CN117808799A - Chamfering equipment processing quality detection method based on artificial intelligence - Google Patents
Chamfering equipment processing quality detection method based on artificial intelligence Download PDFInfo
- Publication number
- CN117808799A CN117808799A CN202410223783.3A CN202410223783A CN117808799A CN 117808799 A CN117808799 A CN 117808799A CN 202410223783 A CN202410223783 A CN 202410223783A CN 117808799 A CN117808799 A CN 117808799A
- Authority
- CN
- China
- Prior art keywords
- hole
- pixel point
- burr
- pixel points
- area
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
- 238000001514 detection method Methods 0.000 title claims abstract description 50
- 238000012545 processing Methods 0.000 title claims abstract description 42
- 238000013473 artificial intelligence Methods 0.000 title claims abstract description 22
- 238000005286 illumination Methods 0.000 claims abstract description 108
- 238000012937 correction Methods 0.000 claims abstract description 47
- 238000009826 distribution Methods 0.000 claims abstract description 11
- 238000003709 image segmentation Methods 0.000 claims abstract description 5
- 230000007547 defect Effects 0.000 claims description 44
- 238000000034 method Methods 0.000 claims description 27
- 238000013507 mapping Methods 0.000 claims description 15
- 238000004364 calculation method Methods 0.000 claims description 12
- 230000003044 adaptive effect Effects 0.000 claims description 8
- 230000002159 abnormal effect Effects 0.000 claims description 6
- 238000010606 normalization Methods 0.000 claims description 6
- 238000003708 edge detection Methods 0.000 claims description 4
- 238000004458 analytical method Methods 0.000 abstract description 10
- 238000003384 imaging method Methods 0.000 abstract description 10
- 239000000047 product Substances 0.000 description 47
- 230000000875 corresponding effect Effects 0.000 description 9
- 238000010586 diagram Methods 0.000 description 6
- 239000002184 metal Substances 0.000 description 6
- 238000003672 processing method Methods 0.000 description 4
- 238000002360 preparation method Methods 0.000 description 3
- 230000009286 beneficial effect Effects 0.000 description 2
- 230000000694 effects Effects 0.000 description 2
- 238000009499 grossing Methods 0.000 description 2
- 230000001788 irregular Effects 0.000 description 2
- 238000004519 manufacturing process Methods 0.000 description 2
- 230000002596 correlated effect Effects 0.000 description 1
- 230000000750 progressive effect Effects 0.000 description 1
- 238000012216 screening Methods 0.000 description 1
- 230000035945 sensitivity Effects 0.000 description 1
- 239000013589 supplement Substances 0.000 description 1
- 230000001502 supplementing effect Effects 0.000 description 1
Landscapes
- Image Processing (AREA)
- Investigating Materials By The Use Of Optical Means Adapted For Particular Applications (AREA)
- Image Analysis (AREA)
Abstract
The invention relates to the technical field of image segmentation and texture analysis, in particular to a chamfering equipment processing quality detection method based on artificial intelligence. Firstly, acquiring a surface gray level image of a product processed by chamfering equipment; further analyzing distribution characteristics and gray change characteristics of the hole pixel points in the hole area to obtain correction burr degree of each hole pixel point; further analyzing gradient amplitude values of pixel points at the edge of the hole, and combining differences of the hole areas and other hole areas to obtain corrected illumination intensity of each hole area; and further acquiring a self-adaptive neighborhood region of each hole pixel point, and detecting the quality of a processed product of chamfering equipment. According to the invention, through analyzing the gray characteristics of each hole pixel point and surrounding pixel points, the influence of illumination factors on hole region imaging is obtained, the self-adaptive neighborhood region of each hole pixel point is obtained, the interference of strong light factors on quality detection is reduced, and the processing quality detection of chamfering equipment is more accurate.
Description
Technical Field
The invention relates to the technical field of image segmentation and texture analysis, in particular to a chamfering equipment processing quality detection method based on artificial intelligence.
Background
When detecting the product quality of chamfering equipment processing, the hole area on the product surface can exist to the incomplete condition of burr defect removal, and the existence of burr can not only destroy the outward appearance of product, can also lead to the product surface to have sharp edge or protruding and then can cause the potential safety hazard to influence product quality. The existing method for detecting the burr defects is a Markov random field (Markov Random Field, MRF) algorithm, which can utilize the spatial relationship among pixels in an image, and the correlation among different areas in the image can be better analyzed by modeling the whole image, so that the defect areas can be accurately detected.
When the image of the chamfer product is acquired, as the chamfer product is made of metal, the metal surface is smooth, a strong light area caused by light reflection is easy to generate, the gray value of the strong light area is usually higher, when the strong light area is overlapped with a hole area to be detected, the contrast of the edge of the hole area is lowered, the neighborhood parameter setting of the MRF algorithm is unreasonable, and the quality detection of the processed product of the chamfer equipment is affected.
Disclosure of Invention
In order to solve the technical problem that the imaging of a hole area is affected by illumination factors, so that accurate quality detection of a processed product cannot be performed, the invention aims to provide an artificial intelligence-based chamfering equipment processing quality detection method, which adopts the following specific technical scheme:
acquiring a surface gray level image of a product processed by chamfering equipment; acquiring a hole area and hole edge pixel points of the surface gray level image;
acquiring the possibility of burr defects of each hole area according to the distribution characteristics of the hole edge pixel points of each hole area; according to the gray scale change characteristics of the pixel points in the preset neighborhood of each hole pixel point in each hole region, combining the gray scale characteristics of each hole pixel point to obtain the burr degree of each hole pixel point; combining the burr degree of each hole pixel point and the burr defect possibility of the hole region to obtain a corrected burr degree of each hole pixel point;
acquiring initial illumination intensity of each hole area according to statistical characteristics and fluctuation characteristics of gradient amplitude values of hole edge pixel points of each hole area; according to the difference characteristics of the initial illumination intensity of each hole area and other hole areas, the initial illumination intensity is corrected by combining the difference characteristics of the correction burr degrees of all hole edge pixel points in each hole area and other hole areas, and the corrected illumination intensity of each hole area is obtained;
according to the corrected burr degree of each hole pixel point and the corrected illumination intensity of the hole region, combining preset neighborhood size parameters to obtain a self-adaptive neighborhood region of each hole pixel point;
and processing the surface gray level image according to the self-adaptive neighborhood region of each hole pixel point, and finally realizing quality detection.
Further, the method for acquiring the possibility of the burr defect comprises the following steps:
and connecting adjacent hole edge pixel points, obtaining the angle values of the connecting lines of the adjacent hole edge pixel points in a plane rectangular coordinate system, summing the absolute values of the differences of the angle values of all the adjacent connecting lines, and taking the sum result as the possibility of burr defects.
Further, the method for acquiring the burr degree comprises the following steps:
taking the ratio of the gray value of the hole pixel point to the average gray value of the surface gray image as the gray characteristic parameter of the hole pixel point;
traversing the pixel points in a preset adjacent area of the hole pixel points according to a set sequence, counting the number of times of changing the gray value of the adjacent pixel points, and taking the product of the number of times of changing and the corresponding gray characteristic parameter as the burr degree of the corresponding hole pixel points.
Further, the method for acquiring the initial illumination intensity comprises the following steps:
taking the sum of gradient magnitudes of all hole edge pixel points in each hole area as a first illumination parameter;
summing the absolute values of the differences between the gradient magnitudes and the gradient magnitude mean values of all the hole edge pixel points in each hole area, and taking the mapping result as a second illumination parameter after the negative correlation of the summation result and the mapping by using an exponential function;
taking the product of the first illumination parameter and the second illumination parameter as an initial illumination intensity.
Further, the method for obtaining the corrected illumination intensity comprises the following steps:
obtaining corrected illumination intensity by using an initial illumination intensity correction formula; the initial illumination intensity correction formula comprises:
the method comprises the steps of carrying out a first treatment on the surface of the Wherein (1)>Indicate->The corrected illumination intensity of the individual hole areas; />Indicate->Initial illumination intensity of individual hole areas; />Representing a normalization function; />Indicate->In each hole area, the sum of the correction burr degrees of all hole edge pixel points;representation of the surface greyscale image except +.>The number of other hole areas outside the hole areas; />Indicating removal of->Outside the hole area->Initial illumination intensity of the other hole areas; />Indicating removal of->Outside the hole area->And in other hole areas, the sum of the correction burr degrees of all hole edge pixel points.
Further, the method for acquiring the adaptive neighborhood region comprises the following steps:
obtaining the self-adaptive neighborhood size parameter of each hole pixel point by utilizing a self-adaptive neighborhood size parameter calculation formula; establishing a square neighborhood region by taking each hole pixel point as a center and taking the corresponding self-adaptive neighborhood size parameter as a side length to obtain the self-adaptive neighborhood region of each hole pixel point; the adaptive neighborhood size parameter calculation formula comprises:
the method comprises the steps of carrying out a first treatment on the surface of the Wherein (1)>Indicate->First part of the hole area>Self-adaptive neighborhood size parameters of the pixel points of the holes; />Representing an upward rounding function; />Representing a preset neighborhood size parameter; />Representing a normalization function; />First->Within the hole region->Correcting burr degree of each hole pixel point; />Indicate->The corrected illumination intensity of the individual hole areas; />Representing a preset scaling factor, and +.>。
Further, the method for processing the surface gray level image and finally realizing quality detection comprises the following steps:
and according to the self-adaptive neighborhood region of each hole pixel point, performing image segmentation on a surface gray level image of a chamfering equipment processing product by utilizing an MRF algorithm to obtain a hole region to be analyzed, calculating the correction burr degree of each pixel point in the hole region to be analyzed, normalizing to obtain an abnormal score of each pixel point in the hole region to be analyzed, and when the abnormal score of each pixel point is larger than a preset burr threshold value, determining the hole region to be analyzed as a burr region, otherwise, determining the hole region to be analyzed as a qualified region.
Further, the method for acquiring the corrected burr degree comprises the following steps:
and taking the product of the burr degree of each hole pixel point and the burr defect possibility of the hole area as the corrected burr degree of each hole pixel point.
Further, the method for obtaining the hole edge pixel points comprises the following steps:
performing edge detection by using a Canny operator to obtain all edge pixel points of the surface gray level image; and detecting holes in the image by using a Hough circle detection algorithm, and taking edge pixel points at the edges of the holes as hole edge pixel points.
Further, the method for acquiring the hole area comprises the following steps:
selecting a first preset number of pixel points on an extension line of the gradient direction of each hole edge pixel point contained in the hole, selecting a second preset number of pixel points on an extension line of the opposite direction of the gradient direction, and taking the hole edge pixel point of each hole and the pixel points selected by the hole edge pixel point as a hole area.
The invention has the following beneficial effects:
firstly, acquiring a surface gray level image of a processed product of chamfering equipment, and providing an analysis basis for subsequent analysis and quality detection; further acquiring hole areas and hole edge pixel points, determining a focus area for quality detection, and eliminating interference of other areas; further, according to the distribution characteristics of the hole edge pixel points of each hole area, the possibility of burr defect of the hole area is measured from the regularity of the hole edge, and a basis is provided for the subsequent correction of the burr degree of the hole pixel points; further analyzing the highlighting degree and the local irregularity degree of the hole pixel points according to the gray level change in the local neighborhood of the hole pixel points and the gray level characteristics of the hole pixel points, acquiring the burr degree of each hole pixel point, measuring the burr degree of the hole pixel points by combining the local characteristics and the global characteristics, and providing a correction basis for correcting the possibility of the subsequent combined burr defects; further correcting the burr degree by utilizing the burr defect possibility, supplementing the distribution characteristics of the edges of the region, obtaining more accurate corrected burr degree, and preparing for the subsequent acquisition of the self-adaptive neighborhood region; further analyzing the statistical characteristics and the fluctuation characteristics of the gradient amplitude values of the pixel points at the edge of each hole area, acquiring the initial illumination intensity of each hole area from the coincidence degree of the edge when the hole is in the strong light area, primarily measuring the illumination intensity characteristics of the hole area, and simultaneously providing a basis for further correction; the difference of the initial illumination intensity and the correction burr degree of the hole pixel points in the area in the hole area and other hole areas is further combined, the initial illumination intensity is corrected, more abundant context information is obtained, objects and structures in the image are better analyzed, so that the accuracy of results is improved, and the robustness of the quality detection method is improved; further, according to the correction burr degree of each hole pixel point and the correction illumination intensity of the hole region, the influence of illumination factors on the hole region is considered, the self-adaptive neighborhood region of each hole pixel point is obtained, and preparation is made for final quality detection; and finally, detecting the quality of the processed product of the chamfering equipment according to the self-adaptive neighborhood region of each hole pixel point. According to the invention, through analyzing the gray characteristics of each hole pixel point and surrounding pixel points, the influence of illumination factors on hole region imaging is obtained, the self-adaptive neighborhood region of each hole pixel point is obtained, the interference of strong light factors on quality detection is reduced, and the processing quality detection of chamfering equipment is more accurate.
Drawings
In order to more clearly illustrate the embodiments of the invention or the technical solutions and advantages of the prior art, the following description will briefly explain the drawings used in the embodiments or the description of the prior art, and it is obvious that the drawings in the following description are only some embodiments of the invention, and other drawings can be obtained according to the drawings without inventive effort for a person skilled in the art.
FIG. 1 is a flow chart of a chamfering device processing quality detection method based on artificial intelligence according to an embodiment of the present invention;
FIG. 2 is an exemplary illustration of a chamfering tool tooling product according to one embodiment of the present invention;
FIG. 3 is an exemplary illustration of a chamfering tool tooling product according to one embodiment of the present invention;
FIG. 4 is a schematic diagram illustrating connection of pixel points at the edge of a hole without burr defect hole area according to an embodiment of the present invention;
FIG. 5 is a schematic diagram illustrating connection of hole edge pixels in a hole region with burr defects according to an embodiment of the present invention;
fig. 6 is an exemplary view of a chamfering apparatus tooling product according to one embodiment of the present invention.
Detailed Description
In order to further describe the technical means and effects adopted by the chamfering device for achieving the preset aim of the invention, the following detailed description is given of a chamfering device processing quality detection method based on artificial intelligence according to the invention, and the detailed description of the specific implementation, structure, characteristics and effects thereof is given below in combination with the accompanying drawings and the preferred embodiment. In the following description, different "one embodiment" or "another embodiment" means that the embodiments are not necessarily the same. Furthermore, the particular features, structures, or characteristics of one or more embodiments may be combined in any suitable manner.
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.
Chamfering equipment processing quality detection method embodiment based on artificial intelligence
The invention provides a chamfering equipment processing quality detection method based on artificial intelligence, which is concretely described below with reference to the accompanying drawings.
Referring to fig. 1, a flowchart of a chamfering device processing quality detection method based on artificial intelligence according to an embodiment of the present invention specifically includes:
step S1: acquiring a surface gray level image of a product processed by chamfering equipment; and acquiring a hole area and hole edge pixel points of the surface gray level image.
In the embodiment of the invention, the gray level image of the surface of the chamfering equipment processed product is obtained in consideration of the fact that the gray level image has smaller data size compared with the color image, the calculated amount is smaller during calculation and analysis, the sensitivity of the image to illumination change can be reduced to a certain extent, and the robustness of the quality detection method is improved. Because the burr of the hole is detected, the hole area and the hole edge pixel point need to be acquired first for subsequent analysis.
Preferably, in one embodiment of the invention, the Canny operator is considered to have strong adaptability and simple implementation process, so that the Canny operator is utilized to detect the edge pixel points; considering that the holes of a chamfering device processing product are round, detecting the holes in the image by using a Hough circle detection algorithm, and screening out hole edge pixel points; in addition, after the imaging of the hole burrs actually processed is considered, the imaging is not a single pixel point, and is not a single circle of pixel points, the hole area is required to be acquired for analysis, and as the pixel points are selected in the gradient direction of the pixel points, the pixel points are selected in the gradient direction of the pixel points to form the hole area because the pixel points can help to capture the edge information and the texture information in the image, and the hole area is formed based on the pixel points in the gradient direction of the pixel points at the edge of the hole:
all edge pixel points of the surface gray image are obtained by utilizing the Canny operator to carry out edge detection; and detecting holes in the image by using a Hough circle detection algorithm, and taking edge pixel points at the edges of the holes as hole edge pixel points. Selecting a first preset number of pixel points on an extension line of the gradient direction of each hole edge pixel point contained in the hole, selecting a second preset number of pixel points on an extension line of the opposite direction of the gradient direction, and taking the hole edge pixel point of each hole and the pixel points selected by the hole edge pixel point as a hole area.
It should be noted that, the Canny operator, the hough circle detection algorithm, and the gradient direction for obtaining the pixel point are all technical means well known to those skilled in the art, and are not described herein again; in one embodiment of the invention, the first preset number is 10 and the second preset number is 3; in other embodiments of the present invention, the practitioner may set other first preset numbers and second preset numbers, and acquire hole edge pixels by using other edge detection algorithms.
Step S2: acquiring the possibility of burr defects of each hole area according to the distribution characteristics of the hole edge pixel points of each hole area; according to the gray scale change characteristics of the pixel points in the preset neighborhood of each hole pixel point in each hole area, combining the gray scale characteristics of each hole pixel point to obtain the burr degree of each hole pixel point; and combining the burr degree of each hole pixel point and the burr defect possibility of the hole region, and acquiring the corrected burr degree of each hole pixel point.
In the embodiment of the invention, the greater the possibility that the hole pixel point is a burr pixel point is considered, the greater the possibility that the burr defect exists in the hole area where the pixel point is positioned is indicated, and more local information is captured in a larger neighborhood range, so that in order to obtain a more accurate self-adaptive neighborhood area later, accurate quality detection is carried out on a processed product of chamfering equipment, and the burr correction degree of the hole pixel point is obtained first.
Referring to fig. 2 and 3, an exemplary diagram of a chamfering apparatus according to an embodiment of the present invention is shown, wherein fig. 2 includes a portion having a burr defect and fig. 3 includes a portion having no burr defect. As can be seen from fig. 2 and 3, due to the influence of the burrs, the hole edges are irregular, the distribution of the hole edge pixels is more irregular, and the burrs influence the flatness of the surface of the processed product, so that the reflection of the surface of the product in the hole area with the burrs is inconsistent, the gray level change of the local pixels is more intense, and the gray level value of the pixel in the burr area is larger than that of the other pixels, so that the distribution characteristics of the hole edge pixels, the gray level characteristics of the hole pixels and the gray level change in the local vicinity of the gray level pixels can be combined to obtain the corrected burr degree of each hole pixel.
It should be noted that, the pixel points in the hole area are called hole pixel points, and the hole pixel points include hole edge pixel points.
Preferably, in an embodiment of the present invention, considering that when burrs occur in a hole area, the connection line of adjacent hole edge pixel points changes more severely, referring to fig. 4 and 5, fig. 4 shows a hole edge pixel point connection schematic diagram of a hole area without burrs defect provided in an embodiment of the present invention, fig. 5 shows a hole edge pixel point connection schematic diagram of a hole area with burrs defect provided in an embodiment of the present invention, and it can be seen by comparing fig. 4 and 5 that the connection line of the hole area without burrs defect forms a relatively complete circle, and the angle change between adjacent connection lines is relatively uniform; the connecting lines of the hole areas with the burr defects are more tortuous, and the angle change between the adjacent connecting lines is larger, so that the adjacent hole edge pixel points are connected, the angle values of the connecting lines of the adjacent hole edge pixel points in a plane rectangular coordinate system are obtained, the absolute values of the difference values of the angle values of all the adjacent connecting lines are summed, and the summation result is used as the burr defect possibility. The larger the corresponding summation result of the hole area is, the greater the possibility of burr defect is, and the greater the possibility of burr exists in the hole area is.
It should be noted that, the connection direction of the hole edge pixel points has no influence on the analysis result, in one embodiment of the present invention, clockwise connection is adopted, and because the hole edge pixel points are connected end to end, any connection line can be selected as a starting point, and all adjacent connection lines can be traversed.
Preferably, in one embodiment of the present invention, it is considered that the larger the gray value of the hole pixel point is relative to the whole image, the more likely the burr pixel point is; meanwhile, the more the gray value fluctuation times of adjacent pixels in the local neighborhood of the hole pixel point are considered, the more severe the fluctuation of the pixels in the local neighborhood is, the more likely the pixels are in the burr area, and the greater the possibility that the hole pixel point is the burr pixel point is; based on the gray value ratio of the gray value of the hole pixel point and the average gray value of the surface gray image is used as the gray characteristic parameter of the hole pixel point;
and traversing the pixel points in a preset adjacent area of the hole pixel points according to a set sequence, counting the number of times of changing the gray value of the adjacent pixel points, and taking the product of the number of times of changing and the corresponding gray characteristic parameter as the burr degree of the corresponding hole pixel points. The calculation formula of the burr degree comprises:
wherein,indicate->First part of the hole area>The burr degree of each hole pixel point; />Indicate->First part of the hole area>Gray values of the pixel points of the holes; />An average gray value representing the gray image of the entire surface;indicate->First part of the hole area>The number of times the gray value of the adjacent pixel changes in the preset neighborhood of the pixel of each hole.
It should be noted that, in one embodiment of the present invention, in order to avoid omission during statistics of gray level variation times of adjacent pixels, the pixels are traversed from left to right and from top to bottom in a preset neighborhood of the hole pixels, and the size of the preset neighborhood isConstructing a preset neighborhood of the hole pixel point by taking the hole pixel point as a center and the preset neighborhood size; in other embodiments of the present invention, an implementer may set a neighborhood range of other sizes, traverse the pixel points in the neighborhood in other manners, count the number of gray level changes of the adjacent pixel points, and implement related mapping by using other basic mathematical operations or function mapping, which are all technical means well known to those skilled in the art, and are not described herein.
The possibility of burr defect represents the possibility that the whole hole area is a defect area, and the burr degree of the hole pixel point reflects the local characteristic that the pixel point is the burr pixel point, so that the whole and local characteristics are combined, the burr degree of each hole pixel point is corrected more accurately, and a more reliable basis is provided for subsequent analysis.
Preferably, in one embodiment of the present invention, considering that the probability of the burr defect and the probability of the burr of the hole pixel point are in positive correlation, the product of the burr degree of each hole pixel point and the probability of the burr defect of the hole region is taken as the corrected burr degree of each hole pixel point.
It should be noted that, in other embodiments of the present invention, other basic mathematical operations or function mapping may be used to implement the related mapping, which are all technical means known to those skilled in the art, and are not described herein.
Step S3: acquiring the initial illumination intensity of each hole area according to the statistical characteristics and the fluctuation characteristics of the gradient amplitude of the hole edge pixel points of each hole area; and according to the difference characteristics of the initial illumination intensity of each hole area and other hole areas, the initial illumination intensity is corrected by combining the difference characteristics of the correction burr degree of all hole edge pixel points in each hole area and other hole areas, and the corrected illumination intensity of each hole area is obtained.
Referring to fig. 6, an example diagram of a chamfering device processing product provided by an embodiment of the present invention is shown, fig. 6 includes a defect portion interfered by light, and it can be seen from fig. 6 that, under strong light interference, the overall gray value of the burr defect area is higher, the contrast in the burr defect area is lower, which is not beneficial to analyzing the texture feature and gray variation of the burr defect area, and at this time, the reliability of the burr correction degree of the hole pixel point is lower, so that the light feature of the hole area needs to be analyzed, and the burr correction degree needs to be further supplemented, so that a more suitable adaptive neighborhood size is obtained later.
As can be seen from fig. 6, under the strong light interference, the gradient amplitude of the hole edge pixel points in the hole area is larger, and the consistency is stronger, so that the initial illumination intensity of each hole area can be obtained according to the statistical characteristic and the fluctuation characteristic of the gradient amplitude of the hole edge pixel points.
Preferably, in one embodiment of the present invention, the sum of the gradient magnitudes of all the hole edge pixels in each hole area is used as the first illumination parameter;
summing absolute values of differences between gradient magnitudes and gradient magnitude mean values of all hole edge pixel points in each hole area, and taking a mapping result as a second illumination parameter after the sum result is inversely related and mapped by using an exponential function;
taking the product of the first illumination parameter and the second illumination parameter as the initial illumination intensity.
The calculation formula of the initial illumination intensity comprises:
wherein,indicate->Initial illumination intensity of individual hole areas; />A first illumination parameter is indicated and a first illumination parameter is indicated,，/>indicate->First part of the hole area>Gradient amplitude values of pixel points at the edges of the holes;representing a second illumination parameter,/->，/>Expressed as natural constant->An exponential function of the base +.>Indicate->The average value of the gradient magnitudes of all the hole edge pixels of the hole areas,indicate->The number of hole edge pixels in each hole area.
In the calculation formula of the initial illumination intensity, the larger the first illumination parameter is, the larger the sum of gradient magnitudes of pixel points at the edge of the hole is, the larger the difference between the pixel points at the edge of the current hole area and the area from which metal is removed is, the more obvious the characteristics of metal reflected light rays of the hole area are, the larger the illumination intensity of the current hole area is, and the larger the initial illumination intensity is; the larger the second parameter is, the smaller the fluctuation of the gradient amplitude of the pixel points at the edge of the hole is, the larger the degree of influence of illumination is, the larger the illumination intensity of the current hole area is, and the larger the initial illumination intensity is.
It should be noted that, in other embodiments of the present invention, other basic mathematical operations or function mapping may be used to implement the related mapping, which are all technical means known to those skilled in the art, and are not described herein.
In the embodiment of the invention, the larger the initial illumination intensity of the current hole area relative to other hole areas is, the stronger the illumination intensity of the current hole area is, the greater the possibility that the pixel points in the current hole area are burr pixel points is, the greater the importance degree of the illumination intensity to the current area is reflected, and the influence degree is greater, so that the difference of the initial illumination intensity and the correction burr degree of the hole pixel points in the hole area is combined with the difference of the initial illumination intensity and the correction burr degree of the hole pixel points in the hole area, the initial illumination intensity is corrected, more abundant context information is obtained, better analysis of objects and structures in an image is facilitated, the accuracy of results is improved, and the robustness of the quality detection method is improved.
Preferably, in one embodiment of the present invention, considering that the illumination intensity has an influence on the burr feature of the hole pixel point in the hole area, and that the greater the possibility of the overall correction burr in the hole area, the greater the possibility that the hole area is in the defect area, the more important the influence of the illumination intensity is, so that the initial illumination intensity and the overall correction burr possibility are positively correlated, so that the difference feature of the product of the initial illumination intensity and the overall correction burr possibility can be analyzed, an initial illumination intensity correction formula is constructed, and the corrected illumination intensity is obtained by using the initial illumination intensity correction formula; the initial illumination intensity correction formula includes:
wherein,indicate->The corrected illumination intensity of the individual hole areas; />Indicate->Initial illumination intensity of individual hole areas; />Representing a normalization function; />Indicate->In each hole area, the sum of the correction burr degrees of all hole edge pixel points; />Representation of the surface greyscale image except +.>The number of other hole areas outside the hole areas;indicating removal of->Outside the hole area->Initial illumination intensity of the other hole areas; />Indicating removal of->Outside the hole area->And in other hole areas, the sum of the correction burr degrees of all hole edge pixel points.
It should be noted that, in other embodiments of the present invention, the initial illumination intensity and the overall burr correction degree in the hole area may be separately compared, and finally, the two difference results are multiplied and normalized, or other basic mathematical operations or function mapping may be selected to implement the relevant mapping, which are all technical means well known to those skilled in the art, and are not described herein.
Step S4: and according to the correction burr degree of each hole pixel point and the correction illumination intensity of the hole region to which the correction burr degree belongs, and combining preset neighborhood size parameters, obtaining a self-adaptive neighborhood region of each hole pixel point.
Through steps S1 to S3, the correction burr degree of each hole area pixel point and the correction illumination degree of the corresponding hole area are obtained, the two characteristic indexes integrate the local characteristic of the hole area pixel point and the global characteristic of the whole image, and the preparation is carried out for obtaining the self-adaptive neighborhood area of each hole pixel point, so that the self-adaptive neighborhood area is obtained immediately, and the preparation is carried out in advance for final quality detection.
Preferably, in one embodiment of the present invention, considering that the greater the burr correction degree of the hole pixel point, the greater the possibility that the burr defect exists in the hole area where the hole pixel point is located, the greater the neighborhood range is required to capture more local information, so as to facilitate subsequent defect detection; the greater the corrected illumination degree of the hole area corresponding to the hole pixel point is, the greater the influence degree of illumination reflection on the hole pixel point in the area is, the greater the possibility that the expression degree of the hole pixel point on the burr defect is interfered by the illumination intensity is, so that the illumination intensity of the hole area is utilized to supplement the burr feature of the hole pixel point of the hole area to a certain extent;
based on the self-adaptive neighborhood size parameter calculation formula, the self-adaptive neighborhood size parameter of each hole pixel point is obtained by utilizing the self-adaptive neighborhood size parameter calculation formula; taking each hole pixel point as a center, taking a corresponding self-adaptive neighborhood size parameter as a side length, and establishing a square neighborhood region to obtain the self-adaptive neighborhood region of each hole pixel point; the adaptive neighborhood size parameter calculation formula comprises:
wherein,indicate->First part of the hole area>Self-adaptive neighborhood size parameters of the pixel points of the holes; />Representing an upward rounding function; />Representing a preset neighborhood size parameter; />Representing a normalization function; />First->Within the hole region->Correcting burr degree of each hole pixel point; />Indicate->The corrected illumination intensity of the individual hole areas; />Representing a preset scaling factor, and +.>。
It should be noted that, in the calculation formula of the adaptive neighborhood size parameter, the preset scaling factor is used to regulate and control the correction range of the preset neighborhood size parameter, and control the value range of the adaptive neighborhood size parameter, in one embodiment of the present invention,，/>in other embodiments of the invention, the practitioner may set other +.>And->Other basic mathematical operations or function mapping may be used to implement the correlation mapping, which are all technical means known to those skilled in the art, and are not described herein.
Step S5: and processing the surface gray level image according to the self-adaptive neighborhood region of each hole pixel point, and finally realizing quality detection.
After the self-adaptive neighborhood region is obtained in the step S4, quality detection can be carried out on the processed product of the chamfering equipment.
Preferably, in one embodiment of the present invention, considering that the spatial relationship between pixels in an image can be utilized by an MRF algorithm, by modeling the whole image, the correlation between different areas in the image can be better understood, the defect area is helped to be accurately detected, and the probability that a pixel belongs to a burr pixel can be measured by correcting the burr degree, so that according to the self-adaptive neighborhood area of each hole pixel, the image segmentation is performed on the surface gray image of the chamfering equipment processing product by utilizing the MRF algorithm, the hole area to be analyzed is obtained, the corrected burr degree of each pixel in the hole area to be analyzed is calculated and normalized, the abnormal score of each pixel in the hole area to be analyzed is obtained, and when the abnormal score of the pixel is greater than the preset burr threshold, the hole area to be analyzed is considered as the burr area, otherwise, the hole area to be analyzed is qualified.
In one embodiment of the present invention, the smoothing coefficient of the MRF algorithm is 0.8, the preset burr threshold is 0.6, and in other embodiments of the present invention, the practitioner may set other suitable smoothing coefficients and preset burr thresholds; it should be noted that the MRF algorithm is a technical means well known to those skilled in the art, and will not be described herein.
After quality detection is carried out on the processed product of the chamfering equipment, the burr area can be marked, related personnel are reminded or automatic repairing equipment is utilized for repairing, the occurrence frequency of the burr area can be counted, and when the frequency is higher than a certain threshold value, related personnel are reminded to check the production state and parameters of the equipment in time, so that the production quality is guaranteed.
In summary, the invention provides an artificial intelligence-based chamfering equipment processing quality detection method aiming at the technical problem that the imaging of a hole area is affected by illumination factors, so that the accurate quality detection of a processed product cannot be performed. Firstly, acquiring a surface gray level image of a product processed by chamfering equipment, and determining a hole area and hole edge pixel points; further analyzing the distribution characteristics of the hole edge pixel points, the gray level change characteristics of the pixel points in the preset neighborhood of each hole pixel point and the gray level characteristics of each hole pixel point, and obtaining the correction burr degree of each hole pixel point; further according to the statistical characteristics and the fluctuation characteristics of the gradient amplitude values of the hole edge pixel points, the difference of the hole areas and other hole areas is combined, and the corrected illumination intensity of each hole area is obtained; and further acquiring a self-adaptive neighborhood region of each hole pixel point, and detecting the quality of a processed product of chamfering equipment. According to the invention, the gray characteristics of each hole pixel point and surrounding pixel points are analyzed, the influence of illumination factors on hole region imaging is obtained, and the self-adaptive neighborhood region of each hole pixel point is obtained, so that the processing quality detection of chamfering equipment is more accurate.
Image processing method embodiment for chamfering equipment processing product
When the image of the chamfer product is acquired, as the chamfer product is made of metal, the metal surface is smooth, a strong light area caused by light reflection is easy to generate, the gray value of the strong light area is usually higher, when the strong light area is overlapped with a hole area to be analyzed, the contrast of the edge of the hole area is lowered, the neighborhood parameter setting of the MRF algorithm is unreasonable, and the determination of the hole area in the image of the chamfer equipment processing product is affected.
In order to solve the technical problem that an accurate hole area of a processed product cannot be obtained due to the fact that an illumination factor affects hole area imaging, the invention aims to provide an image processing method for a processed product of chamfering equipment, and the adopted technical scheme is as follows:
step S1: acquiring a surface gray level image of a product processed by chamfering equipment; and acquiring a hole area and hole edge pixel points of the surface gray level image.
Step S2: acquiring the possibility of burr defects of each hole area according to the distribution characteristics of the hole edge pixel points of each hole area; according to the gray scale change characteristics of the pixel points in the preset neighborhood of each hole pixel point in each hole area, combining the gray scale characteristics of each hole pixel point to obtain the burr degree of each hole pixel point; and combining the burr degree of each hole pixel point and the burr defect possibility of the hole region, and acquiring the corrected burr degree of each hole pixel point.
Step S3: acquiring the initial illumination intensity of each hole area according to the statistical characteristics and the fluctuation characteristics of the gradient amplitude of the hole edge pixel points of each hole area; and according to the difference characteristics of the initial illumination intensity of each hole area and other hole areas, the initial illumination intensity is corrected by combining the difference characteristics of the correction burr degree of all hole edge pixel points in each hole area and other hole areas, and the corrected illumination intensity of each hole area is obtained.
Step S4: and according to the correction burr degree of each hole pixel point and the correction illumination intensity of the hole region to which the correction burr degree belongs, and combining preset neighborhood size parameters, obtaining a self-adaptive neighborhood region of each hole pixel point.
Step S5: and processing the surface gray level image according to the self-adaptive neighborhood region of each hole pixel point to obtain a hole region to be analyzed.
Because the specific implementation process of the image processing method for the chamfering equipment processing product is already given in the above-mentioned chamfering equipment processing quality detection method based on artificial intelligence, the detailed description is omitted.
In summary, aiming at the technical problem that the accurate hole area of the processed product cannot be obtained due to the fact that the imaging of the hole area is affected by illumination factors, an image processing method for the processed product of the chamfering equipment is provided. Firstly, obtaining a surface gray level image of a product processed by chamfering equipment, and determining a hole area and hole edge pixel points; further analyzing the distribution characteristics of the hole edge pixel points, the gray level change characteristics of the pixel points in the preset neighborhood of each hole pixel point and the gray level characteristics of each hole pixel point, and obtaining the correction burr degree of each hole pixel point; further according to the statistical characteristics and the fluctuation characteristics of the gradient amplitude values of the hole edge pixel points, the difference of the hole areas and other hole areas is combined, and the corrected illumination intensity of each hole area is obtained; and further acquiring a self-adaptive neighborhood region of each hole pixel point, and processing the surface gray level image to acquire a hole region to be analyzed. According to the invention, the self-adaptive neighborhood region of each hole pixel point is obtained by analyzing the gray characteristics of each hole pixel point and surrounding pixel points, and the influence of illumination factors on the imaging of the hole region, so that an accurate hole region is obtained.
It should be noted that: the sequence of the 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 particular order shown, or sequential order, to achieve desirable results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.
In this specification, each embodiment is described in a progressive manner, and identical and similar parts of each embodiment are all referred to each other, and each embodiment mainly describes differences from other embodiments.
Claims (10)
1. The chamfering equipment processing quality detection method based on artificial intelligence is characterized by comprising the following steps:
acquiring a surface gray level image of a product processed by chamfering equipment; acquiring a hole area and hole edge pixel points of the surface gray level image;
acquiring the possibility of burr defects of each hole area according to the distribution characteristics of the hole edge pixel points of each hole area; according to the gray scale change characteristics of the pixel points in the preset neighborhood of each hole pixel point in each hole region, combining the gray scale characteristics of each hole pixel point to obtain the burr degree of each hole pixel point; combining the burr degree of each hole pixel point and the burr defect possibility of the hole region to obtain a corrected burr degree of each hole pixel point;
acquiring initial illumination intensity of each hole area according to statistical characteristics and fluctuation characteristics of gradient amplitude values of hole edge pixel points of each hole area; according to the difference characteristics of the initial illumination intensity of each hole area and other hole areas, the initial illumination intensity is corrected by combining the difference characteristics of the correction burr degrees of all hole edge pixel points in each hole area and other hole areas, and the corrected illumination intensity of each hole area is obtained;
according to the corrected burr degree of each hole pixel point and the corrected illumination intensity of the hole region, combining preset neighborhood size parameters to obtain a self-adaptive neighborhood region of each hole pixel point;
and processing the surface gray level image according to the self-adaptive neighborhood region of each hole pixel point, and finally realizing quality detection.
2. The chamfering equipment processing quality detecting method based on artificial intelligence according to claim 1, wherein the burr defect possibility obtaining method comprises:
and connecting adjacent hole edge pixel points, obtaining the angle values of the connecting lines of the adjacent hole edge pixel points in a plane rectangular coordinate system, summing the absolute values of the differences of the angle values of all the adjacent connecting lines, and taking the sum result as the possibility of burr defects.
3. The chamfering equipment processing quality detection method based on artificial intelligence according to claim 1, wherein the burr degree acquisition method comprises the steps of:
taking the ratio of the gray value of the hole pixel point to the average gray value of the surface gray image as the gray characteristic parameter of the hole pixel point;
traversing the pixel points in a preset adjacent area of the hole pixel points according to a set sequence, counting the number of times of changing the gray value of the adjacent pixel points, and taking the product of the number of times of changing and the corresponding gray characteristic parameter as the burr degree of the corresponding hole pixel points.
4. The chamfering equipment processing quality detection method based on artificial intelligence according to claim 1, wherein the initial illumination intensity acquisition method comprises:
taking the sum of gradient magnitudes of all hole edge pixel points in each hole area as a first illumination parameter;
summing the absolute values of the differences between the gradient magnitudes and the gradient magnitude mean values of all the hole edge pixel points in each hole area, and taking the mapping result as a second illumination parameter after the negative correlation of the summation result and the mapping by using an exponential function;
taking the product of the first illumination parameter and the second illumination parameter as an initial illumination intensity.
5. The chamfering equipment processing quality detection method based on artificial intelligence according to claim 1, wherein the modified illumination intensity acquisition method comprises:
obtaining corrected illumination intensity by using an initial illumination intensity correction formula; the initial illumination intensity correction formula comprises:
the method comprises the steps of carrying out a first treatment on the surface of the Wherein (1)>Indicate->The corrected illumination intensity of the individual hole areas; />Indicate->Initial illumination intensity of individual hole areas; />Representing a normalization function; />Indicate->In each hole area, the sum of the correction burr degrees of all hole edge pixel points; />Representation of the surface greyscale image except +.>The number of other hole areas outside the hole areas; />Indicating removal of->Outside the hole area->Initial illumination intensity of the other hole areas; />Indicating removal of->Outside the hole area, the firstAnd in other hole areas, the sum of the correction burr degrees of all hole edge pixel points.
6. The chamfering equipment processing quality detection method based on artificial intelligence according to claim 1, wherein the adaptive neighborhood region acquisition method comprises:
obtaining the self-adaptive neighborhood size parameter of each hole pixel point by utilizing a self-adaptive neighborhood size parameter calculation formula; establishing a square neighborhood region by taking each hole pixel point as a center and taking the corresponding self-adaptive neighborhood size parameter as a side length to obtain the self-adaptive neighborhood region of each hole pixel point; the adaptive neighborhood size parameter calculation formula comprises:
the method comprises the steps of carrying out a first treatment on the surface of the Wherein (1)>Indicate->First part of the hole area>Self-adaptive neighborhood size parameters of the pixel points of the holes; />Representing an upward rounding function; />Representing a preset neighborhood size parameter;representing a normalization function; />First->Within the hole region->Correcting burr degree of each hole pixel point;indicate->The corrected illumination intensity of the individual hole areas; />Representing a preset scaling factor, and +.>。
7. The chamfering equipment processing quality detection method based on artificial intelligence according to claim 1, wherein the method for processing the surface gray-scale image and finally realizing quality detection comprises the following steps:
and according to the self-adaptive neighborhood region of each hole pixel point, performing image segmentation on a surface gray level image of a chamfering equipment processing product by utilizing an MRF algorithm to obtain a hole region to be analyzed, calculating the correction burr degree of each pixel point in the hole region to be analyzed, normalizing to obtain an abnormal score of each pixel point in the hole region to be analyzed, and when the abnormal score of each pixel point is larger than a preset burr threshold value, determining the hole region to be analyzed as a burr region, otherwise, determining the hole region to be analyzed as a qualified region.
8. The chamfering equipment processing quality detection method based on artificial intelligence according to claim 1, wherein the method for obtaining the corrected burr degree comprises the steps of:
and taking the product of the burr degree of each hole pixel point and the burr defect possibility of the hole area as the corrected burr degree of each hole pixel point.
9. The chamfering equipment processing quality detection method based on artificial intelligence according to claim 1, wherein the hole edge pixel point acquisition method comprises the following steps:
performing edge detection by using a Canny operator to obtain all edge pixel points of the surface gray level image; and detecting holes in the image by using a Hough circle detection algorithm, and taking edge pixel points at the edges of the holes as hole edge pixel points.
10. The chamfering equipment processing quality detection method based on artificial intelligence according to claim 9, wherein the hole area acquisition method comprises:
selecting a first preset number of pixel points on an extension line of the gradient direction of each hole edge pixel point contained in the hole, selecting a second preset number of pixel points on an extension line of the opposite direction of the gradient direction, and taking the hole edge pixel point of each hole and the pixel points selected by the hole edge pixel point as a hole area.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410223783.3A CN117808799B (en) | 2024-02-29 | 2024-02-29 | Chamfering equipment processing quality detection method based on artificial intelligence |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410223783.3A CN117808799B (en) | 2024-02-29 | 2024-02-29 | Chamfering equipment processing quality detection method based on artificial intelligence |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN117808799A true CN117808799A (en) | 2024-04-02 |
| CN117808799B CN117808799B (en) | 2024-05-07 |
Family
ID=90423853
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202410223783.3A Active CN117808799B (en) | 2024-02-29 | 2024-02-29 | Chamfering equipment processing quality detection method based on artificial intelligence |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117808799B (en) |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114943848A (en) * | 2022-07-25 | 2022-08-26 | 南通德晋昌光电科技有限公司 | Crack identification method in nickel screen laser cladding process |
| CN117078672A (en) * | 2023-10-13 | 2023-11-17 | 深圳市凯尔文电子有限公司 | Intelligent detection method for mobile phone screen defects based on computer vision |
| CN117437191A (en) * | 2023-10-24 | 2024-01-23 | 辽宁工程技术大学 | Rail surface detection method and system integrating inspection image and magnetic flux leakage detection |
| CN117541588A (en) * | 2024-01-10 | 2024-02-09 | 大连建峰印业有限公司 | Printing defect detection method for paper product |
-
2024
- 2024-02-29 CN CN202410223783.3A patent/CN117808799B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114943848A (en) * | 2022-07-25 | 2022-08-26 | 南通德晋昌光电科技有限公司 | Crack identification method in nickel screen laser cladding process |
| CN117078672A (en) * | 2023-10-13 | 2023-11-17 | 深圳市凯尔文电子有限公司 | Intelligent detection method for mobile phone screen defects based on computer vision |
| CN117437191A (en) * | 2023-10-24 | 2024-01-23 | 辽宁工程技术大学 | Rail surface detection method and system integrating inspection image and magnetic flux leakage detection |
| CN117541588A (en) * | 2024-01-10 | 2024-02-09 | 大连建峰印业有限公司 | Printing defect detection method for paper product |
Non-Patent Citations (2)
| Title |
|---|
| 潘密，等: "灰铸铁件加工面孔洞缺陷的原因分析及对策", 铸造工艺技术, vol. 72, no. 10, 31 December 2023 (2023-12-31) * |
| 牛晗，等: "基于OTSU与CANNY算法的竹片缺陷图像检测", 森林工程, vol. 38, no. 6, 30 November 2022 (2022-11-30) * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN117808799B (en) | 2024-05-07 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108921176B (en) | Pointer instrument positioning and identifying method based on machine vision | |
| WO2022110804A1 (en) | Image noise measurement method and device based on local statistical information | |
| CN116168026B (en) | Water quality detection method and system based on computer vision | |
| CN115020267B (en) | Semiconductor surface defect detection method | |
| CN115222733A (en) | Metal component surface defect detection method based on image recognition | |
| CN107490582B (en) | Assembly line workpiece detection system | |
| CN115018844B (en) | Plastic film quality evaluation method based on artificial intelligence | |
| CN116843678A (en) | Hard carbon electrode production quality detection method | |
| CN115984271A (en) | Metal burr identification method based on angular point detection | |
| CN116883408B (en) | Integrating instrument shell defect detection method based on artificial intelligence | |
| CN114881915A (en) | Symmetry-based mobile phone glass cover plate window area defect detection method | |
| CN116934763B (en) | Medical rubber plug defect detection method based on visual characteristics | |
| CN112884746A (en) | Character defect intelligent detection algorithm based on edge shape matching | |
| CN114581805A (en) | Coating roller surface defect detection method adopting 3D line laser profile technology | |
| CN109544513A (en) | A kind of steel pipe end surface defect extraction knowledge method for distinguishing | |
| CN116128873A (en) | Bearing retainer detection method, device and medium based on image recognition | |
| CN117237747B (en) | Hardware defect classification and identification method based on artificial intelligence | |
| CN116993742B (en) | Nickel alloy rolling defect detection method based on machine vision | |
| CN117011291B (en) | Watch shell quality visual detection method | |
| CN111815575B (en) | Bearing steel ball part detection method based on machine vision | |
| CN116612112B (en) | Visual inspection method for surface defects of bucket | |
| CN111539951B (en) | Visual detection method for outline size of ceramic grinding wheel head | |
| CN117808799B (en) | Chamfering equipment processing quality detection method based on artificial intelligence | |
| CN115880280A (en) | Detection method for quality of steel structure weld joint | |
| CN108171691B (en) | Detection method of plastic container |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |