Visual measurement method and system for contour of gear block part of bulldozer
By analyzing the grayscale image of the bulldozer tooth block parts and filtering out the true contour edge lines, the problems of texture and noise interference in the prior art are solved, and the accurate evaluation of the production quality of the tooth block parts is achieved.
Patent Information
- Application Number
- CN202411507069.3
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-10-28
- Publication Date
- 2024-11-29
- Estimated Expiration
- Not applicable · inactive patent
AI Technical Summary
The prior art cannot accurately extract the contour lines of bulldozer tooth block parts, resulting in a decrease in accuracy in production quality evaluation, mainly due to the presence of texture edge lines and noise interference in the tooth block images.
A visual measurement method is adopted to obtain the grayscale image of the tooth block part, perform edge detection, analyze the noise possibility and contour authenticity of the closed edge line, filter out the real contour edge line, and compare it with the standard contour line. Evaluate production quality.
It improves the accuracy of the production quality evaluation of the tooth block parts, eliminates texture and noise interference, and accurately extracts the contour edge lines of the tooth block parts, improving the accuracy of the evaluation.
Smart Images

Figure CN119048578A_ABST
Abstract
Description
Technical Field
[0001] The invention relates to the field of gear segment profile measurement, and in particular to a visual measurement method and system for the profile of a bulldozer gear segment part. Background Art
[0002] The gear block of a bulldozer is an important part of its transmission system. It is mainly used to transmit driving force and provide traction, driving the bulldozer tracks to move and operate in various complex terrains. Usually, a driving wheel is composed of multiple gear block parts, and the matching accuracy between the gear block and the track affects the traction and walking stability of the bulldozer. Therefore, whether the shape or contour of the gear block parts meets the production standards is an important indicator for evaluating the production quality of the gear block parts.
[0003] In the related art, edge detection is usually performed on the gear segment part image, and the extracted contour lines are compared with the contour lines of the standard gear segment parts to evaluate the production quality of the gear segment parts. However, due to the presence of relatively obvious texture edge lines in the gear segment image and the noise in the image may also form edge lines, the existing methods are unable to accurately extract the contour lines of the gear segment parts, thereby reducing the accuracy of the gear segment part production quality assessment. Summary of the invention
[0004] In order to solve the technical problem that the existing method cannot accurately extract the contour line of the gear segment part and reduces the accuracy of the production quality assessment of the gear segment part, the purpose of the present invention is to provide a visual measurement method and system for the contour of the bulldozer gear segment part. The technical solution adopted is as follows: The present invention proposes a visual measurement method for the profile of a bulldozer gear segment part, the method comprising: Obtaining a grayscale image of each gear segment part of the bulldozer, taking the grayscale image of any gear segment part as a target grayscale image, performing edge detection on the target grayscale image, and obtaining a plurality of closed edge lines in the target grayscale image; Taking any closed edge line in the target grayscale image as the target closed edge line, taking the area enclosed by the target closed edge line as the inner area of the target closed edge line, and obtaining the noise possibility of the target closed edge line according to the distribution of the gradient amplitude of each pixel point on the target closed edge line, the direction of the target closed edge line, and the number of pixels in the inner area; taking other areas except the inner area as the outer area of the target closed edge line, and obtaining the contour authenticity of the target closed edge line according to the difference in the grayscale value of the pixel points between the inner area and the outer area of the target closed edge line; According to the noise possibility and the contour authenticity, the contour possibility of the target closed edge line is obtained; based on the contour possibility, the contour edge line is screened out from all closed edge lines of the target grayscale image; according to the number of pixel points at the same position on the contour edge line of each grayscale image and the standard contour line of the standard gear segment part image, the quality assessment parameter of each gear segment part is obtained; Based on the quality assessment parameters, the production quality of each gear segment component is assessed.
[0005] Further, the noise possibility of obtaining the target closed edge line includes: Analyze the discrete degree of the gradient amplitude of all pixel points on the target closed edge line to obtain the first noise parameter of the target closed edge line; Performing negative correlation mapping on the number of all pixel points in the inner area of the target closed edge line to obtain a second noise parameter of the target closed edge line; Acquire an eight-chain code sequence of a target closed edge line, and obtain a third noise parameter of the target closed edge line according to differences between adjacent elements in the eight-chain code sequence; The first noise parameter, the second noise parameter and the third noise parameter of the target closed edge line are integrated to obtain the noise possibility of the target closed edge line.
[0006] Further, obtaining the third noise parameter of the target closed edge line according to the difference between adjacent elements in the eight-chain code sequence includes: In the eight-chain code sequence, the absolute value of the difference between each element and the next adjacent element is taken as the variation of each element; The average value of the variation of all elements in the eight-chain code sequence is used as the third noise parameter of the target closed edge line.
[0007] Furthermore, obtaining the contour authenticity of the target closed edge line includes: Take any pixel point on the target closed edge line as the target pixel point; In a preset window centered on the target pixel point, the pixel points belonging to the inner area are used as the inner pixel points of the preset window, and the pixel points belonging to the outer area are used as the outer pixel points of the preset window; The average grayscale value of all inner pixels in the preset window is used as the inner overall grayscale value of the preset window, and the average grayscale value of all outer pixels in the preset window is used as the outer overall grayscale value of the preset window; the absolute value of the difference between the inner overall grayscale value and the outer overall grayscale value is used as the grayscale difference value of the target pixel; The overall level of the grayscale difference values of all pixels on the target closed edge line is analyzed to obtain the contour authenticity of the target closed edge line.
[0008] Furthermore, the possibility of obtaining the contour of the target closed edge line includes: Performing negative correlation mapping on the noise possibility to obtain a contour judgment value of a target closed edge line; The contour judgment value and the contour authenticity are integrated and normalized to obtain the contour possibility of the target closed edge line.
[0009] Further, the step of selecting contour edge lines from all closed edge lines of the target grayscale image based on the contour possibility includes: In the target grayscale image, the closed edge lines whose contour likelihood is greater than a preset likelihood threshold are taken as contour edge lines.
[0010] Furthermore, obtaining the quality assessment parameters of each gear segment part includes: Performing image alignment processing on the target grayscale image and the standard gear segment part image, and obtaining the coincident pixel points on the contour edge line of the target grayscale image, wherein the positions of the coincident pixel points are the same as the positions of the pixel points on the standard contour line of the standard gear segment part image; The number of all the overlapping pixel points on the contour edge line of the target grayscale image is normalized to obtain the quality assessment parameter of the gear segment part corresponding to the target grayscale image.
[0011] Further, the evaluating the production quality of each gear segment part based on the quality evaluation parameter includes: If the quality assessment parameter is greater than the preset assessment threshold, the block part is qualified for production, otherwise, the gear block part is unqualified for production.
[0012] Furthermore, performing edge detection on the target grayscale image to obtain a plurality of closed edge lines in the target grayscale image includes: Using the Canny edge detection algorithm, the target grayscale image is processed to obtain multiple edge lines in the target grayscale image; Using the endpoint detection algorithm, each edge line of the target grayscale image is processed to obtain the number of endpoints of each edge line; The edge line with the number of endpoints equal to the value 1 is regarded as a closed edge line.
[0013] The present invention also proposes a visual measurement system for the contour of a bulldozer tooth segment part. The system includes a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, it implements any step of a visual measurement method for the contour of a bulldozer tooth segment part.
[0014] The present invention has the following beneficial effects: The present invention takes into account that the existing methods cannot accurately extract the contour lines of the gear block parts, which reduces the accuracy of the production quality assessment of the gear block parts. First, a grayscale image of each gear block part is obtained. Considering that there are obvious textures on the surface of the gear block parts, the texture edge lines are usually non-closed, while the contour edge lines of the gear block parts are closed. Because the closed edge lines in the target grayscale image are first extracted to eliminate the interference of the surface texture on the contour line extraction, considering that the more concentrated noise phenomenon in the target grayscale image will also form closed edge lines, due to the strong randomness of the influence of noise on the grayscale value of pixel points, the closed edge lines formed by the noise are less smooth and have a higher degree of irregularity in shape. At the same time, the area affected by the concentrated noise is smaller, and the area surrounded by the closed edge lines formed by the noise is Small, so the noise possibility can be used to reflect the possibility that the target closed edge line is formed by the influence of noise. Subsequently, the interference of noise on the extraction of the gear block contour can be eliminated based on the noise possibility. Taking into account the large grayscale difference between the gear block part and the background, resulting in a large difference in the grayscale values of the pixels on the inner and outer sides of the contour of the gear block part, while the difference in the grayscale values of the pixels on the inner and outer sides of the closed edge line formed by the noise is relatively small. Therefore, the contour authenticity can be used to reflect the possibility that the target closed edge line is the true contour of the gear block part. Further, the contour edge line of the gear block part in the target grayscale image can be accurately extracted through the contour possibility, and the contour edge line is compared with the standard contour line of the standard gear block part image to accurately obtain the quality evaluation parameters of the gear block part, thereby improving the accuracy of the production quality evaluation of the gear block part. BRIEF DESCRIPTION OF THE DRAWINGS
[0015] In order to more clearly illustrate the technical solutions and advantages in the embodiments of the present invention 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 invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.
[0016] Figure 1 A flow chart of a visual measurement method for the profile of a bulldozer gear segment part provided by one embodiment of the present invention; Figure 2A flow chart of a method for acquiring the noise probability of a target closed edge line provided by one embodiment of the present invention. DETAILED DESCRIPTION
[0017] In order to further explain the technical means and effects adopted by the present invention to achieve the predetermined invention purpose, the following is a detailed description of the visual measurement method and system for the profile of the gear segment parts of a bulldozer proposed in accordance with the present invention, its specific implementation method, 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 invention belongs.
[0019] The following is a detailed description of a method and system for visually measuring the profile of a bulldozer gear segment part provided by the present invention in conjunction with the accompanying drawings.
[0020] See also Figure 1 , which shows a flow chart of a visual measurement method for the profile of a bulldozer gear segment part provided by an embodiment of the present invention, the method comprising: Step S1: obtaining a grayscale image of each tooth segment part of the bulldozer, taking the grayscale image of any tooth segment part as a target grayscale image, performing edge detection on the target grayscale image, and obtaining a plurality of closed edge lines in the target grayscale image.
[0021] The embodiment of the present invention first uses an industrial camera to photograph the gear block parts of the bulldozer, and collects the original image of each gear block part. In order to reduce the calculation amount of subsequent image processing and improve the processing speed, in one embodiment of the present invention, the collected original image is grayed and converted into a single-channel gray image. It should be noted that graying is a technical means well known to those skilled in the art and will not be described in detail here.
[0022] In the subsequent steps, it is necessary to analyze the grayscale image of each gear segment part, so the grayscale image of any gear segment part is used as the target grayscale image. Since the gear segment parts will have obvious textures on their surfaces during the production and processing process, and the textures on the surface of the gear segment parts are usually shown as non-closed edge lines in the grayscale images, and the contours of the gear segment parts are usually shown as closed edge lines in the grayscale images, in order to eliminate the interference of surface texture on contour extraction, edge detection can be performed on the target grayscale image first, and multiple closed edge lines in the target grayscale image can be obtained. Subsequently, the contours of the gear segment parts can be further extracted from the closed edge lines to improve the accuracy of the production quality assessment of the gear segment parts.
[0023] Preferably, in one embodiment of the present invention, a method for acquiring multiple closed edge lines in a target grayscale image specifically includes: First, the Canny edge detection algorithm is used to process the target grayscale image to obtain multiple edge lines in the target grayscale image. For a non-closed edge line, there are usually two endpoints. Therefore, the endpoint detection algorithm can be used to process each edge line of the target grayscale image to obtain the number of endpoints of each edge line, and the edge line with the number of endpoints equal to the value 1 is regarded as a closed edge line. It should be noted that when the endpoint detection algorithm processes a closed edge line, the starting point and the end point detected are the same point, and the closed edge line is considered to have one endpoint. When processing a non-closed edge line, the starting point and the end point are not the same point, and the non-closed edge line is considered to have two endpoints.
[0024] Among them, the Canny edge detection algorithm and the endpoint detection algorithm are both technical fields well known to those skilled in the art and will not be described in detail here.
[0025] Step S2: taking any closed edge line in the target grayscale image as the target closed edge line, taking the area enclosed by the target closed edge line as the inner area of the target closed edge line, and obtaining the noise possibility of the target closed edge line according to the distribution of the gradient amplitude of each pixel point on the target closed edge line, the direction of the target closed edge line, and the number of pixels in the inner area; taking other areas except the inner area as the outer area of the target closed edge line, and obtaining the contour authenticity of the target closed edge line according to the difference in the grayscale values of the pixels between the inner area and the outer area of the target closed edge line.
[0026] Since there is a small range of noise influence in the target grayscale image, and the more concentrated noise phenomenon in the target grayscale image will also form closed edge lines, the closed edge lines obtained above are not all the contours of the gear block parts, and there are also closed edge lines caused by the concentrated influence of noise. Therefore, it is necessary to analyze each closed edge line. First, any closed edge line in the target grayscale image is used as the target closed edge line. Considering the strong randomness of the influence of noise on the grayscale value of pixel points, the smoothness of the closed edge line formed by the noise is poor, that is, the gradient amplitude distribution of the pixel points on the closed edge line formed by the noise is less consistent, and the shape of the closed edge line formed by the noise is more irregular. At the same time, the area affected by the concentrated noise is smaller, and the area formed by the noise is The area of the region enclosed by the closed edge line is small, so the existing gradient operators such as the Sobel operator or the Roberts operator can be used to calculate the gradient amplitude of the pixel points in the target grayscale image first, and the area enclosed by the target closed edge line is used as the inner area of the target closed edge line, and then the distribution of the gradient amplitude of each pixel point on the target closed edge line, the direction of the target closed edge line, and the number of pixels in the inner area are analyzed. The noise possibility obtained reflects the possibility that the target closed edge line is formed by the influence of noise. The greater the noise possibility, the greater the possibility that the target closed edge line is formed by the influence of noise, and then the possibility that the target closed edge line is the contour of the gear block part is smaller. Subsequently, the closed edge line formed by noise can be screened out based on the noise possibility to improve the accuracy of gear block contour extraction.
[0027] Preferably, in one embodiment of the present invention, the method for obtaining the noise possibility of the target closed edge line specifically includes: See also Figure 2 , which shows a flow chart of a method for acquiring the noise possibility of a target closed edge line provided by an embodiment of the present invention.
[0028] Step S201: Analyze the discrete degree of the gradient amplitude of all pixel points on the target closed edge line to obtain the first noise parameter of the target closed edge line.
[0029] From the above analysis, it can be seen that compared with the contour line of the gear block part, the gradient amplitude distribution of each pixel point on the closed edge line formed by noise is less consistent. Therefore, the discrete degree of the gradient amplitude of all pixel points on the target closed edge line can be analyzed to obtain the first noise parameter of the target closed edge line. The larger the first noise parameter, the greater the possibility that the target closed edge line is formed by noise, which provides a data basis for the subsequent accurate calculation of the noise possibility of the target closed edge line.
[0030] In an embodiment of the present invention, the variance or standard deviation of the gradient amplitude of all pixel points on the target closed edge line can be used as the first noise parameter of the target closed edge line to analyze the discrete degree of the gradient amplitude of all pixel points on the target closed edge line, which is not limited here.
[0031] Step S202: performing negative correlation mapping on the number of all pixel points in the inner area of the target closed edge line to obtain a second noise parameter of the target closed edge line.
[0032] Since the range of concentrated noise influence is small, if the target closed edge line is formed by the influence of noise, the area enclosed by the target closed edge line is small, that is, the number of pixels in the inner area is small. Therefore, the number of all pixels in the inner area of the target closed edge line can be negatively correlated to obtain the second noise parameter of the target closed edge line. The larger the second noise parameter, the greater the possibility that the target closed edge line is formed by the influence of noise, which provides a data basis for the subsequent accurate calculation of the noise possibility of the target closed edge line.
[0033] Step S203: Obtain an eight-chain code sequence of the target closed edge line, and obtain a third noise parameter of the target closed edge line according to the difference between adjacent elements in the eight-chain code sequence.
[0034] From the above analysis, it can be seen that compared with the contour line of the gear block part, the shape of the closed edge line formed by the influence of noise is more irregular, and the chain code can reflect the shape characteristics of the edge line. Therefore, the eight-chain code sequence of the target closed edge line can be obtained first, wherein the eight-chain code is a technical means well known to those skilled in the art and will not be elaborated here. The greater the difference between adjacent elements in the eight-chain code sequence, the more irregular the shape or direction of the target closed edge line. Therefore, the third noise parameter of the target closed edge line can be obtained according to the difference between adjacent elements in the eight-chain code sequence. The larger the third noise parameter, the greater the possibility that the target closed edge line is formed by the influence of noise, which provides a data basis for the subsequent accurate calculation of the noise possibility of the target closed edge line.
[0035] Preferably, in one embodiment of the present invention, the method for obtaining the third noise parameter of the target closed edge line specifically includes: In the eight-chain code sequence, the absolute value of the difference between each element and the next adjacent element is taken as the variation of each element. The larger the variation, the more inconsistent the directions of adjacent pixels on the target closed edge line, and further the more irregular the shape of the target closed edge line. Therefore, the average value of the variation of all elements in the eight-chain code sequence can be used as the third noise parameter of the target closed edge line.
[0036] It should be noted that since there is no adjacent next element to the last element, but the target closed edge line has a closed feature, the absolute value of the difference between the last element and the first element can be used as the variation of the last element.
[0037] Step S204: synthesizing the first noise parameter, the second noise parameter and the third noise parameter of the target closed edge line to obtain the noise possibility of the target closed edge line.
[0038] The larger the first noise parameter, the second noise parameter and the third noise parameter of the target closed edge line are, the greater the possibility that the target closed edge line is formed by the influence of noise. Therefore, the first noise parameter, the second noise parameter and the third noise parameter of the target closed edge line can be combined to obtain the noise possibility of the target closed edge line.
[0039] In an embodiment of the present invention, the sum or product of the first noise parameter, the second noise parameter and the third noise parameter of the target closed edge line may be used as the noise possibility of the target closed edge line to achieve a combination of the three, which is not limited here.
[0040] As an example, in one embodiment of the present invention, the expression of the noise possibility of the target closed edge line can be specifically, for example, as follows: in, Represents the noise probability of the closed edge line of the target; A first noise parameter representing a closed edge line of a target; Represents the number of all pixels in the inner area of the target closed edge line; A second noise parameter representing a closed edge line of the target; The first The amount of change in an element; The number of elements in the octach sequence representing the target closed edge line; The third noise parameter representing the closed edge line of the target.
[0041] It should be noted that in other embodiments of the present invention, the positive and negative correlations may be represented by other basic mathematical operations, which will not be elaborated herein.
[0042] Taking into account the large grayscale difference between the gear block parts and the background, resulting in a large difference in the grayscale values of the pixels on the inner and outer sides of the contour of the gear block parts, while the difference in the grayscale values of the pixels on the inner and outer sides of the closed edge line formed by noise is relatively small. Therefore, in the target grayscale image, other areas except the inner area can be first used as the outer area of the target closed edge line, and then the difference in the grayscale values of the pixels between the inner and outer areas of the target closed edge line is analyzed. The obtained contour authenticity reflects the possibility that the target closed edge line is the true contour of the gear block parts. The greater the contour authenticity, the more likely the target closed edge line is the true contour of the gear block parts. Subsequently, the contour edge line of the gear block parts can be accurately extracted by combining the contour authenticity and the noise possibility.
[0043] Preferably, in one embodiment of the present invention, a method for obtaining the contour authenticity of a target closed edge line specifically includes: First, any pixel on the target closed edge line is taken as the target pixel. In the preset window centered on the target pixel, the pixels belonging to the inner area are taken as the inner pixels of the preset window, and the pixels belonging to the outer area are taken as the outer pixels of the preset window. The size of the preset window is set to The specific size of the preset window can also be set by the implementer according to the specific actual scenario and is not limited here.
[0044] Then, the average grayscale value of all inner pixels in the preset window is taken as the overall grayscale value of the inner side of the preset window, and the average grayscale value of all outer pixels in the preset window is taken as the overall grayscale value of the outer side of the preset window; the absolute value of the difference between the overall grayscale value of the inner side and the overall grayscale value of the outer side is taken as the grayscale difference value of the target pixel. The larger the grayscale difference value, the greater the difference in the grayscale values of the pixels on both sides of the target closed edge line.
[0045] The grayscale difference value of each pixel on the target closed edge line can be obtained by the same method as above, and then the overall level of the grayscale difference values of all pixels on the target closed edge line can be analyzed to obtain the contour authenticity of the target closed edge line.
[0046] In an embodiment of the present invention, the average or median of the grayscale difference values of all pixels on the target closed edge line can be used as the contour authenticity of the target closed edge line to achieve an overall level analysis of the grayscale difference values of all pixels on the target closed edge line, which is not limited here.
[0047] As an example, in one embodiment of the present invention, the expression of the contour authenticity of the target closed edge line can be specifically, for example, as follows: in, Indicates the contour authenticity of the closed edge line of the target; Indicates the first The overall grayscale value inside the preset window of pixels; Indicates the first The overall gray value outside the preset window of pixels; Indicates the first Grayscale difference value of pixels; Represents the number of all pixels on the target closed edge line.
[0048] Step S3: According to the noise possibility and the contour authenticity, the contour possibility of the target closed edge line is obtained; based on the contour possibility, the contour edge line is screened out from all the closed edge lines of the target grayscale image; according to the number of pixel points at the same position on the contour edge line of each grayscale image and the standard contour line of the standard gear block part image, the quality evaluation parameters of each gear block part are obtained.
[0049] From the above analysis, it can be seen that the greater the noise possibility of the target closed edge line, the smaller the possibility that the target edge line is the contour of the gear block part, and the greater the contour authenticity of the target closed edge line, the greater the possibility that the target edge line is the contour of the gear block part. Therefore, the noise possibility and contour authenticity of the target closed edge line can be combined to obtain the contour possibility of the target closed edge line. Subsequently, based on the contour possibility, the contour edge line of the gear block part can be accurately extracted in the target grayscale image, eliminating the interference of noise on contour extraction, and improving the accuracy of subsequent production quality assessment of gear block parts.
[0050] Preferably, in one embodiment of the present invention, the method for obtaining the contour possibility of the target closed edge line specifically includes: The noise possibility is negatively correlated to obtain the contour judgment value of the target closed edge line; the contour judgment value and contour authenticity are integrated and normalized to obtain the contour possibility of the target closed edge line.
[0051] In the embodiment of the present invention, the integration of the contour judgment value and the contour authenticity can be achieved by calculating the sum or product of the two, which is not limited here.
[0052] In one embodiment of the present invention, the normalization processing can be specifically, for example, maximum and minimum value normalization processing, and the normalization in subsequent steps can all adopt maximum and minimum value normalization processing. In other embodiments of the present invention, other normalization methods can be selected according to the specific range of numerical values, which will not be described in detail.
[0053] As an example, in one embodiment of the present invention, the expression of the contour possibility of the target closed edge line can be specifically, for example, as follows: in, Represents the contour possibility of the closed edge line of the target; Represents the noise probability of the closed edge line of the target; Indicates the contour authenticity of the closed edge line of the target; represents the normalization function; Indicates the preset adjustment parameter, which is used to prevent the denominator from being 0. The value range is In one embodiment of the present invention, Set to 0.01, The specific value of can also be set by the implementer according to the specific implementation scenario and is not limited here.
[0054] It should be noted that in other embodiments of the present invention, the positive and negative correlations may be represented by other basic mathematical operations, which will not be elaborated herein.
[0055] The contour possibility of each closed edge line in the target grayscale image can be obtained by the same method as mentioned above. The greater the contour possibility, the more likely the closed edge line is to be the contour line of the gear block part. Therefore, based on the contour possibility, the contour edge line can be accurately screened out from all the closed edge lines in the target grayscale image, thereby eliminating the interference of noise on the contour extraction of the gear block part. Subsequently, the extracted contour edge line can be compared with the contour line of the standard gear block part, thereby realizing the evaluation and detection of the production quality of the gear block part.
[0056] Preferably, in one embodiment of the present invention, a closed edge line in the target grayscale image whose contour possibility is greater than a preset possibility threshold can be used as a contour edge line, wherein the preset possibility threshold is set to 0.6, and the specific value of the preset possibility threshold can also be set by the implementer according to the specific implementation scenario, which is not limited here.
[0057] The key to the production quality of the bulldozer's gear segment lies in whether it meets the production standards. The more pixels that overlap between the contour edge line of the gear segment part extracted from the grayscale image and the standard contour line, the more the gear segment part meets the standards. Therefore, the number of pixels at the same position on the contour edge line of each grayscale image and the standard contour line of the standard gear segment part image can be analyzed to obtain the quality assessment parameters of each gear segment part. Subsequently, the quality assessment parameters can be used to evaluate whether the production quality of each gear segment part meets the standards. It should be noted that the standard contour line of the standard gear segment part image is a known contour line manually set during the gear segment part production process.
[0058] Preferably, in one embodiment of the present invention, the method for obtaining the quality assessment parameters of each gear segment part specifically includes: The target grayscale image and the standard gear block part image are image aligned, and the overlapping pixel points on the contour edge line of the target grayscale image are obtained, wherein the result of the image alignment is to adjust the two images to the same spatial coordinate system so that their same parts or same features overlap as much as possible to minimize the difference between them. Image alignment is a technical means well known to those skilled in the art and will not be elaborated here. The position of the overlapping pixel points is the same as the position of the pixel points on the standard contour line of the standard gear block part image, that is, for the position of a certain pixel point on the standard contour line, if the pixel point of the target grayscale image at that position is on the contour edge line, then the pixel point of the target grayscale image at that position is the overlapping pixel point.
[0059] The more the number of overlapping pixels is, the more consistent the contour edge line is with the standard contour line. Therefore, the number of all overlapping pixels on the contour edge line of the target grayscale image can be normalized to obtain the quality assessment parameters of the gear block parts corresponding to the target grayscale image.
[0060] As an example, in one embodiment of the present invention, the expression of the quality assessment parameter of the gear segment part corresponding to the target grayscale image can be specifically, for example, as follows: in, Indicates the quality assessment parameters of the gear block parts corresponding to the target grayscale image; Indicates the number of overlapping pixels on the contour edge line of the target grayscale image; Represents the normalization function.
[0061] The quality assessment parameters of the gear segment parts corresponding to each grayscale image can be obtained by the same method as above.
[0062] Step S4: Based on the quality evaluation parameters, the production quality of each gear segment component is evaluated.
[0063] The larger the quality assessment parameter of each gear segment part is, the more the production quality of the gear segment part meets the standard. Therefore, the production quality of each gear segment part can be accurately assessed based on the quality assessment parameter.
[0064] Preferably, in one embodiment of the present invention, the method for evaluating the production quality of each gear segment part specifically includes: If the quality assessment parameter is greater than the preset assessment threshold, it means that the outer contour of the gear segment part meets the production standard, and the gear segment part is considered to be qualified. Otherwise, the gear segment part is considered to be unqualified. The preset assessment threshold is set to 0.7, and the specific value of the preset assessment threshold can also be set by the implementer according to the specific implementation scenario, and is not limited here.
[0065] An embodiment of the present invention provides a visual measurement system for the contour of a bulldozer gear segment part, the system comprising a memory, a processor and a computer program, wherein the memory is used to store a corresponding computer program, the processor is used to run the corresponding computer program, and when the computer program runs in the processor, the method described in steps S1 to S4 can be implemented.
[0066] In summary, the embodiment of the present invention first obtains a grayscale image of each tooth segment part of the bulldozer, takes the grayscale image of any tooth segment part as the target grayscale image, performs edge detection on the target grayscale image, and obtains multiple closed edge lines in the target grayscale image; takes any closed edge line in the target grayscale image as the target closed edge line, and takes the area enclosed by the target closed edge line as the inner area of the target closed edge line, and obtains the noise possibility of the target closed edge line according to the distribution of the gradient amplitude of each pixel point on the target closed edge line, the direction of the target closed edge line, and the number of pixels in the inner area; and takes the area except the inner area as the target closed edge line. The other areas outside the target closed edge line are taken as the outer areas of the target closed edge line, and the contour authenticity of the target closed edge line is obtained according to the difference in the grayscale values of the pixel points between the inner area and the outer area of the target closed edge line; the contour possibility of the target closed edge line is obtained according to the noise possibility and the contour authenticity; based on the contour possibility, the contour edge lines are screened out from all the closed edge lines of the target grayscale image; the quality evaluation parameters of each gear block part are obtained according to the number of pixels at the same position on the contour edge line of each grayscale image and the standard contour line of the standard gear block part image; based on the quality evaluation parameters, the production quality of each gear block part is evaluated.
[0067] It should be noted that the sequence of the above embodiments of the present invention is only for description and does not represent the advantages and disadvantages of the embodiments. The processes depicted in the accompanying drawings do not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.
[0068] The various embodiments in this specification 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.
Claims
1. A visual measurement method for the contour of a bulldozer gear segment, characterized in that: The method comprises: Obtaining a grayscale image of each gear segment part of the bulldozer, taking the grayscale image of any gear segment part as a target grayscale image, performing edge detection on the target grayscale image, and obtaining a plurality of closed edge lines in the target grayscale image; Taking any closed edge line in the target grayscale image as the target closed edge line, taking the area enclosed by the target closed edge line as the inner area of the target closed edge line, and obtaining the noise possibility of the target closed edge line according to the distribution of the gradient amplitude of each pixel point on the target closed edge line, the direction of the target closed edge line, and the number of pixels in the inner area; taking other areas except the inner area as the outer area of the target closed edge line, and obtaining the contour authenticity of the target closed edge line according to the difference in the grayscale value of the pixel points between the inner area and the outer area of the target closed edge line; According to the noise possibility and the contour authenticity, the contour possibility of the target closed edge line is obtained; based on the contour possibility, the contour edge line is screened out from all closed edge lines of the target grayscale image; according to the number of pixel points at the same position on the contour edge line of each grayscale image and the standard contour line of the standard gear segment part image, the quality assessment parameter of each gear segment part is obtained; Based on the quality assessment parameters, the production quality of each gear segment component is assessed.
2. A visual measurement method for the profile of a bulldozer gear segment according to claim 1, characterized in that: The noise possibilities of obtaining the target closed edge line include: Analyze the discrete degree of the gradient amplitude of all pixel points on the target closed edge line to obtain the first noise parameter of the target closed edge line; Performing negative correlation mapping on the number of all pixel points in the inner area of the target closed edge line to obtain a second noise parameter of the target closed edge line; Acquire an eight-chain code sequence of a target closed edge line, and obtain a third noise parameter of the target closed edge line according to differences between adjacent elements in the eight-chain code sequence; The first noise parameter, the second noise parameter and the third noise parameter of the target closed edge line are integrated to obtain the noise possibility of the target closed edge line.
3. The visual measurement method for the profile of a bulldozer gear segment according to claim 2, characterized in that: The step of obtaining the third noise parameter of the target closed edge line according to the difference between adjacent elements in the eight-chain code sequence comprises: In the eight-chain code sequence, the absolute value of the difference between each element and the next adjacent element is taken as the variation of each element; The average value of the variation of all elements in the eight-chain code sequence is used as the third noise parameter of the target closed edge line.
4. The visual measurement method for the profile of a bulldozer gear segment according to claim 1, characterized in that: The step of obtaining the contour authenticity of the target closed edge line comprises: Take any pixel point on the target closed edge line as the target pixel point; In a preset window centered on the target pixel point, the pixel points belonging to the inner area are used as the inner pixel points of the preset window, and the pixel points belonging to the outer area are used as the outer pixel points of the preset window; The average grayscale value of all inner pixels in the preset window is used as the inner overall grayscale value of the preset window, and the average grayscale value of all outer pixels in the preset window is used as the outer overall grayscale value of the preset window; the absolute value of the difference between the inner overall grayscale value and the outer overall grayscale value is used as the grayscale difference value of the target pixel; The overall level of the grayscale difference values of all pixels on the target closed edge line is analyzed to obtain the contour authenticity of the target closed edge line.
5. The visual measurement method for the profile of a bulldozer gear segment according to claim 1, characterized in that: The possibilities of obtaining the contour of the target closed edge line include: Performing negative correlation mapping on the noise possibility to obtain a contour judgment value of a target closed edge line; The contour judgment value and the contour authenticity are integrated and normalized to obtain the contour possibility of the target closed edge line.
6. The visual measurement method for the profile of a bulldozer gear segment according to claim 1, characterized in that: The step of selecting the contour edge line from all closed edge lines of the target grayscale image based on the contour possibility comprises: In the target grayscale image, the closed edge lines whose contour likelihood is greater than a preset likelihood threshold are taken as contour edge lines.
7. The visual measurement method for the profile of a bulldozer gear segment according to claim 1, characterized in that: The quality assessment parameters of each gear segment component are obtained as follows: Performing image alignment processing on the target grayscale image and the standard gear segment part image, and obtaining the coincident pixel points on the contour edge line of the target grayscale image, wherein the positions of the coincident pixel points are the same as the positions of the pixel points on the standard contour line of the standard gear segment part image; The number of all the overlapping pixel points on the contour edge line of the target grayscale image is normalized to obtain the quality assessment parameter of the gear segment part corresponding to the target grayscale image.
8. The visual measurement method for the profile of a bulldozer gear segment according to claim 1, characterized in that: The evaluating of the production quality of each gear segment part based on the quality evaluation parameter includes: If the quality assessment parameter is greater than the preset assessment threshold, the block part is qualified for production, otherwise, the gear block part is unqualified for production.
9. The visual measurement method for the profile of a bulldozer gear segment according to claim 1, characterized in that: The performing edge detection on the target grayscale image to obtain a plurality of closed edge lines in the target grayscale image comprises: Using the Canny edge detection algorithm, the target grayscale image is processed to obtain multiple edge lines in the target grayscale image; Using the endpoint detection algorithm, each edge line of the target grayscale image is processed to obtain the number of endpoints of each edge line; The edge line with the number of endpoints equal to the value 1 is regarded as a closed edge line.
10. A visual measurement system for the profile of a bulldozer gear segment, the system 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 the method according to any one of claims 1 to 9 are implemented.
Citation Information
Patent Citations
Gear surface defect detection method and system based on image processing
CN113658133A
Extrusion molding rubber sealing ring surface abnormity detection method based on edge detection
CN114926404A
Welding image molten pool edge detection method
CN115330774A
Environment-friendly and energy-saving aluminum silicate plate quality detection method
CN116168039A
Intelligent detection method for quality of concrete prefabricated bottom plate
CN116703910A
Cited By
3D printing modeling method and system for vehicle parts
CN119636077A
Abnormal region screening method, device and equipment, medium and program product
CN120088446A
An abnormal area screening method, device, equipment, medium and program product
CN120088446B
