate radius r of the edge contour of the j-th hole position, . Get the minimum circumscribed rectangle of the j-th hole edge contour, and use the center of the minimum circumscribed rectangle as the center coordinate of the j-th hole edge contour.

[0051] (2) Further, the edge contour Hough matching degree is constructed, and the edge contour Hough matching degree is used to screen out the matching Hough circle of each hole edge contour. The specific screening method is as follows: If the jth hole edge contour and the mth Hough circle are the same edge contour of the same hole, then the distance between the center point of the jth hole edge contour and the center point of the mth Hough circle is shorter, and the approximate radius of the jth edge contour is less different from the radius of the mth Hough circle.

[0052] Based on the above features, the edge contour Hough matching degree is constructed to characterize the matching degree between the edge contour of the jth hole position and the mth Hough circle.

[0053] Preferably, in an embodiment of the present application, a method for constructing the edge contour Hough matching degree is: obtaining the center distance and radius difference between the hole edge contour and the Hough circle; the edge contour Hough matching degree is negatively correlated with the center distance and radius difference, respectively.

[0054] It can be understood that a negative correlation means that the dependent variable will decrease as the independent variable increases, and the dependent variable will increase as the independent variable decreases. This is determined by actual application and is not specifically limited in this application.

[0055] In one implementation of the embodiment of the present application, the Hough matching degree between the edge profile of the j-th hole position and the edge profile of the m-th Hough circle is calculated. For example, the expression is: ; In the formula, is the Hough matching degree between the edge contour of the j-th hole position and the edge contour of the m-th Hough circle; Represents the center coordinate of the edge contour of the jth hole; Represents the coordinates of the center of the mth Hough circle; is a Euclidean distance function, which is used to calculate the Euclidean distance between two input coordinate points. The calculation of the Euclidean distance is a well-known technique, and the specific process will not be described in detail. represents the approximate radius of the edge contour of the jth hole; is the radius of the mth Hough circle; It is a constant to prevent the denominator from being 0, and its value is 0.01.

[0056] It should be understood that when When the value of is small, it means that the distance between the center coordinate of the edge contour of the jth hole position and the center of the mth Hough circle is very close, which means that the edge contour of the jth hole position and the mth Hough circle are likely to be the same hole position; when When the value of is small, it means that the approximate radius of the edge contour of the jth hole position is very close to the radius of the mth Hough circle, which means that the edge contour of the jth hole position and the mth Hough circle are likely to be the same edge contour of the same hole position. The larger the value of , the more likely it is that the edge contour of the j-th hole position and the m-th Hough circle are the same edge contour of the same hole position.

[0057] Preferably, in the embodiment of the present application, the matching Hough circle of each hole edge contour is: the Hough circle corresponding to the maximum Hough matching degree of the edge contour calculated with the hole edge contour.

[0058] In one implementation of the embodiments of the present application, taking the matching Hough circle of the edge contour of the j-th hole position as an example: the edge contour Hough matching degree between the edge contour of the j-th hole position and each Hough circle is calculated, and the Hough circle corresponding to the maximum value of the edge contour Hough matching degree is used as the matching Hough circle of the edge contour of the j-th hole position.

[0059] Similarly, a Hough circle matching the edge contour of each hole position is obtained, and the successfully matched hole position edge contour and the Hough circle represent the same edge contour of the same hole position.

[0060] A3, based on the edge between the hole edge contour and its matching Hough circle, and the overlapping difference between the hole edge contour and the enclosed area of ​​the matching Hough circle, a hole deformation discrimination value of the hole edge contour is constructed to identify whether the hole edge contour is deformed.

[0061] Furthermore, by analyzing the overlap difference between the matched hole edge contour and the Hough circle, it is determined whether the hole is deformed.

[0062] If a hole is deformed due to extrusion, the ratio of the number of non-overlapping edge points to the number of overlapping pixel points in the edge points of the hole edge contour and the edge points of the Hough circle is large; when a normal circular hole is deformed, the perimeter of the edge contour after deformation is equal to the perimeter before deformation, but the shape is no longer circular, then the area of ​​the hole after deformation will be smaller than the area before deformation, and the intersection ratio of the area enclosed by the hole edge contour of the hole and the area enclosed by the Hough circle will be smaller.

[0063] Based on the above characteristics, a hole position deformation discrimination value is constructed to characterize whether a hole position is deformed. The specific expression is: ; In the formula, represents the hole position deformation discrimination value of the jth hole position edge contour; It represents the ratio of the number of non-overlapping pixels to the number of overlapping pixels of the edge contour of the j-th hole position and all edge pixels of its matching Hough circle; represents the area of ​​the region enclosed by the matching Hough circle of the jth hole edge contour pixel point, represents the area of ​​the area surrounded by the edge contour of the jth hole position; exp is an exponential function with a natural constant as the base; iou is an intersection-and-union ratio function, which is used to calculate the intersection-and-union ratio of the areas of the two input areas. The calculation of the intersection-and-union ratio is a well-known technology, and the specific process is not repeated here. The area is the number of pixels in the area it surrounds.

[0064] It should be understood that when When the value of is large, it means that many of the pixel points of the jth hole edge contour and its matching Hough circle edge points do not overlap, which further indicates that the hole represented by the hole edge contour may be deformed. When the value of is large, it means that the area of ​​the matching Hough circle is larger than the area of ​​the edge contour of the jth hole position, which further indicates that the edge contour of the jth hole position may have been deformed; when When the value of is small, it means that the ratio of the area where the j-th hole edge contour intersects with its matching Hough circle to the area where they merge is small, which further indicates that the hole represented by the j-th hole edge contour may have deformed.

[0065] When the j-th hole edge contour matches its matching Hough circle, the hole deformation discrimination value of the j-th hole edge contour is calculated. When it is greater than a preset threshold value b, it indicates that the hole position represented by the edge contour of the j-th hole position has been deformed. In this embodiment, b is 10.

[0066] At this point, it can be determined whether each hole in the hole side image of the motorcycle cylinder head is deformed.

[0067] The embodiment of the present application also proposes a motorcycle cylinder head hole position deformation side detection device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements any of the steps of the above-mentioned motorcycle cylinder head hole position deformation side detection method when executing the computer program. Since a motorcycle cylinder head hole position deformation side detection method has been described in detail above, it will not be repeated here.

[0068] The various embodiments in the present application are described in a progressive manner, and the same or similar parts between the various embodiments can be referenced to each other, and each embodiment focuses on the differences from other embodiments.

[0069] It should be noted that, unless otherwise specified and limited, terms such as "include", "comprises" or any other variants thereof are intended to cover non-exclusive inclusion, so that a circuit structure, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such article or device. In the absence of further restrictions, an element defined by the sentence "including one..." does not exclude the existence of other identical elements in the article or device including the element. In addition, the term "and\or" used herein includes any and all combinations of one or more related listed items.

[0070] Those skilled in the art will readily appreciate other embodiments of the present application after considering the specification and practicing the invention herein. The present application is intended to cover any variations, uses or adaptations of the present application, which follow the general principles of the present application and include common knowledge or customary technical means in the art that are not invented by the present application.

[0071] It should be understood that the present application is not limited to the exact construction that has been described above and shown in the drawings, and that various modifications and changes may be made without departing from the scope thereof.

Claims

1. A method for detecting the deformation of a motorcycle cylinder head hole, characterized in that: The method comprises the following steps: Collect the side image of the motorcycle cylinder head hole; The edge contour features in the side image of the motorcycle cylinder head hole are used to identify whether the hole edge contour is deformed. Specifically: A1, based on the shape features of the edge contour detected in the side image of the motorcycle cylinder head hole, the edge contour circle-like value of the edge contour is constructed, and the edge contour of the hole is screened out using the edge contour circle-like value; A2: Based on the center distance and radius difference between the hole edge contour and the Hough circle, the edge contour Hough matching degree is constructed, and the matching Hough circle of each hole edge contour is screened out using the edge contour Hough matching degree; A3, based on the edge between the hole edge contour and its matching Hough circle, and the overlapping difference between the hole edge contour and the enclosed area of ​​the matching Hough circle, a hole deformation discrimination value of the hole edge contour is constructed to identify whether the hole edge contour is deformed.

2. A method for detecting the deformation of a motorcycle cylinder head hole according to claim 1, characterized in that: The shape features include the curvature of edge points on the edge contour and the overall distribution of gradient directions.

3. A method for detecting the deformation of a motorcycle cylinder head hole according to claim 1, characterized in that: The screening method of the hole edge profile is: Get the segmentation threshold value in the edge contour circle-like value of all edge contours; The edge contour corresponding to the circle-like value of the edge contour greater than the segmentation threshold is taken as the hole edge contour.

4. A method for detecting the deformation of a motorcycle cylinder head hole according to claim 1, characterized in that: The center coordinates of the hole edge contour used in calculating the center distance are determined by the center of the minimum circumscribed rectangle of the hole edge contour.

5. A method for detecting the deformation of a motorcycle cylinder head hole according to claim 1, characterized in that: The radius of the hole edge profile used in calculating the radius difference is determined by half of the average value of the distance between the two edge points that are farthest apart in the longitudinal direction and the distance between the two edge points that are farthest apart in the transverse direction on the hole edge profile.

6. A method for detecting the deformation of a motorcycle cylinder head hole according to claim 1, characterized in that: The Hough circle is obtained by using a Hough circle detection algorithm from a binary image of a motorcycle cylinder head hole side image.

7. A method for detecting the deformation of a motorcycle cylinder head hole according to claim 1, characterized in that: The matching Hough circle of each hole edge contour is: the Hough circle corresponding to the maximum value of the edge contour Hough matching degree calculated from the hole edge contour.

8. The method for detecting the hole position deformation of a motorcycle cylinder head according to claim 1, characterized in that: The construction expression of the hole position deformation discrimination value is: ; In the formula, represents the hole position deformation discrimination value of the jth hole position edge contour; It represents the ratio of the number of non-overlapping pixels to the number of overlapping pixels of the edge contour of the j-th hole position and all edge pixels of its matching Hough circle; represents the area of ​​the region enclosed by the matching Hough circle of the jth hole edge contour pixel point, It represents the area of ​​the region enclosed by the edge contour of the jth hole; exp is an exponential function with a natural constant as the base; iou is the intersection-union ratio function.

9. A method for detecting the deformation of a motorcycle cylinder head hole according to claim 1, characterized in that: The identifying whether the hole edge contour is deformed includes: when the hole deformation discrimination value of the hole edge contour is greater than a preset threshold, identifying the hole edge contour as a deformed hole; otherwise, identifying the hole edge contour as a non-deformed hole.

10. A motorcycle cylinder head hole position deformation detection device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the computer program, the steps of a method for detecting the deformation side of a motorcycle cylinder head hole as described in any one of claims 1 to 9 are implemented.

Citation Information

Patent Citations

  • Machine vision-based engine cylinder block casting locating method

    CN106097323A

  • Button hole defect detection method

    CN115272305A

  • Stamping part circular hole positioning and deformation detection method based on sub-pixel edge extraction

    CN116542910A

  • Method for quickly tracing quality of blast bead product

    CN116721102A

  • Welded pipe surface defect edge visual extraction method

    CN117314922A