CN117011297A - Aluminum alloy automobile accessory die defect detection method based on image processing - Google Patents
Aluminum alloy automobile accessory die defect detection method based on image processing Download PDFInfo
- Publication number
- CN117011297A CN117011297A CN202311279017.0A CN202311279017A CN117011297A CN 117011297 A CN117011297 A CN 117011297A CN 202311279017 A CN202311279017 A CN 202311279017A CN 117011297 A CN117011297 A CN 117011297A
- Authority
- CN
- China
- Prior art keywords
- gray
- growth
- value
- coefficient
- neighborhood
- 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
- 230000007547 defect Effects 0.000 title claims abstract description 59
- 229910000838 Al alloy Inorganic materials 0.000 title claims abstract description 48
- 238000012545 processing Methods 0.000 title claims abstract description 25
- 238000001514 detection method Methods 0.000 title claims abstract description 22
- 238000009826 distribution Methods 0.000 claims abstract description 39
- 238000000034 method Methods 0.000 claims abstract description 33
- 238000012937 correction Methods 0.000 claims abstract description 27
- 238000012163 sequencing technique Methods 0.000 claims description 12
- 238000012216 screening Methods 0.000 claims description 6
- 230000001502 supplementing effect Effects 0.000 claims description 4
- 230000002950 deficient Effects 0.000 claims description 3
- 230000000694 effects Effects 0.000 abstract description 10
- 238000010191 image analysis Methods 0.000 abstract description 5
- 238000004458 analytical method Methods 0.000 description 16
- 238000004364 calculation method Methods 0.000 description 5
- 238000010606 normalization Methods 0.000 description 5
- 230000000750 progressive effect Effects 0.000 description 2
- 241001270131 Agaricus moelleri Species 0.000 description 1
- 230000002159 abnormal effect Effects 0.000 description 1
- 230000005856 abnormality Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000005266 casting Methods 0.000 description 1
- 238000003708 edge detection Methods 0.000 description 1
- 238000001914 filtration Methods 0.000 description 1
- 239000003102 growth factor Substances 0.000 description 1
- 238000005286 illumination Methods 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 238000005259 measurement Methods 0.000 description 1
- 230000035772 mutation Effects 0.000 description 1
- 239000013307 optical fiber Substances 0.000 description 1
- 239000004576 sand Substances 0.000 description 1
- 239000002893 slag Substances 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/0002—Inspection of images, e.g. flaw detection
- G06T7/0004—Industrial image inspection
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/60—Analysis of geometric attributes
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/10—Image acquisition modality
- G06T2207/10004—Still image; Photographic image
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/30—Subject of image; Context of image processing
- G06T2207/30108—Industrial image inspection
- G06T2207/30164—Workpiece; Machine component
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02P—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN THE PRODUCTION OR PROCESSING OF GOODS
- Y02P90/00—Enabling technologies with a potential contribution to greenhouse gas [GHG] emissions mitigation
- Y02P90/30—Computing systems specially adapted for manufacturing
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Quality & Reliability (AREA)
- Geometry (AREA)
- Image Analysis (AREA)
Abstract
The invention relates to the technical field of image analysis, in particular to an aluminum alloy steam fitting die defect detection method based on image processing. The method comprises the steps of obtaining an initial gray image of an aluminum alloy fitting, and determining seed points of region growth; optionally selecting a target pixel point, determining a neighborhood direction, and determining the initial influence degree of the target pixel point according to the gray value change and the gradient included angle change; determining a correction coefficient according to the gray level distribution and the gradient direction distribution of the pixel points, and determining a target influence degree according to the correction coefficient and the initial influence degree; obtaining a growth coefficient according to the gray value of the seed point, the gray value of the target pixel point and the target influence degree, and carrying out regional growth treatment on the seed point according to the growth coefficient to obtain a growth region; and determining a defect area according to the gray value of the pixel point in the growth area. The invention can effectively analyze and identify the defect area, improve the detection precision of defect detection and enhance the detection effect.
Description
Technical Field
The invention relates to the technical field of image analysis, in particular to an aluminum alloy steam fitting die defect detection method based on image processing.
Background
The aluminum alloy steam fitting is widely applied to steam repair service, and the defect detection of the aluminum alloy steam fitting mold is generally carried out according to the defect detection of the produced aluminum alloy steam fitting, so that the aluminum alloy is easy to have casting defects such as shrinkage cavity, sand holes, air holes, slag inclusion and the like in the production process, and the requirement on the detection precision of the aluminum alloy steam fitting is higher.
In the related art, the surface defects of the aluminum alloy fitting are detected by using the modes of edge detection and region growth, in the mode, because the aluminum alloy fitting can reflect optical fibers during shooting, namely, because the incident conditions of light rays are different, the difference of corresponding gray scales is generated in the same normal region, so that the gray scale distribution in an image is uneven, and the region growth algorithm carries out region growth processing based on the gray scale change, so that the reliability of a defect region obtained through analysis is lower, the identification effect is poor, the defect region cannot be accurately identified, and the detection precision of defect detection is insufficient, and the detection effect is poor.
Disclosure of Invention
In order to solve the technical problems that a defect area cannot be accurately identified in the related art, the detection precision of defect detection is insufficient and the detection effect is poor, the invention provides an aluminum alloy automobile accessory die defect detection method based on image processing, and the adopted technical scheme is as follows:
the invention provides an image processing-based aluminum alloy automobile accessory die defect detection method, which comprises the following steps:
acquiring an initial gray image of an aluminum alloy fitting, and determining seed points for region growth according to gray value distribution of pixel points in the initial gray image;
taking any pixel point in the neighborhood of the seed point as a target pixel point, determining a neighborhood direction according to the direction of other pixel points in the neighborhood of the target pixel point from the seed point, and determining the initial influence degree of the target pixel point according to the gray value change and gradient included angle change of the pixel points in different neighborhood directions;
according to the gray level distribution and gradient direction distribution of the pixel points in different neighborhood directions, determining a correction coefficient, and correcting the initial influence degree according to the correction coefficient to obtain a target influence degree;
obtaining a growth coefficient according to the gray value of the seed point, the gray value of the target pixel point and the target influence degree, and carrying out regional growth treatment on the seed point according to the growth coefficient to obtain a growth region; and determining a defect area according to the gray value of the pixel point in the growth area.
Further, the determining the seed points for the region growth according to the gray value distribution of the pixel points in the initial gray image includes:
constructing a gray distribution histogram of pixel points in the initial gray image, determining the number of pixel points in different gray intervals, and determining the duty ratio of the pixel points in different gray intervals according to the number of the pixel points in different gray intervals;
distributing a preset first number of seed points into corresponding gray intervals according to the duty ratio of the pixel points in different gray intervals, determining the number of the seed points in each gray interval, and randomly screening the seed points according to the number of the corresponding seed points in any gray interval.
Further, the determining the neighborhood direction according to the direction of other pixels in the neighborhood of the target pixel from the seed point includes:
taking the pixel points except the seed point in the eight adjacent points of the target pixel point as other pixel points;
and determining the straight line direction between the seed point and any other pixel point as a neighborhood direction by taking the seed point as a starting point.
Further, determining the initial influence degree of the target pixel point according to the gray value change and the gradient included angle change of the pixel point in different neighborhood directions includes:
in any neighborhood direction, taking a preset second number of pixel points which are closest to the seed point as reference pixel points; sequencing the reference pixel points according to the sequence from near to far from the seed points to obtain a reference sequence;
calculating the absolute value of the difference value of the gray values of two adjacent reference pixel points in the reference sequence as the adjacent gray difference value; when the adjacent gray level difference value is larger than a preset gray level difference value threshold value, adjusting the adjacent gray level difference value to be a first constant coefficient; when the adjacent gray level difference value is smaller than or equal to a preset gray level difference value threshold value, the adjacent gray level difference value is adjusted to be a second constant coefficient, wherein the first constant coefficient is smaller than the second constant coefficient; accumulating all the adjusted adjacent gray level difference values to obtain a gray level change coefficient;
calculating the difference value of gradient included angles of two adjacent reference pixel points in the reference sequence as an adjacent angle difference value; when the adjacent angle difference value is larger than a preset angle difference value threshold value, adjusting the adjacent angle difference value to be a third constant coefficient; when the adjacent angle difference value is smaller than or equal to a preset angle difference value threshold value, adjusting the adjacent angle difference value to be a fourth constant coefficient, wherein the third constant coefficient is smaller than the fourth constant coefficient; accumulating all the adjusted adjacent angle differences to obtain gradient included angle change coefficients;
calculating a normalized value of the product of the gray value change coefficient and the gradient included angle change coefficient as a change index of the neighborhood direction;
and taking the average value of the change indexes of all the neighborhood directions as the initial influence degree of the target pixel point.
Further, the determining the correction coefficient according to the gray scale distribution and the gradient direction distribution of the pixel points in different neighborhood directions includes:
calculating average normalized values of adjacent gray level difference values of all reference pixel points in each neighborhood direction as gray level average coefficients of the corresponding neighborhood directions; sequencing the neighborhood direction based on a preset sequencing order, and performing curve fitting on the gray average coefficient according to the sequencing order to obtain a neighborhood gray scale curve;
calculating average normalized values of adjacent angle differences of all reference pixel points in each neighborhood direction as angle average coefficients of corresponding neighborhood directions; performing curve fitting on the angle mean coefficients according to the ordering sequence to obtain a neighborhood angle curve;
and calculating the similarity of the neighborhood gray level curve and the neighborhood angle curve based on a structural similarity algorithm, and taking the similarity as a correction coefficient.
Further, the correcting the initial influence degree according to the correction coefficient to obtain a target influence degree includes:
and calculating the product of the correction coefficient and the initial influence degree as a target influence degree.
Further, the obtaining a growth coefficient according to the gray value of the seed point, the gray value of the target pixel point and the target influence degree includes:
calculating the absolute value of the difference between the gray value of the seed point and the gray value of the target pixel point as gray difference;
and obtaining a growth coefficient according to the gray level difference and the target influence degree, wherein the gray level difference and the growth coefficient are in an inverse correlation relationship, the target influence degree and the growth coefficient are in a positive correlation relationship, and the value of the growth coefficient is a normalized value.
Further, the performing a region growing process on the seed points according to the growth coefficients to obtain a growing region, including:
when the growth coefficient meets a preset growth condition, supplementing the target pixel point to an area where the seed point is located, and taking the target pixel point as a new seed point to perform area growth treatment;
and stopping the growth of the region when the growth coefficient does not meet the preset growth condition, and taking the obtained region as a growth region after stopping the growth in all directions.
Further, the preset growth condition is that the growth coefficient is greater than a preset coefficient threshold.
Further, the determining the defect area according to the gray value of the pixel point in the growth area includes:
calculating the average value of the gray values of all pixel points in any growth area as an area gray average value;
clustering the area gray average value of all the growing areas, and screening defective areas from the growing areas according to the clustering result.
The invention has the following beneficial effects:
the method is applied to the technical field of image analysis, an initial seed point is determined according to the distribution of the gray values of the pixel points in the initial gray image, then the initial influence degree of the target pixel point is determined by combining the change of the gray values of the pixel points around the seed point and the change of the gradient direction, and as the gradient change of the surface of the aluminum alloy vapor fitting is consistent, namely when the gray values and the gradient direction generate abrupt changes, the corresponding defects can be represented, therefore, the initial influence degree can be used for carrying out initial analysis on the defect condition, the correction coefficient can be obtained according to the similarity of the gray distribution of the pixel points and the gradient direction distribution, the target influence degree can be obtained by combining the initial influence degree and the correction coefficient, the target influence degree can be quickly and accurately analyzed on the gray distribution of the pixel points around the target pixel point, the gray values of the target pixel point and the target influence degree, and the growth coefficient can be obtained according to the gray values of the target pixel points, compared with the method of carrying out regional growth based on the difference of the gray values in the traditional regional growth algorithm, the defect region can be obtained, the defect region growth region can be reliably detected, the defect region can be effectively detected and the defect region can be accurately detected, the defect region can be accurately is improved, the defect region growth region can be accurately is detected and the defect region can be accurately is detected, and the defect region can be detected.
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 flowchart of a method for detecting defects of an aluminum alloy steam fitting mold based on image processing according to an embodiment of the invention.
Detailed Description
In order to further describe the technical means and effects adopted by the invention to achieve the preset aim, the following is a detailed description of specific implementation, structure, characteristics and effects of the method for detecting the defects of the aluminum alloy steam fitting die based on image processing according to the invention, which is provided by the invention, with reference to 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.
The invention provides a specific scheme of an aluminum alloy steam fitting die defect detection method based on image processing, which is specifically described below with reference to the accompanying drawings.
Referring to fig. 1, a flowchart of a method for detecting defects of an aluminum alloy auto parts mold based on image processing according to an embodiment of the present invention is shown, where the method includes:
s101: acquiring an initial gray image of the aluminum alloy fitting, and determining seed points for region growth according to gray value distribution of pixel points in the initial gray image.
In the embodiment of the invention, a high-resolution camera can be arranged right above the aluminum alloy fitting, then the high-resolution camera is used for collecting the original image of the aluminum alloy fitting, and the image denoising and the image graying treatment are carried out on the original image to obtain an initial gray image.
It is understood that the image graying may be, for example, mean graying, and the image denoising may be, for example, mean filtering denoising, which are all well known to those skilled in the art, and are not further limited and described herein.
The invention analyzes the defect area by analyzing the gray information and gradient information of the pixel points in the initial gray image, and particularly refers to the following embodiment.
Optionally, in some embodiments of the present invention, determining the seed point for the region growing according to the gray value distribution of the pixel points in the initial gray image includes: constructing a gray distribution histogram of pixel points in an initial gray image, determining the number of pixel points in different gray intervals, and determining the duty ratio of the pixel points in different gray intervals according to the number of the pixel points in different gray intervals; distributing a preset first number of seed points into corresponding gray intervals according to the duty ratio of the pixel points in different gray intervals, determining the number of the seed points in each gray interval, and randomly screening the seed points according to the number of the corresponding seed points in any gray interval.
In the embodiment of the present invention, gray values from 0 to 255 can be uniformly divided into 16 ranges, namely [0,16 ], [16,32 ], …, and [240,255], and then, the number of pixels corresponding to the gray values in the initial gray image is counted and processed in a histogram to obtain a gray distribution histogram of the pixels, and of course, the number of the divided ranges can be adjusted according to the actual situation.
The preset first number may be determined empirically, and optionally, the preset first number may be, for example, 16, which is not limited in this embodiment.
According to the embodiment of the invention, the seed point quantity is distributed according to the ratio of the quantity of the pixel points in different gray value ranges to the quantity of all the pixel points, so that the seed point quantity in each gray value range is obtained, the seed point quantity obtained in each gray value range is ensured to be matched with the gray distribution of the pixel points in the initial gray image, and the initial gray image is analyzed more comprehensively.
In the embodiment of the invention, in any gray scale interval, seed points are randomly selected according to the corresponding seed point number, that is, after the number of the seed points distributed in any gray scale value range is determined, the seed points can be randomly selected from the corresponding pixel points, so as to obtain the seed points. After the seed points are determined, the embodiment of the invention can perform region growth treatment according to the seed points, but only perform region growth according to the gray values, which finally leads to lower defect detection precision, so that the invention introduces gradient change of the neighborhood pixel points and gray distribution of the neighborhood pixel points, thereby performing integral analysis on the periphery of the seed points and ensuring the reliability of region growth.
S102: and taking any pixel point in the neighborhood of the seed point as a target pixel point, determining the neighborhood direction according to the direction of other pixel points in the neighborhood of the target pixel point from the seed point, and determining the initial influence degree of the target pixel point according to the gray value change and gradient included angle change of the pixel points in different neighborhood directions.
In the embodiment of the present invention, the neighborhood may specifically be, for example, an eight neighborhood, that is, one pixel point is optionally selected as a target pixel point in the eight neighborhood of the seed point, and it is to be understood that the region growth of the present invention uses the seed point as a starting point, and performs region growth on surrounding pixel points, that is, all pixel points in the eight neighborhood need to be respectively subjected to region growth analysis, and for convenience of explanation, a certain pixel point is optionally selected as a target pixel point, and other pixel points in the eight neighborhood need to perform corresponding region growth analysis processes, which are not described herein.
Optionally, in some embodiments of the present invention, determining the neighborhood direction according to the direction of other pixels in the neighborhood of the target pixel from the seed point includes: taking the pixel points except the seed point in the eight adjacent points of the target pixel point as other pixel points; and determining the straight line direction between the seed point and any other pixel point as a neighborhood direction by taking the seed point as a starting point.
In the embodiment of the invention, the pixel points except the seed point in the eight adjacent areas of the target pixel point can be used as other pixel points, then the seed point is used as a starting point, the seed point is connected with any other pixel point, and the direction corresponding to the connection, namely the direction from the seed point to the other pixel points, is used as the neighborhood direction.
It can be understood that in some embodiments of the present invention, since the target pixel point is a pixel point in the eight neighboring areas of the seed point, the azimuth between the target pixel point and the seed point may be divided into a positive azimuth and an oblique azimuth, where the positive azimuth is that the straight line connecting the seed point and the target pixel point is parallel to the coordinate axis, and the oblique direction indicates that the straight line connecting the seed point and the target pixel point is not parallel to the coordinate axis, and the number of neighboring directions corresponding to different situations is different, but the calculation processes are the same, so that the analysis of the pixel point distribution is uniformly performed.
Optionally, in some embodiments of the present invention, determining the initial influence degree of the target pixel according to the gray value change and the gradient angle change of the pixel in different neighborhood directions includes: in any neighborhood direction, taking a preset second number of pixel points which are closest to the seed point as reference pixel points; sequencing the reference pixel points according to the sequence from near to far from the seed points to obtain a reference sequence; calculating the absolute value of the difference value of the gray values of two adjacent reference pixel points in the reference sequence as the adjacent gray difference value; when the adjacent gray difference value is larger than a preset gray difference value threshold value, adjusting the adjacent gray difference value to be a first constant coefficient; when the adjacent gray level difference value is smaller than or equal to a preset gray level difference value threshold value, adjusting the adjacent gray level difference value to a second constant coefficient, wherein the first constant coefficient is smaller than the second constant coefficient; accumulating all the adjusted adjacent gray level difference values to obtain a gray level change coefficient; calculating the difference value of gradient included angles of two adjacent reference pixel points in the reference sequence as an adjacent angle difference value; when the adjacent angle difference value is larger than a preset angle difference value threshold value, adjusting the adjacent angle difference value to be a third constant coefficient; when the adjacent angle difference value is smaller than or equal to a preset angle difference value threshold value, adjusting the adjacent angle difference value to be a fourth constant coefficient, wherein the third constant coefficient is smaller than the fourth constant coefficient; accumulating all the adjusted adjacent angle differences to obtain gradient included angle change coefficients; calculating a normalized value of the product of the gray value change coefficient and the gradient included angle change coefficient as a change index of the neighborhood direction; and taking the average value of the change indexes of all the neighborhood directions as the initial influence degree of the target pixel point.
The preset second number is the number of reference pixel points to be analyzed in any one of the neighborhood directions, and optionally, the preset second number may be, for example, specifically 5, that is, 5 pixel points, which are closest to the seed point in any one of the neighborhood directions, are used as reference pixel points in the corresponding neighborhood directions.
It can be understood that, because the capability of reflecting the light source on the surface of the aluminum alloy fitting is stronger, the light source irradiates the aluminum alloy surface to easily generate texture features which attenuate smoothly along with the distance, and the closer the light source points are, the larger the corresponding illumination intensity is, the larger the gray value of the pixel point is, but because the defect area, namely the area such as the bulge, the crease, the dent and the like can influence the light receiving effect of the aluminum alloy surface, the corresponding light is bent, the aluminum alloy surface, namely the partial area, can suddenly lighten and suddenly darken, or the situation that the normal light intensity attenuation rule is not followed is generated, and the analysis is performed based on the defect area.
In the embodiment of the invention, the reference pixel points are sequenced from near to far from the seed points to obtain the reference sequence, and as the distance from the light source is longer, the gray value between the corresponding adjacent pixel points is smaller, the absolute value of the difference value of the gray values of the two adjacent reference pixel points in the reference sequence is calculated as the adjacent gray difference value.
The preset gray level difference threshold is a threshold value of an adjacent gray level difference, the preset angle difference threshold is a threshold value of an adjacent angle difference, the value of the preset gray level difference threshold can be adjusted according to actual requirements, optionally, the preset gray level difference threshold is 50, and the preset angle difference threshold is 45 degrees, which is not limited.
The first constant coefficient, the second constant coefficient, the third constant coefficient and the fourth constant coefficient are constant values used for facilitating statistical calculation, and the gray value and the gradient angle of the pixel point are replaced by corresponding constant values, so that the calculation process is simplified while the dimension influence is eliminated, and the processing efficiency is improved. Alternatively, the first constant coefficient and the third constant coefficient are 0, and the second constant coefficient and the fourth constant coefficient are 1, or may be adjusted according to actual requirements, which is not limited.
It can be understood that, since the gray value gradually becomes smaller along with the distance between the gray value and the light source on the surface of the normal aluminum alloy fitting, the process is a progressive process, and when the influence of the bulge, the recess or other defects occurs, the abrupt change of the gray value can be generated, so that the constant corresponding to the pixel point with the abrupt change of the gray value is adjusted to be a smaller value, the constant corresponding to the normal pixel point is adjusted to be a larger value, and the subsequent analysis of the pixel point is facilitated. The same applies to the gradient angle treatment.
In the related art, the processing is only carried out according to the gray value, the region growth cannot be effectively carried out when the corresponding noise is generated in the mode, and the reliability of the final region growth processing can be higher by combining the gray value and the gradient angle joint analysis due to the smooth surface of the aluminum alloy fitting.
In the embodiment of the invention, all the adjusted adjacent angle differences are accumulated to obtain gradient included angle change coefficients; calculating a normalized value of the product of the gray value change coefficient and the gradient included angle change coefficient as a change index of the neighborhood direction; and taking the average value of the change indexes of all the neighborhood directions as the initial influence degree of the target pixel point.
The larger the change coefficient of the gray value is, the smaller the gray abrupt change condition of the corresponding pixel point is, the larger the change coefficient of the gradient included angle is, the smaller the gradient direction change of the corresponding pixel point is, namely, when the change coefficient of the gray value and the change coefficient of the gradient included angle are both larger, the corresponding pixel point is more in accordance with the texture condition of the surface of a normal aluminum alloy automobile accessory, the more normal the texture is in the corresponding neighborhood direction.
S103: and determining a correction coefficient according to the gray level distribution and the gradient direction distribution of the pixel points in different neighborhood directions, and correcting the initial influence degree according to the correction coefficient to obtain the target influence degree.
Optionally, in some embodiments of the present invention, determining the correction coefficient according to the gray scale distribution and the gradient direction distribution of the pixel points in different neighborhood directions includes: calculating average normalized values of adjacent gray level difference values of all reference pixel points in each neighborhood direction as gray level average coefficients of the corresponding neighborhood directions; sequencing the neighborhood direction based on a preset sequencing order, and performing curve fitting on the gray average coefficient according to the sequencing order to obtain a neighborhood gray scale curve; calculating average normalized values of adjacent angle differences of all reference pixel points in each neighborhood direction as angle average coefficients of corresponding neighborhood directions; performing curve fitting on the angle mean coefficients according to the ordering sequence to obtain a neighborhood angle curve; and calculating the similarity of the neighborhood gray level curve and the neighborhood angle curve based on the structural similarity algorithm, and taking the similarity as a correction coefficient.
In the embodiment of the invention, since the gray value change and the gradient angle change both accord with a certain change rule when analysis is carried out, namely, in a normal aluminum alloy surface, the change rule of gray of adjacent pixel points and the change rule of gradient angles are similar, and when abnormality occurs, the change of gray of one party possibly occurs, for example, when crease occurs on the aluminum alloy surface, the influence of the gray change is small, but the gradient direction can change greatly, the embodiment of the invention carries out finer analysis on the image texture by calculating the similarity of the gray change and the gradient angle change.
In the embodiment of the invention, curve fitting is performed on the gray average coefficient to obtain a neighborhood gray scale curve, curve fitting is performed on the angle average coefficient to obtain a neighborhood angle curve, and under normal conditions, the neighborhood gray scale curve and the neighborhood angle curve have the same fluctuation effect, namely have larger similarity, and under abnormal conditions, the similarity is smaller, the embodiment of the invention calculates the similarity based on a structural similarity algorithm, wherein the structural similarity algorithm is an existing algorithm well known to a person skilled in the art, and the structural similarity algorithm is not further described.
Optionally, in some embodiments of the present invention, correcting the initial influence level according to the correction coefficient to obtain the target influence level includes: and calculating the product of the correction coefficient and the initial influence degree as a target influence degree.
In the embodiment of the invention, the correction coefficient can be directly used as the weight coefficient of the initial influence degree, and the product of the correction coefficient and the initial influence degree is calculated to be used as the target influence degree. The larger the correction coefficient is, the higher the similarity between the corresponding neighborhood gray scale curve and the neighborhood angle curve is, that is, the higher the possibility that the target pixel point is the pixel point of the normal aluminum alloy automobile accessory surface is.
Thus, embodiments of the present invention analyze region growth by targeted impact levels, see in particular the examples that follow.
S104: obtaining a growth coefficient according to the gray value of the seed point, the gray value of the target pixel point and the target influence degree, and carrying out regional growth treatment on the seed point according to the growth coefficient to obtain a growth region; and determining a defect area according to the gray value of the pixel point in the growth area.
Optionally, in some embodiments of the present invention, obtaining the growth coefficient according to the gray value of the seed point, the gray value of the target pixel point, and the target influence degree includes: calculating the absolute value of the difference between the gray value of the seed point and the gray value of the target pixel point as the gray difference; and obtaining a growth coefficient according to the gray level difference and the target influence degree, wherein the gray level difference and the growth coefficient are in an inverse relation, the target influence degree and the growth coefficient are in a positive relation, and the value of the growth coefficient is a normalized value.
When the target influence degree is obtained, analysis is only carried out according to the change of the gray value and the gradient angle, but a certain highlight area and a certain backlight area are formed on the surface of a real aluminum alloy automobile accessory, wherein the gray value in the highlight area is higher, the change of the gray value and the gradient angle is smaller, the gray value in the backlight area is lower, and the change of the gray value and the gradient angle is smaller, so that the effective analysis can not be carried out according to the change only in the case, the invention combines the gray values of a seed point and a target pixel point, and the calculation formula of the growth coefficient can be specifically as follows:
；
in the method, in the process of the invention,representing the growth factor of the target pixel, +.>Gray value representing seed point,/->Gray value representing target pixel, +.>Representing absolute value>Representing gray scale difference +.>Indicating the target influence degree of the target pixel, e indicating the natural constant, < >>In one embodiment of the present invention, the normalization process may be specifically, for example, maximum and minimum normalization processes, and the normalization in the subsequent steps may be performed by using the maximum and minimum normalization processes, and in other embodiments of the present invention, other normalization methods may be selected according to a specific range of values, which will not be described herein.
It can be understood that when the gray level difference is larger, the situation that gray level mutation is more likely to occur is indicated, and the larger the target influence degree is, the more normal the gray level change of the pixel points around the target pixel point is indicated, so that the corresponding growth coefficient is calculated by combining the gray level difference and the target influence degree, the growth coefficient not only can accurately represent the gray level value characteristic at the position of the target pixel point, but also can be combined with the gray level change of the surrounding pixel points, the reliability of the growth coefficient is improved, and the growth effect of the region is enhanced.
Optionally, in some embodiments of the present invention, performing a region growing process on the seed points according to the growth coefficients to obtain a growing region, including: when the growth coefficient meets the preset growth condition, supplementing the target pixel point to the area where the seed point is located, and taking the target pixel point as a new seed point to perform area growth treatment; and stopping the growth of the region when the growth coefficient does not meet the preset growth condition, and taking the obtained region as a growth region after stopping the growth in all directions.
Wherein the preset growth condition is that the growth coefficient is larger than a preset coefficient threshold. The preset coefficient threshold may be, for example, specifically 0.75, or may be adjusted according to the actual situation, which is not limited in the embodiment of the present invention.
In the embodiment of the invention, when the growth coefficient is greater than 0.75, supplementing the target pixel point to the area where the seed point is located, and taking the target pixel point as a new seed point to perform area growth treatment; when the growth coefficient is 0.75 or less, the region growth is stopped, and after stopping the growth in all directions, the obtained region is taken as a growth region.
It can be understood that the region growing is to grow towards surrounding pixel points, the process of the region growing is to put all the pixel points in the eight neighborhoods of the current seed point into stacks, and then sequentially calculate the similarity between the seed point and the pixel points in each eight neighborhoods so as to select whether to add the pixel point into the region, in the process, if the pixel points in the growth of other regions exist in the eight neighborhoods of the current seed point, the pixel points in the growth of other regions can be directly put out of stacks, thereby avoiding repeated calculation. The growth coefficient is obtained by carrying out image analysis on the gray value, gradient direction and position relation of adjacent pixel points around the target pixel point in a multi-scale manner, and the growth coefficient is used for replacing similarity measurement in the traditional growth strategy, so that the regional growth can adapt to the texture characteristics of the aluminum alloy surface, the accuracy of the regional growth is further improved, and the completeness and accuracy of the growth region are improved.
Optionally, in some embodiments of the present invention, determining the defect region according to the gray value of the pixel point in the growth region includes: calculating the average value of the gray values of all pixel points in any growth area as an area gray average value; clustering the area gray average value of all the growing areas, and screening the defect areas from the growing areas according to the clustering result.
In the embodiment of the invention, the gray value average value of all the pixel points in each growth area can be used as the area gray value average value, and because the seed points are selected based on different gray values, the area gray value average values of all the growth areas are clustered, when the clustering result is not matched with the corresponding seed points, the unmatched growth areas can be accurately screened out to be used as defect areas, or in other embodiments of the invention, the growth areas with excessively deviated clusters (i.e. the growth areas with excessively large or excessively small overall gray values) can be used as defect areas, so that the method is not limited.
The invention is applied to the technical field of image analysis, an initial seed point is determined by acquiring an initial gray image of an aluminum alloy fitting and according to the distribution of pixel point gray values in the initial gray image, then, the initial influence degree of a target pixel point is determined by combining the change of the pixel point gray values around the seed point and the change of gradient directions, as the surface gradient change of the aluminum alloy fitting is consistent, namely, when the gray values and the gradient directions are suddenly changed, the corresponding defect can be represented, therefore, the initial influence degree can carry out initial analysis on the defect condition, a correction coefficient is obtained according to the similarity of the gray distribution of the pixel point and the gradient direction distribution, and then, the target influence degree is obtained by combining the initial influence degree and the correction coefficient, so that the target influence degree can rapidly and accurately analyze the gray distribution and the gradient distribution of the pixel points around the target pixel point, compared with the mode of carrying out region growth based on the difference of gray values in the traditional region growth algorithm, the method and the device combine the gray distribution and gradient distribution of surrounding pixel points to obtain the growth coefficients, judge the region growth according to the growth coefficients, improve the growth strategy of the region growth, effectively improve the stability and analysis effect of the region growth, enhance the reliability of the obtained growth region, and further improve the accuracy and reliability of analysis of the defect region when determining the defect region according to the gray values of the pixel points in the growth region.
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. An aluminum alloy automobile accessory die defect detection method based on image processing is characterized by comprising the following steps:
acquiring an initial gray image of an aluminum alloy fitting, and determining seed points for region growth according to gray value distribution of pixel points in the initial gray image;
taking any pixel point in the neighborhood of the seed point as a target pixel point, determining a neighborhood direction according to the direction of other pixel points in the neighborhood of the target pixel point from the seed point, and determining the initial influence degree of the target pixel point according to the gray value change and gradient included angle change of the pixel points in different neighborhood directions;
according to the gray level distribution and gradient direction distribution of the pixel points in different neighborhood directions, determining a correction coefficient, and correcting the initial influence degree according to the correction coefficient to obtain a target influence degree;
obtaining a growth coefficient according to the gray value of the seed point, the gray value of the target pixel point and the target influence degree, and carrying out regional growth treatment on the seed point according to the growth coefficient to obtain a growth region; and determining a defect area according to the gray value of the pixel point in the growth area.
2. The method for detecting defects of an aluminum alloy auto parts mold based on image processing according to claim 1, wherein determining seed points for region growth according to gray value distribution of pixel points in the initial gray image comprises:
constructing a gray distribution histogram of pixel points in the initial gray image, determining the number of pixel points in different gray intervals, and determining the duty ratio of the pixel points in different gray intervals according to the number of the pixel points in different gray intervals;
distributing a preset first number of seed points into corresponding gray intervals according to the duty ratio of the pixel points in different gray intervals, determining the number of the seed points in each gray interval, and randomly screening the seed points according to the number of the corresponding seed points in any gray interval.
3. The method for detecting defects of an aluminum alloy auto parts mold based on image processing according to claim 1, wherein determining a neighborhood direction according to a direction from other pixels in the neighborhood of the target pixel to a seed point comprises:
taking the pixel points except the seed point in the eight adjacent points of the target pixel point as other pixel points;
and determining the straight line direction between the seed point and any other pixel point as a neighborhood direction by taking the seed point as a starting point.
4. The method for detecting the defects of the aluminum alloy auto parts die based on image processing according to claim 1, wherein the determining the initial influence degree of the target pixel point according to the gray value change and the gradient included angle change of the pixel point in different neighborhood directions comprises:
in any neighborhood direction, taking a preset second number of pixel points which are closest to the seed point as reference pixel points; sequencing the reference pixel points according to the sequence from near to far from the seed points to obtain a reference sequence;
calculating the absolute value of the difference value of the gray values of two adjacent reference pixel points in the reference sequence as the adjacent gray difference value; when the adjacent gray level difference value is larger than a preset gray level difference value threshold value, adjusting the adjacent gray level difference value to be a first constant coefficient; when the adjacent gray level difference value is smaller than or equal to a preset gray level difference value threshold value, the adjacent gray level difference value is adjusted to be a second constant coefficient, wherein the first constant coefficient is smaller than the second constant coefficient; accumulating all the adjusted adjacent gray level difference values to obtain a gray level change coefficient;
calculating the difference value of gradient included angles of two adjacent reference pixel points in the reference sequence as an adjacent angle difference value; when the adjacent angle difference value is larger than a preset angle difference value threshold value, adjusting the adjacent angle difference value to be a third constant coefficient; when the adjacent angle difference value is smaller than or equal to a preset angle difference value threshold value, adjusting the adjacent angle difference value to be a fourth constant coefficient, wherein the third constant coefficient is smaller than the fourth constant coefficient; accumulating all the adjusted adjacent angle differences to obtain gradient included angle change coefficients;
calculating a normalized value of the product of the gray value change coefficient and the gradient included angle change coefficient as a change index of the neighborhood direction;
and taking the average value of the change indexes of all the neighborhood directions as the initial influence degree of the target pixel point.
5. The method for detecting defects of aluminum alloy auto parts mold based on image processing as claimed in claim 4, wherein the determining correction factors according to gray scale distribution and gradient direction distribution of pixels in different neighborhood directions comprises:
calculating average normalized values of adjacent gray level difference values of all reference pixel points in each neighborhood direction as gray level average coefficients of the corresponding neighborhood directions; sequencing the neighborhood direction based on a preset sequencing order, and performing curve fitting on the gray average coefficient according to the sequencing order to obtain a neighborhood gray scale curve;
calculating average normalized values of adjacent angle differences of all reference pixel points in each neighborhood direction as angle average coefficients of corresponding neighborhood directions; performing curve fitting on the angle mean coefficients according to the ordering sequence to obtain a neighborhood angle curve;
and calculating the similarity of the neighborhood gray level curve and the neighborhood angle curve based on a structural similarity algorithm, and taking the similarity as a correction coefficient.
6. The method for detecting defects of an aluminum alloy auto parts mold based on image processing according to claim 1, wherein the correcting the initial influence degree according to the correction coefficient to obtain a target influence degree comprises:
and calculating the product of the correction coefficient and the initial influence degree as a target influence degree.
7. The method for detecting defects of an aluminum alloy auto parts mold based on image processing according to claim 1, wherein the obtaining a growth coefficient according to the gray value of the seed point, the gray value of the target pixel point and the target influence degree comprises:
calculating the absolute value of the difference between the gray value of the seed point and the gray value of the target pixel point as gray difference;
and obtaining a growth coefficient according to the gray level difference and the target influence degree, wherein the gray level difference and the growth coefficient are in an inverse correlation relationship, the target influence degree and the growth coefficient are in a positive correlation relationship, and the value of the growth coefficient is a normalized value.
8. The method for detecting defects of an aluminum alloy auto parts mold based on image processing according to claim 1, wherein the performing the region growing process on the seed points according to the growth coefficients to obtain the growing regions comprises:
when the growth coefficient meets a preset growth condition, supplementing the target pixel point to an area where the seed point is located, and taking the target pixel point as a new seed point to perform area growth treatment;
and stopping the growth of the region when the growth coefficient does not meet the preset growth condition, and taking the obtained region as a growth region after stopping the growth in all directions.
9. The image processing-based aluminum alloy auto part mold defect detection method according to claim 8, wherein the preset growth condition is that the growth coefficient is greater than a preset coefficient threshold.
10. The method for detecting defects of an aluminum alloy auto parts mold based on image processing according to claim 1, wherein the determining a defective area according to gray values of pixel points in the growth area comprises:
calculating the average value of the gray values of all pixel points in any growth area as an area gray average value;
clustering the area gray average value of all the growing areas, and screening defective areas from the growing areas according to the clustering result.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311279017.0A CN117011297B (en) | 2023-10-07 | 2023-10-07 | Aluminum alloy automobile accessory die defect detection method based on image processing |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311279017.0A CN117011297B (en) | 2023-10-07 | 2023-10-07 | Aluminum alloy automobile accessory die defect detection method based on image processing |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN117011297A true CN117011297A (en) | 2023-11-07 |
| CN117011297B CN117011297B (en) | 2024-02-02 |
Family
ID=88569442
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202311279017.0A Active CN117011297B (en) | 2023-10-07 | 2023-10-07 | Aluminum alloy automobile accessory die defect detection method based on image processing |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117011297B (en) |
Cited By (12)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117197140A (en) * | 2023-11-07 | 2023-12-08 | 东莞市恒兴隆实业有限公司 | Irregular metal buckle forming detection method based on machine vision |
| CN117237344A (en) * | 2023-11-13 | 2023-12-15 | 深圳市帝一通讯有限公司 | Machine vision-based method for detecting appearance quality of electronic atomizer chip |
| CN117351214A (en) * | 2023-12-06 | 2024-01-05 | 上海电气集团腾恩驰科技(苏州)有限公司 | Heat tracing cable detection method and system based on image processing |
| CN117372435A (en) * | 2023-12-08 | 2024-01-09 | 智联信通科技股份有限公司 | Connector pin detection method based on image characteristics |
| CN117372436A (en) * | 2023-12-08 | 2024-01-09 | 红叶风电设备(营口)有限公司 | Wind power blade fault detection method based on image data |
| CN117388302A (en) * | 2023-12-13 | 2024-01-12 | 南通纳科达聚氨酯科技有限公司 | Defect detection method and system for anti-icing TPU film |
| CN117593305A (en) * | 2024-01-19 | 2024-02-23 | 天津康博轻钢制造有限公司 | Visual positioning system for high-frequency welding H-shaped steel |
| CN117611583A (en) * | 2024-01-22 | 2024-02-27 | 张家港飞腾复合新材料股份有限公司 | Artificial intelligence-based aluminum composite panel defect detection method and system |
| CN117788459A (en) * | 2024-02-23 | 2024-03-29 | 青岛正大正电力环保设备有限公司 | Chain abrasion detection method based on image characteristics |
| CN117773334A (en) * | 2024-02-28 | 2024-03-29 | 无锡丰毅汽车配件有限公司 | Automatic welding device for automobile skylight frame |
| CN117853487A (en) * | 2024-03-07 | 2024-04-09 | 浙江合丰科技有限公司 | FPC connector crack detection method and system based on image processing technology |
| CN117853487B (en) * | 2024-03-07 | 2024-05-14 | 浙江合丰科技有限公司 | FPC connector crack detection method and system based on image processing technology |
Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20120298862A1 (en) * | 2011-05-25 | 2012-11-29 | Chien-Huei Chen | Contour-based defect detection using an inspection apparatus |
| CN113689428A (en) * | 2021-10-25 | 2021-11-23 | 江苏南通元辰钢结构制造有限公司 | Mechanical part stress corrosion detection method and system based on image processing |
| WO2022027949A1 (en) * | 2020-08-04 | 2022-02-10 | 湖南大学 | Machine vision-based detecting method and system for glass bottle bottom defects |
| CN115100171A (en) * | 2022-07-11 | 2022-09-23 | 常宝云 | Steel die welding defect detection method and system based on machine vision |
| CN115311290A (en) * | 2022-10-12 | 2022-11-08 | 南通市通州区精华电器有限公司 | Method for detecting defects of metal parts of precision instrument |
| CN115861313A (en) * | 2023-02-24 | 2023-03-28 | 东莞市春草研磨科技有限公司 | Abrasion detection method for grinding head |
| CN115861317A (en) * | 2023-02-27 | 2023-03-28 | 深圳市伟利达精密塑胶模具有限公司 | Plastic mold production defect detection method based on machine vision |
| CN116309537A (en) * | 2023-04-24 | 2023-06-23 | 东莞市京品精密模具有限公司 | Defect detection method for oil stain on surface of tab die |
| CN116611748A (en) * | 2023-07-20 | 2023-08-18 | 吴江市高瑞庭园金属制品有限公司 | Titanium alloy furniture production quality monitoring system |
-
2023
- 2023-10-07 CN CN202311279017.0A patent/CN117011297B/en active Active
Patent Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20120298862A1 (en) * | 2011-05-25 | 2012-11-29 | Chien-Huei Chen | Contour-based defect detection using an inspection apparatus |
| WO2022027949A1 (en) * | 2020-08-04 | 2022-02-10 | 湖南大学 | Machine vision-based detecting method and system for glass bottle bottom defects |
| CN113689428A (en) * | 2021-10-25 | 2021-11-23 | 江苏南通元辰钢结构制造有限公司 | Mechanical part stress corrosion detection method and system based on image processing |
| CN115100171A (en) * | 2022-07-11 | 2022-09-23 | 常宝云 | Steel die welding defect detection method and system based on machine vision |
| CN115311290A (en) * | 2022-10-12 | 2022-11-08 | 南通市通州区精华电器有限公司 | Method for detecting defects of metal parts of precision instrument |
| CN115861313A (en) * | 2023-02-24 | 2023-03-28 | 东莞市春草研磨科技有限公司 | Abrasion detection method for grinding head |
| CN115861317A (en) * | 2023-02-27 | 2023-03-28 | 深圳市伟利达精密塑胶模具有限公司 | Plastic mold production defect detection method based on machine vision |
| CN116309537A (en) * | 2023-04-24 | 2023-06-23 | 东莞市京品精密模具有限公司 | Defect detection method for oil stain on surface of tab die |
| CN116611748A (en) * | 2023-07-20 | 2023-08-18 | 吴江市高瑞庭园金属制品有限公司 | Titanium alloy furniture production quality monitoring system |
Non-Patent Citations (1)
| Title |
|---|
| 乔佳亮;李晓阳;: "基于边缘检测和种子自动区域生长算法的铝铸件图像分割", 伺服控制, no. 03, pages 49 - 51 * |
Cited By (22)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117197140B (en) * | 2023-11-07 | 2024-02-20 | 东莞市恒兴隆实业有限公司 | Irregular metal buckle forming detection method based on machine vision |
| CN117197140A (en) * | 2023-11-07 | 2023-12-08 | 东莞市恒兴隆实业有限公司 | Irregular metal buckle forming detection method based on machine vision |
| CN117237344A (en) * | 2023-11-13 | 2023-12-15 | 深圳市帝一通讯有限公司 | Machine vision-based method for detecting appearance quality of electronic atomizer chip |
| CN117237344B (en) * | 2023-11-13 | 2024-01-19 | 深圳市帝一通讯有限公司 | Machine vision-based method for detecting appearance quality of electronic atomizer chip |
| CN117351214A (en) * | 2023-12-06 | 2024-01-05 | 上海电气集团腾恩驰科技(苏州)有限公司 | Heat tracing cable detection method and system based on image processing |
| CN117351214B (en) * | 2023-12-06 | 2024-02-13 | 上海电气集团腾恩驰科技(苏州)有限公司 | Heat tracing cable detection method and system based on image processing |
| CN117372435A (en) * | 2023-12-08 | 2024-01-09 | 智联信通科技股份有限公司 | Connector pin detection method based on image characteristics |
| CN117372436A (en) * | 2023-12-08 | 2024-01-09 | 红叶风电设备(营口)有限公司 | Wind power blade fault detection method based on image data |
| CN117372435B (en) * | 2023-12-08 | 2024-02-06 | 智联信通科技股份有限公司 | Connector pin detection method based on image characteristics |
| CN117372436B (en) * | 2023-12-08 | 2024-02-13 | 红叶风电设备(营口)有限公司 | Wind power blade fault detection method based on image data |
| CN117388302A (en) * | 2023-12-13 | 2024-01-12 | 南通纳科达聚氨酯科技有限公司 | Defect detection method and system for anti-icing TPU film |
| CN117388302B (en) * | 2023-12-13 | 2024-04-19 | 南通纳科达聚氨酯科技有限公司 | Defect detection method and system for anti-icing TPU film |
| CN117593305A (en) * | 2024-01-19 | 2024-02-23 | 天津康博轻钢制造有限公司 | Visual positioning system for high-frequency welding H-shaped steel |
| CN117593305B (en) * | 2024-01-19 | 2024-04-05 | 天津康博轻钢制造有限公司 | Visual positioning system for high-frequency welding H-shaped steel |
| CN117611583A (en) * | 2024-01-22 | 2024-02-27 | 张家港飞腾复合新材料股份有限公司 | Artificial intelligence-based aluminum composite panel defect detection method and system |
| CN117611583B (en) * | 2024-01-22 | 2024-04-19 | 张家港飞腾复合新材料股份有限公司 | Artificial intelligence-based aluminum composite panel defect detection method and system |
| CN117788459A (en) * | 2024-02-23 | 2024-03-29 | 青岛正大正电力环保设备有限公司 | Chain abrasion detection method based on image characteristics |
| CN117788459B (en) * | 2024-02-23 | 2024-04-30 | 青岛正大正电力环保设备有限公司 | Chain abrasion detection method based on image characteristics |
| CN117773334A (en) * | 2024-02-28 | 2024-03-29 | 无锡丰毅汽车配件有限公司 | Automatic welding device for automobile skylight frame |
| CN117773334B (en) * | 2024-02-28 | 2024-05-14 | 无锡丰毅汽车配件有限公司 | Automatic welding device for automobile skylight frame |
| CN117853487A (en) * | 2024-03-07 | 2024-04-09 | 浙江合丰科技有限公司 | FPC connector crack detection method and system based on image processing technology |
| CN117853487B (en) * | 2024-03-07 | 2024-05-14 | 浙江合丰科技有限公司 | FPC connector crack detection method and system based on image processing technology |
Also Published As
| Publication number | Publication date |
|---|---|
| CN117011297B (en) | 2024-02-02 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN117011297B (en) | Aluminum alloy automobile accessory die defect detection method based on image processing | |
| CN114972343B (en) | Method for detecting production abnormity of laser welding product | |
| CN116843678B (en) | Hard carbon electrode production quality detection method | |
| CN111667470B (en) | Industrial pipeline flaw detection inner wall detection method based on digital image | |
| CN116091504B (en) | Connecting pipe connector quality detection method based on image processing | |
| CN116934740B (en) | Plastic mold surface defect analysis and detection method based on image processing | |
| CN116977358B (en) | Visual auxiliary detection method for corrugated paper production quality | |
| CN115294159B (en) | Method for dividing corroded area of metal fastener | |
| CN116228758B (en) | Internal bubble detection method for polarizer production | |
| CN116402808B (en) | Intelligent detection method for manufacturing cable clamp plate die | |
| CN116309599B (en) | Water quality visual monitoring method based on sewage pretreatment | |
| CN116416252B (en) | Method for detecting sedimentation image of wastewater in boehmite production process | |
| CN117351019B (en) | Welding defect detection method | |
| CN117557820B (en) | Quantum dot optical film damage detection method and system based on machine vision | |
| CN116012357A (en) | Automobile circular trailer hole air line defect detection method based on machine vision | |
| CN116152247B (en) | Quality evaluation method of polyurethane coating film | |
| CN117237646B (en) | PET high-temperature flame-retardant adhesive tape flaw extraction method and system based on image segmentation | |
| CN110930425B (en) | Damaged target detection method based on neighborhood vector inner product local contrast image enhancement | |
| CN117095004A (en) | Excavator walking frame main body welding deformation detection method based on computer vision | |
| CN116245880A (en) | Electric vehicle charging pile fire risk detection method based on infrared identification | |
| CN117315289B (en) | Aeroengine blade contour edge detection method based on image processing | |
| CN116452589A (en) | Intelligent detection method for surface defects of artificial board based on image processing | |
| CN116883446B (en) | Real-time monitoring system for grinding degree of vehicle-mounted camera lens | |
| CN116883408B (en) | Integrating instrument shell defect detection method based on artificial intelligence | |
| CN117522719A (en) | Bronchoscope image auxiliary optimization system based on machine learning |
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 |