CN113686874B - Mechanical part damage detection method and system based on artificial intelligence - Google Patents
Mechanical part damage detection method and system based on artificial intelligence Download PDFInfo
- Publication number
- CN113686874B CN113686874B CN202110936579.2A CN202110936579A CN113686874B CN 113686874 B CN113686874 B CN 113686874B CN 202110936579 A CN202110936579 A CN 202110936579A CN 113686874 B CN113686874 B CN 113686874B
- Authority
- CN
- China
- Prior art keywords
- crack
- edge
- section
- segment
- branch
- 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.)
- Active
Links
Images
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N21/00—Investigating or analysing materials by the use of optical means, i.e. using sub-millimetre waves, infrared, visible or ultraviolet light
- G01N21/84—Systems specially adapted for particular applications
- G01N21/88—Investigating the presence of flaws or contamination
- G01N21/8851—Scan or image signal processing specially adapted therefor, e.g. for scan signal adjustment, for detecting different kinds of defects, for compensating for structures, markings, edges
-
- 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
- G06T7/10—Segmentation; Edge detection
- G06T7/13—Edge detection
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N21/00—Investigating or analysing materials by the use of optical means, i.e. using sub-millimetre waves, infrared, visible or ultraviolet light
- G01N21/84—Systems specially adapted for particular applications
- G01N21/88—Investigating the presence of flaws or contamination
- G01N21/8851—Scan or image signal processing specially adapted therefor, e.g. for scan signal adjustment, for detecting different kinds of defects, for compensating for structures, markings, edges
- G01N2021/8887—Scan or image signal processing specially adapted therefor, e.g. for scan signal adjustment, for detecting different kinds of defects, for compensating for structures, markings, edges based on image processing techniques
-
- 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
Landscapes
- Engineering & Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Life Sciences & Earth Sciences (AREA)
- Signal Processing (AREA)
- Health & Medical Sciences (AREA)
- Quality & Reliability (AREA)
- Chemical & Material Sciences (AREA)
- Analytical Chemistry (AREA)
- Biochemistry (AREA)
- General Health & Medical Sciences (AREA)
- Immunology (AREA)
- Pathology (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 artificial intelligence, in particular to a method and a system for detecting damage of a mechanical part based on artificial intelligence. The method comprises the following steps: and carrying out edge detection on the surface image of the part to obtain a crack edge. And thinning the crack edge to obtain a thinned edge. And segmenting the refined edge through the end points and the cross points of the refined edge to obtain the refined crack segment and the crack segment edge corresponding to the crack segment. And screening out a main crack section and a branch crack section according to the width of the edge of the crack section. And obtaining the damage degree of the first part according to the length and the width of the main crack section and the branch crack section. And connecting the central point of the branch crack section with the central point of the main crack section to obtain a crack vector. And acquiring crack dispersion information and crack propagation information according to the crack vector, and acquiring the damage degree of the second part according to the crack dispersion information and the crack propagation information. The invention comprehensively judges the damage condition of the parts and improves the detection accuracy.
Description
Technical Field
The invention relates to the technical field of artificial intelligence, in particular to a method and a system for detecting damage of a mechanical part based on artificial intelligence.
Background
Mechanical equipment parts, under the combined action of stress and corrosive media, will exhibit brittle cracking below the material strength limit, a phenomenon known as stress corrosion cracking. The occurrence of cracks can reduce the safety of the structural system and even lead to failure of the entire part. Therefore, it is necessary to detect the stress corrosion cracks on the surface of the part and obtain the damage degree of the part surface, so as to determine the subsequent processing operation on the mechanical part.
In the prior art, machine vision technology can be used to obtain damage information of a mechanical part through a surface image of the part. Because of the irregular formation of cracks, many divergent branch cracks appear on the trunk cracks during the damage process. Since the crack information is rich, if the overall crack state and crack distribution cannot be considered, the reliability and accuracy of the damage detection are reduced.
Disclosure of Invention
In order to solve the technical problems, the invention aims to provide a mechanical part damage detection method and system based on artificial intelligence, and the adopted technical scheme is as follows:
the invention provides a mechanical part damage detection method based on artificial intelligence, which comprises the following steps
Acquiring a part surface image; carrying out edge detection on the surface image of the part to obtain a crack edge;
thinning the crack edge to obtain a thinned edge; acquiring the end point and the intersection point of the refined edge; segmenting the refined edge according to the end point and the intersection point to obtain a refined crack segment; obtaining the corresponding crack section edge on the crack edge according to the refined crack section; screening out a main crack section and a branch crack section according to the width of the edge of the crack section;
obtaining the damage degree of a first part according to the lengths and the widths of the main crack section and the branch crack section;
connecting the central point of the branch crack section with the central point of the main crack section to obtain a crack vector; one said branch crack segment corresponds to a plurality of said crack vectors; acquiring crack dispersion information according to the angle information of the crack vector; acquiring crack propagation information according to the length of the crack vector; obtaining a second part damage degree according to the crack dispersion information and the crack propagation information;
and judging the damage condition of the part according to the damage degree of the first part and the damage degree of the second part.
Further, the obtaining the end point and the intersection point of the refined edge comprises:
detecting the number of adjacent pixels of each pixel in the refined edge; if the number of the adjacent pixel points is one, the corresponding pixel point is the endpoint; and if the number of the adjacent pixel points is more than or equal to three, the corresponding pixel point is the cross point.
Further, the obtaining of the corresponding crack segment edge on the part edge according to the refined crack segment includes:
subtracting the edge of the part from the refined crack section to obtain a difference pixel point; obtaining the distance between the difference pixel point and the refined crack segment; and taking the difference pixel point with the minimum distance and the refined crack segment as similar crack segment pixels, wherein the similar crack segment pixels form the crack segment edge.
Further, screening out a main crack section and a branch crack section according to the width of the crack section edge comprises:
obtaining an average width of the crack segment edge; taking the crack segment edge with the largest average width as a main crack segment; taking the crack section edge intersected with the main crack section as a related crack section; obtaining a difference in average width of the associated crack segment and the trunk crack segment; if the difference is greater than a preset width threshold value, the corresponding related crack segment is the branch crack segment; and otherwise, the corresponding related crack segment is the main crack segment.
Further, the obtaining an average width of the crack segment edge comprises; obtaining the average width according to an average width obtaining formula, wherein the average width obtaining formula comprises:
wherein d is i Is the average width, n, of the ith crack segment edge i2 The number n of the difference pixel points in the ith crack segment edge i1 The number of the pixel points of the refined crack segment in the ith crack segment edge is shown.
Further, the obtaining the first part damage degree according to the lengths and the widths of the main crack section and the branch crack section comprises:
taking the product of the length of the main crack section, the width of the main crack section and a preset damage weight of the main crack section as the damage degree of the main crack; taking the product of the length of the branch crack section, the width of the branch crack section and a preset damage weight of the branch crack section as the damage degree of the branch crack; the damage weight of the main crack section is greater than that of the branch crack section; taking the sum of all of the trunk crack damage levels and all of the branch crack damage levels as the first part damage level.
Further, the obtaining crack dispersion information according to the angle information of the crack vector comprises:
obtaining the average vector angle of all the crack vectors corresponding to the branch crack sections; and taking the variance of the average vector angle as the crack dispersion information.
Further, the obtaining crack propagation information according to the length of the crack vector comprises:
acquiring the average vector length of all the crack vectors corresponding to the branch crack sections; accumulating all the average vector lengths to obtain the crack propagation length; obtaining the surface area of the part according to the part surface image; and taking the ratio of the crack propagation length to the surface area of the part as the crack propagation information.
Further, the obtaining a second part damage degree according to the crack propagation information and the crack dispersion information comprises:
and taking the product of the crack dispersion information and the crack propagation information as the damage degree of the second part.
The invention also provides an artificial intelligence based mechanical part damage detection system, which comprises a memory, a processor and a computer program which is stored in the memory and can run on the processor, wherein when the processor executes the computer program, any one of the steps of the artificial intelligence based mechanical part damage detection method is realized.
The invention has the following beneficial effects:
1. in the exemplary embodiment of the invention, a first degree of damage, which is representative of the influence of the crack itself on the component, is obtained by the length and width of the crack. And obtaining a second damage degree representing the influence of the crack distribution on the part through the crack dispersion information and the crack propagation information. The part damage condition is comprehensively judged through the first damage degree and the second damage degree, and the detection accuracy is improved.
2. In the embodiment of the invention, the crack edge is divided into a plurality of crack section edges through the end points and the cross points, and the complex disordered integral cracks are classified according to the width of the crack section edges to obtain the main crack section and the branch crack section, so that the orderliness of subsequent damage detection is improved, and the complexity of damage detection is reduced.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions and advantages of the prior art, the drawings used in the description of the embodiments or the prior art will be briefly described below, it is obvious that the drawings in the following description are only some embodiments of the present invention, and other drawings can be obtained by those skilled in the art without creative efforts.
Fig. 1 is a flowchart of a method for detecting damage to a mechanical part based on artificial intelligence according to an embodiment of the present invention.
Detailed Description
In order to further explain the technical means and effects of the present invention adopted to achieve the predetermined objects, the following detailed description, the structure, the features and the effects of the method and the system for detecting damage to a mechanical part based on artificial intelligence according to the present invention are provided with reference to the accompanying drawings and the preferred embodiments. In the following description, different "one embodiment" or "another embodiment" refers to not necessarily the same embodiment. Furthermore, the particular features, structures, or characteristics may be combined in any suitable manner in one or more embodiments.
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 following describes a specific scheme of a method and a system for detecting damage to a mechanical part based on artificial intelligence in detail with reference to the accompanying drawings.
Referring to fig. 1, a flowchart of a method for detecting damage to a mechanical part based on artificial intelligence according to an embodiment of the present invention is shown, where the method includes:
step S1: acquiring a part surface image; and carrying out edge detection on the surface image of the part to obtain a crack edge.
And arranging a camera above the surface of the part to be detected, so that the camera can shoot a clear and complete part surface image. In order to make the information in the part surface image more obvious in the subsequent processing process, in the embodiment of the present invention, obtaining the part surface image includes: and carrying out graying processing on the surface image of the part to obtain a grayscale image. And processing the gray image by a threshold segmentation technology, setting the gray value of the pixel point with the gray value lower than 80 as 0, and setting the gray value of the pixel point with the gray value not lower than 80 as 255 to obtain a binary image. The binary image only contains crack information and background information, so that subsequent edge detection is facilitated.
And carrying out edge detection on the binary image of the part surface image to obtain a crack edge. In the embodiment of the invention, a Sobel operator is adopted to extract the high-frequency component of the binary image, and the high-frequency component is superposed on the binary image to obtain a sharpened image. And (3) highlighting outline and detail information in the sharpened image, namely obtaining the crack edge.
Step S2: thinning the crack edge to obtain a thinned edge; acquiring end points and cross points of the refined edges; and segmenting the refined edge according to the end points and the intersection points to obtain a refined crack segment. Obtaining the corresponding crack section edge on the crack edge according to the refined crack section; and screening out a main crack section and a branch crack section according to the width of the edge of the crack section.
For stress corrosion cracking, the propagation of the main crack is often accompanied by multiple branch cracks, and in order to make the damage detection more comprehensive and accurate, the cracks need to be classified. The method specifically comprises the following steps:
1) thinning the crack edge, namely maintaining the basic shape and structure of the crack, converting the crack into a group of simple single-pixel arcs and obtaining the thinned edge
2) And detecting the quantity of adjacent pixels of each pixel in the refined edge. If the number of the adjacent pixel points is one, the corresponding pixel point is an end point; and if the number of the adjacent pixel points is more than or equal to three, the corresponding pixel points are cross points. The end points and intersections of the refined edges are obtained. In the embodiment of the invention, in order to reduce the calculation amount, the number of adjacent pixel points is detected in the range of 8 neighborhoods of each pixel point in the refined edge.
3) And segmenting the refined edge according to the end points and the intersection points, taking the end points or the intersection points as the starting points of a section of crack, and continuously taking the adjacent pixel points as the components until the next adjacent pixel point intersection point or the end point to obtain the refined crack section. The refined crack segments comprise (end points, intersections) composed crack segments and (intersections ) composed crack segments.
4) Obtaining a corresponding crack segment edge on the crack edge from the refined crack segment includes: subtracting the edge of the part from the thinned crack section to obtain a difference pixel point; obtaining the distance between the difference pixel point and the refined crack segment; and taking the difference pixel point with the minimum distance and the refined crack segment as the same type of crack segment pixels, wherein the same type of crack segment pixels form the crack segment edge.
5) Because the width difference of main crack section and branch crack section is great, consequently can screen out main crack section and branch crack section according to the width at crack section edge, specifically include: the average width of the crack segment edge is obtained. And taking the crack section edge with the largest average width as a main crack section. And taking the crack segment edge intersected with the main crack segment as a related crack segment. The difference of the average width of the related crack segment and the main crack segment is obtained. If the difference is greater than the preset width threshold value, the corresponding related crack segment is a branch crack segment; otherwise, the corresponding related crack segment is the main crack segment. And traversing all the crack segment edges to obtain the category of each crack segment edge. In the embodiment of the invention, the width threshold value is set to be 0.1d in consideration of the difference of the widths of different main dry crack sections max Wherein d is max Different main crack sections have different width thresholds in the classification process for the edges of the main crack sections, and related crack sections are screened according to the width thresholds corresponding to the main crack sections, so that the classification accuracy is improved.
Preferably, because of the irregularity of the crack edge, the widths of different positions of the same crack segment are different, so that the average width can be quantized by the ratio of the number of the different pixel points in the crack segment edge to the number of the pixel points in the refined crack segment. The thinned crack segment is a single-pixel edge, so that the number of the pixel points of the thinned crack segment represents the length of the crack segment edge, and the larger the ratio is, the larger the average width of the crack segment edge is. Further adjusting the value of the average width, fitting by a mathematical modeling method to obtain an average width formula, and obtaining the average width according to the average width obtaining formula, wherein the average width obtaining formula comprises:
wherein d is i Is the average width of the i-th crack segment edge, n i2 Number of different pixel points in the ith crack segment edge, n i1 And the number of pixel points of the thinned crack segment in the ith crack segment edge is counted.
In the embodiment of the invention, the image of the crack edge is labeled according to the crack segment type, the main crack segment label is set to be 1, and the branch crack segment is set to be 2. Taking the number of the pixel points of the refined crack segment corresponding to the edge of each crack segment as the length of the edge of the crack segment, and combining the average width to obtain the state information of each crack segment: [ b ] a i ，n i1 ，d i ]Wherein b is i Is the index of the ith crack segment edge, n i1 Is the length of the edge of the ith crack segment, d i The average width of the ith crack segment edge.
Step S3: and obtaining the damage degree of the first part according to the length and the width of the main crack section and the branch crack section.
The length and width of the crack segment edge represent the effect of the crack itself on the part. Because the main cracks and the branch cracks have different influence degrees, the main cracks and the branch cracks need to be analyzed separately according to the types of the edges of each crack segment, and the method specifically comprises the following steps:
taking the product of the length of the main crack section, the width of the main crack section and a preset damage weight of the main crack section as the damage degree of the main crack; and taking the product of the length of the branch crack section, the width of the branch crack section and a preset damage weight of the branch crack section as the damage degree of the branch crack. The damage weight of the main crack section is larger than that of the branch crack section. The sum of all trunk crack damage levels and all branch crack damage levels is taken as the first part damage level. In the embodiment of the invention, the damage weight of the main crack section is set to be 0.6, and the damage weight of the branch crack section is set to be 0.4.
Step S4: connecting the central point of the branch crack section with the central point of the main crack section to obtain a crack vector; one branch crack segment corresponds to a plurality of crack vectors; acquiring crack dispersion information according to the angle information of the crack vector; obtaining crack propagation information according to the length of the crack vector; and obtaining the damage degree of the second part according to the crack dispersion information and the crack propagation information.
In the embodiment of the invention, the sum of the distances between each pixel point and other pixel points in the edge of the crack segment is obtained, and the minimum distance and the corresponding pixel point are used as the central point of the edge of the crack segment.
Connecting the central point of the branch crack section with the central point of the main crack section to obtain crack vectors, wherein one branch crack section corresponds to a plurality of crack vectors, the crack vectors contain angle information and length information, and the angle information and the length information are marked as (theta) ij ，d ij ) Wherein theta ij For the angle of the crack vector pointing to the jth main crack segment for the ith branch crack segment, d ij The ith branch crack segment points to the length of the crack vector of the jth main crack segment.
Obtaining the average vector angle of all crack vectors corresponding to the branch crack section
Namely, it is
Wherein m is 1 The number of the main crack sections is the number of the main crack sections,
is the average vector angle of the ith branch crack segment. The variance of the mean vector angle is taken as crack dispersion information V 1 . Namely, it is
Wherein m is 2 Is the number of branch crack segments.
Obtaining the average vector length of all crack vectors corresponding to the branch crack section
Namely, it is
Wherein the content of the first and second substances,
is the average vector length of the ith branch crack segment. The crack propagation length is obtained by adding up all the average vector lengths. And obtaining the surface area of the part according to the surface image of the part. Taking the ratio of the crack propagation length to the surface area of the part as crack propagation information V 2 . Namely, it is
Wherein N is the surface area of the part. In the embodiment of the invention, the surface area of the part is divided according to the surface image of the part, and the area of the surface area of the part under a pixel coordinate system, namely the surface area of the part, is obtained.
Combining the crack propagation information and the crack dispersion information, and taking the product of the crack propagation information and the crack dispersion information as the second part damage degree V, namely V ═ V 1 V 2 。
Step S5: and judging the damage condition of the part according to the damage degree of the first part and the damage degree of the second part.
The first part damage level represents the effect of the crack itself on the part, and the second part damage level represents the effect of the crack distribution on the part. Therefore, the damage condition of the part is jointly judged by combining the damage degree of the first part and the damage degree of the second part.
In the embodiment of the invention, the damage degree of the first part is divided by the surface area of the part to realize normalization. And normalizing the parameters related to the damage degree of the second part to obtain the damage degree of the second part. And taking the product of the damage degree of the first part and the damage degree of the second part as the damage degree of the part. When the damage degree of the part is greater than a preset first damage threshold value, the damage condition of the current part is considered to be large, the part is at the risk of failure and scrapping, and in order to ensure the smooth operation and the safety of the working process, the damaged part needs to be replaced in time, and the working environment is checked; when the damage degree of the part is greater than or equal to the preset second loss threshold value and less than or equal to the first damage threshold value, the damage condition of the current part is considered to be general, the part can be repaired or replaced in consideration of the working efficiency and the working safety, and the working environment is checked; when the damage degree of the part is less than or equal to the second damage threshold, the damage condition of the current part is considered to be small, the part can be repaired or replaced without need, the working environment is checked, and the probability of further expansion of the crack is reduced. Because the first part damage level and the second part damage level are normalized by the part surface area and have smaller values, the first damage threshold is set to 0.001 and the second damage threshold is set to 0.00005.
In summary, the crack edge is obtained by performing edge detection on the surface image of the part in the embodiment of the invention. And thinning the crack edge to obtain a thinned edge. And segmenting the refined edge through the end points and the cross points of the refined edge to obtain the refined crack segment and the crack segment edge corresponding to the crack segment. And screening out a main crack section and a branch crack section according to the width of the crack section edge. And obtaining the damage degree of the first part according to the length and the width of the main crack section and the branch crack section. And connecting the central point of the branch crack section with the central point of the main crack section to obtain a crack vector. And acquiring crack dispersion information and crack propagation information according to the crack vector, and acquiring the damage degree of the second part according to the crack dispersion information and the crack propagation information. The part damage condition is comprehensively judged, and the detection accuracy is improved.
The invention also provides a mechanical part damage detection system based on artificial intelligence, which comprises a memory, a processor and a computer program which is stored in the memory and can run on the processor, wherein when the processor executes the computer program, any step of the mechanical part damage detection method based on artificial intelligence is realized.
It should be noted that: the precedence order of the above embodiments of the present invention is only for description, and does not represent the merits of the embodiments. And that specific embodiments have been described above. Other embodiments are within the scope of the following claims. In some cases, the actions or steps recited in the claims may be performed in a different order than in the embodiments and still achieve desirable results. In addition, the processes depicted in the accompanying figures do not necessarily require the particular order shown, or sequential order, to achieve desirable results. In some embodiments, multitasking and parallel processing may also be possible or may be advantageous.
The embodiments in the present specification are described in a progressive manner, and the same and similar parts among the embodiments are referred to each other, and each embodiment focuses on the differences from the other embodiments.
The above description is only for the purpose of illustrating the preferred embodiments of the present invention and is not to be construed as limiting the invention, and any modifications, equivalents, improvements and the like that fall within the spirit and principle of the present invention are intended to be included therein.
Claims (4)
1. A mechanical part damage detection method based on artificial intelligence is characterized by comprising the following steps:
acquiring a part surface image; carrying out edge detection on the surface image of the part to obtain a crack edge;
thinning the crack edge to obtain a thinned edge; acquiring the end point and the intersection point of the refined edge; segmenting the refined edge according to the end points and the cross points to obtain refined crack segments; obtaining the corresponding crack section edge on the crack edge according to the refined crack section; screening out a main crack section and a branch crack section according to the width of the edge of the crack section;
obtaining the damage degree of the first part according to the length and the width of the main crack section and the branch crack section;
connecting the central point of the branch crack section with the central point of the main crack section to obtain a crack vector; one said branch crack segment corresponds to a plurality of said crack vectors; acquiring crack dispersion information according to the angle information of the crack vector; acquiring crack propagation information according to the length of the crack vector; obtaining a second part damage degree according to the product of the crack dispersion information and the crack propagation information;
judging the damage condition of the part according to the damage degree of the first part and the damage degree of the second part;
the obtaining of the damage degree of the first part according to the length and the width of the main crack section and the branch crack section comprises:
taking the product of the length of the main crack section, the width of the main crack section and a preset damage weight of the main crack section as the damage degree of the main crack; taking the product of the length of the branch crack section, the width of the branch crack section and a preset damage weight of the branch crack section as the damage degree of the branch crack; the damage weight of the main crack section is greater than that of the branch crack section; taking the sum of all of the trunk crack damage degrees and all of the branch crack damage degrees as the first part damage degree;
the obtaining crack dispersion information according to the angle information of the crack vector includes:
obtaining the average vector angle of all the crack vectors corresponding to the branch crack sections; taking the variance of the mean vector angle as the crack dispersion information;
screening out a main crack section and a branch crack section according to the width of the crack section edge comprises the following steps:
obtaining an average width of the crack segment edge; taking the crack segment edge with the largest average width as a main crack segment; taking the crack section edge intersected with the main crack section as a related crack section; obtaining a difference in average width of the associated crack segment and the trunk crack segment; if the difference is greater than a preset width threshold value, the corresponding related crack segment is the branch crack segment; otherwise, the corresponding related crack segment is the main crack segment;
the obtaining crack propagation information according to the length of the crack vector comprises:
acquiring the average vector length of all the crack vectors corresponding to the branch crack sections; accumulating all the average vector lengths to obtain the crack propagation length; obtaining the surface area of the part according to the part surface image; taking the ratio of the crack propagation length to the surface area of the part as the crack propagation information; the obtaining of the average width of the crack segment edge comprises: obtaining the average width according to an average width obtaining formula, wherein the average width obtaining formula comprises:
wherein d is i Is the average width, n, of the ith crack segment edge i2 The number of the different pixel points in the ith crack segment edge, n i1 The number of pixel points of the refined crack segment in the ith crack segment edge is counted; and the difference pixel point is obtained by subtracting the refined crack section from the edge of the part.
2. The method for detecting damage to mechanical parts based on artificial intelligence of claim 1, wherein the obtaining of the end points and intersection points of the refined edges comprises:
detecting the number of adjacent pixels of each pixel in the refined edge; if the number of the adjacent pixel points is one, the corresponding pixel point is the endpoint; and if the number of the adjacent pixel points is more than or equal to three, the corresponding pixel point is the cross point.
3. The artificial intelligence based mechanical part damage detection method according to claim 1, wherein the obtaining of the corresponding crack segment edge on the crack edge according to the refined crack segment includes: subtracting the edge of the part from the refined crack section to obtain a difference pixel point; obtaining the distance between the difference pixel point and the refined crack segment; and taking the difference pixel point with the minimum distance and the refined crack segment as similar crack segment pixels, wherein the similar crack segment pixels form the crack segment edge.
4. An artificial intelligence based damage detection system for a mechanical part, comprising a memory, a processor and a computer program stored in the memory and executable on the processor, wherein the processor implements the method according to any one of claims 1 to 3 when executing the computer program.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110936579.2A CN113686874B (en) | 2021-08-16 | 2021-08-16 | Mechanical part damage detection method and system based on artificial intelligence |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110936579.2A CN113686874B (en) | 2021-08-16 | 2021-08-16 | Mechanical part damage detection method and system based on artificial intelligence |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN113686874A CN113686874A (en) | 2021-11-23 |
| CN113686874B true CN113686874B (en) | 2022-08-02 |
Family
ID=78580268
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110936579.2A Active CN113686874B (en) | 2021-08-16 | 2021-08-16 | Mechanical part damage detection method and system based on artificial intelligence |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113686874B (en) |
Families Citing this family (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114723700B (en) * | 2022-03-31 | 2023-05-12 | 河北鹰眼智能科技有限公司 | Method and system for predicting surface crack growth of iron furnace based on computer vision |
| CN114972187B (en) * | 2022-04-20 | 2024-01-02 | 烟台大视工业智能科技有限公司 | Crack defect evaluation method based on artificial intelligence |
| CN117876359B (en) * | 2024-03-04 | 2024-06-14 | 金品计算机科技(天津)有限公司 | Rubber product defect detection method, device, equipment and medium |
Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104392224A (en) * | 2014-12-04 | 2015-03-04 | 西南交通大学 | Crack detection method for road surface |
| CN106127777A (en) * | 2016-06-27 | 2016-11-16 | 中山大学 | A kind of three dimensions crack separation identification and characterizing method |
| CN109886921A (en) * | 2019-01-16 | 2019-06-14 | 新而锐电子科技(上海)有限公司 | Crack size measure, device and electronic equipment based on digital picture |
| CN110298802A (en) * | 2019-06-28 | 2019-10-01 | 广东诚泰交通科技发展有限公司 | Road surface crack detection method and device |
| CN110751644A (en) * | 2019-10-23 | 2020-02-04 | 上海应用技术大学 | Road surface crack detection method |
| CN111256594A (en) * | 2020-01-18 | 2020-06-09 | 中国人民解放军国防科技大学 | Method for measuring physical characteristics of surface state of aircraft skin |
| CN112348770A (en) * | 2020-09-09 | 2021-02-09 | 陕西师范大学 | Bridge crack detection method based on multi-resolution convolution network |
| CN113160202A (en) * | 2021-04-30 | 2021-07-23 | 汕头大学 | Crack detection method and system |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108463717B (en) * | 2016-01-26 | 2021-05-25 | 富士胶片株式会社 | Crack information detection device, crack information detection method, and storage medium |
-
2021
- 2021-08-16 CN CN202110936579.2A patent/CN113686874B/en active Active
Patent Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104392224A (en) * | 2014-12-04 | 2015-03-04 | 西南交通大学 | Crack detection method for road surface |
| CN106127777A (en) * | 2016-06-27 | 2016-11-16 | 中山大学 | A kind of three dimensions crack separation identification and characterizing method |
| CN109886921A (en) * | 2019-01-16 | 2019-06-14 | 新而锐电子科技(上海)有限公司 | Crack size measure, device and electronic equipment based on digital picture |
| CN110298802A (en) * | 2019-06-28 | 2019-10-01 | 广东诚泰交通科技发展有限公司 | Road surface crack detection method and device |
| CN110751644A (en) * | 2019-10-23 | 2020-02-04 | 上海应用技术大学 | Road surface crack detection method |
| CN111256594A (en) * | 2020-01-18 | 2020-06-09 | 中国人民解放军国防科技大学 | Method for measuring physical characteristics of surface state of aircraft skin |
| CN112348770A (en) * | 2020-09-09 | 2021-02-09 | 陕西师范大学 | Bridge crack detection method based on multi-resolution convolution network |
| CN113160202A (en) * | 2021-04-30 | 2021-07-23 | 汕头大学 | Crack detection method and system |
Also Published As
| Publication number | Publication date |
|---|---|
| CN113686874A (en) | 2021-11-23 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN113686874B (en) | Mechanical part damage detection method and system based on artificial intelligence | |
| CN115330783A (en) | Steel wire rope defect detection method | |
| CN113205063A (en) | Visual identification and positioning method for defects of power transmission conductor | |
| CN114359270A (en) | Computer vision-based automobile engine oil way copper bush defect detection method | |
| CN114897896B (en) | Building wood defect detection method based on gray level transformation | |
| CN111008961A (en) | Transmission line equipment defect detection method and system, equipment and medium thereof | |
| WO2021118463A1 (en) | Defect detection in image space | |
| CN115294159B (en) | Method for dividing corroded area of metal fastener | |
| CN114863492B (en) | Method and device for repairing low-quality fingerprint image | |
| CN117689655B (en) | Metal button surface defect detection method based on computer vision | |
| CN114972187A (en) | Crack defect evaluation method based on artificial intelligence | |
| CN112991283A (en) | Flexible IC substrate line width detection method based on super-pixels, medium and equipment | |
| CN114897773A (en) | Distorted wood detection method and system based on image processing | |
| CN115222743A (en) | Furniture surface paint spraying defect detection method based on vision | |
| Truong et al. | Automatic multi-thresholds selection for image segmentation based on evolutionary approach | |
| CN115100174A (en) | Ship sheet metal part paint surface defect detection method | |
| CN117788472B (en) | Method for judging corrosion degree of rivet on surface of aircraft skin based on DBSCAN algorithm | |
| Mishchenko et al. | Model-based chart image classification | |
| CN115082449B (en) | Electronic component defect detection method | |
| CN110175257B (en) | Method for matching line manuscript images, electronic equipment and storage medium | |
| Basu et al. | Recognition of similar shaped handwritten characters using logistic regression | |
| CN115984219A (en) | Product surface defect detection method and device, electronic equipment and storage medium | |
| Tabassum et al. | An approach to recognize book title from multi-cell bookshelf images | |
| CN112418210A (en) | Intelligent classification method for tower inspection information | |
| Sanders et al. | Pre-locating corners in images in order to improve the extraction of Fourier descriptors and subsequent recognition of shipbuilding parts |
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 | ||
| TR01 | Transfer of patent right | ||
| TR01 | Transfer of patent right |
Effective date of registration: 20230411 Address after: No. 136-23, Wisdom Second Street, Hunnan District, Shenyang City, Liaoning Province, 110000 (1-23-2) Patentee after: Shenyang Chenyang Information Technology Co.,Ltd. Address before: 223600 No. 8, Wuji Town Industrial Park, Shuyang County, Suqian City, Jiangsu Province Patentee before: Shuyang Linran Plastic Industry Co.,Ltd. |