CN116433665B - Aircraft component defect online identification system based on visual detection - Google Patents

Aircraft component defect online identification system based on visual detection Download PDF

Info

Publication number
CN116433665B
CN116433665B CN202310695760.8A CN202310695760A CN116433665B CN 116433665 B CN116433665 B CN 116433665B CN 202310695760 A CN202310695760 A CN 202310695760A CN 116433665 B CN116433665 B CN 116433665B
Authority
CN
China
Prior art keywords
index
pixel point
color
defect
target
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
Application number
CN202310695760.8A
Other languages
Chinese (zh)
Other versions
CN116433665A (en
Inventor
许凯宁
李顶河
万傲霜
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Civil Aviation University of China
Original Assignee
Civil Aviation University of China
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Civil Aviation University of China filed Critical Civil Aviation University of China
Priority to CN202310695760.8A priority Critical patent/CN116433665B/en
Publication of CN116433665A publication Critical patent/CN116433665A/en
Application granted granted Critical
Publication of CN116433665B publication Critical patent/CN116433665B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

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
    • G06T7/0002—Inspection of images, e.g. flaw detection
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/10—Segmentation; Edge detection
    • G06T7/11—Region-based segmentation
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/90—Determination of colour characteristics
    • 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)
  • Computer Vision & Pattern Recognition (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • Quality & Reliability (AREA)
  • Image Analysis (AREA)

Abstract

The invention relates to the technical field of image data processing, in particular to an on-line identification system for defects of an aircraft component based on visual detection, which comprises the following components: the following steps can be realized by the mutual cooperation among a plurality of modules: acquiring a target space image corresponding to an aircraft component to be detected; performing color difference analysis processing on each pixel point in the target space image; performing color contrast analysis processing on each pixel point in the target space image in a preset number of preset directions; determining defect abnormality indexes corresponding to each pixel point in the target space image; and identifying crack defects of the aircraft component to be detected according to the defect abnormal index and the color abnormal index corresponding to the pixel points in the target space image. According to the invention, through data processing of the target space image, the efficiency of crack defect identification of the aircraft component is improved, and the method is applied to crack defect identification of the aircraft component.

Description

Aircraft component defect online identification system based on visual detection
Technical Field
The invention relates to the technical field of image data processing, in particular to an on-line identification system for defects of an aircraft component based on visual detection.
Background
With the development of technology, visual inspection is becoming more and more widely used, for example, for defect identification of aircraft parts. Wherein, the visual inspection can be to replace human eyes with a machine to make measurement and judgment. Crack defects are relatively common aircraft component defects. Currently, when defect identification is performed on an aircraft component based on visual inspection, the following methods are generally adopted: based on the aircraft component image, a neural network is adopted to identify defects of the aircraft component.
However, when using neural networks for crack defect identification of aircraft components, the following technical problems often exist:
when training the neural network, a large number of images of the aircraft component marked with the crack defect area are often required, a large amount of time is required for collecting the images, and the time for training the neural network is also often long, so that the efficiency of identifying the crack defect of the aircraft component is often low.
Disclosure of Invention
The summary of the invention is provided to introduce a selection of concepts in a simplified form that are further described below in the detailed description. The summary of the invention is not intended to identify key features or essential features of the claimed subject matter, nor is it intended to be used to limit the scope of the claimed subject matter.
In order to solve the technical problem of low efficiency of identifying crack defects of an aircraft component, the invention provides an online identification system for the defects of the aircraft component based on visual detection.
The invention provides an aircraft component defect online identification system based on visual detection, which comprises the following steps:
the image acquisition module is used for acquiring a target space image corresponding to the aircraft component to be detected;
the color difference analysis module is used for carrying out color difference analysis processing on each pixel point in the target space image to obtain a color anomaly index corresponding to the pixel point;
the color contrast analysis module is used for carrying out color contrast analysis processing on each pixel point in the target space image in a preset number of preset directions according to the color anomaly indexes to obtain a color contrast index set corresponding to the pixel point;
the determining module is used for determining defect abnormality indexes corresponding to the pixel points according to the color contrast index set corresponding to each pixel point in the target space image;
and the defect identification module is used for identifying the crack defect of the aircraft component to be detected according to the defect abnormality index and the color abnormality index corresponding to the pixel points in the target space image.
Optionally, the target spatial image is a LAB image.
Optionally, performing color difference analysis processing on each pixel point in the target spatial image to obtain a color anomaly index corresponding to the pixel point, including:
respectively normalizing an L value, an A value and a B value which are included in the LAB value corresponding to the pixel point to obtain a first channel value, a second channel value and a third channel value corresponding to the pixel point;
determining the sum of the first channel value, the second channel value and the third channel value corresponding to the pixel point as an integral channel value corresponding to the pixel point;
the duty ratios of the first channel value, the second channel value and the third channel value corresponding to the pixel point in the whole channel value are respectively determined to be the first channel duty ratio, the second channel duty ratio and the third channel duty ratio corresponding to the pixel point;
determining color feature entropy corresponding to the pixel point according to the first channel duty ratio, the second channel duty ratio and the third channel duty ratio corresponding to the pixel point;
determining color feature entropy corresponding to each neighborhood pixel point in a preset neighborhood corresponding to the pixel point, and obtaining a first feature entropy set corresponding to the pixel point by taking the color feature entropy corresponding to each neighborhood pixel point as a first feature entropy corresponding to the neighborhood pixel point;
Determining the absolute value of the difference value between the color characteristic entropy corresponding to the pixel point and each first characteristic entropy in the first characteristic entropy set as a color difference index to obtain a color difference index set corresponding to the pixel point;
and determining color anomaly indexes corresponding to the pixel points according to the color difference index sets corresponding to the pixel points, wherein the color difference indexes in the color difference index sets are positively correlated with the color anomaly indexes.
Optionally, performing color contrast analysis processing on each pixel point in the target space image in a preset number of preset directions according to the color anomaly index to obtain a color contrast index set corresponding to the pixel point, where the color contrast index set includes:
grouping a preset number of preset directions to obtain a preset direction group set;
and determining the color contrast index corresponding to the pixel point in the preset direction group according to the color anomaly index and each preset direction group in the preset direction group set, and obtaining the color contrast index set corresponding to the pixel point.
Optionally, the determining, according to the color anomaly index and each preset direction group in the preset direction group set, a color contrast index corresponding to the pixel point in the preset direction group includes:
Screening out a reference pixel point group corresponding to each preset direction of the pixel points in the preset direction groups from the target space image to obtain a reference pixel point group set;
determining an abnormal representing index corresponding to each reference pixel point in each reference pixel point set in the reference pixel point set according to the color abnormal index corresponding to each reference pixel point in the reference pixel point set, and obtaining an abnormal representing index set, wherein the color abnormal index and the abnormal representing index are positively correlated;
and determining the ratio of the first representative index to the second representative index as the color comparison index corresponding to the pixel point in the preset direction group, wherein the first representative index is the largest abnormal representative index in the abnormal representative index set, and the second representative index is the smallest abnormal representative index in the abnormal representative index set.
Optionally, the determining, according to the color contrast index set corresponding to each pixel point in the target spatial image, a defect abnormality index corresponding to the pixel point includes:
determining a first defect index according to a color contrast index set corresponding to each window pixel point in a preset window corresponding to the pixel point, wherein the color contrast index and the first defect index are positively correlated;
Combining color contrast index sets corresponding to each window pixel point in a preset window into target feature coordinates corresponding to the window pixel points;
determining the Euclidean distance between the target feature coordinates corresponding to the pixel points and the target feature coordinates corresponding to each window pixel point in a preset window as a color fluctuation index corresponding to the window pixel points to obtain a color fluctuation index set;
screening out the maximum color fluctuation index from the color fluctuation index set, and taking the maximum color fluctuation index as a target fluctuation index;
and determining a defect abnormality index corresponding to the pixel point according to the first defect index and the target fluctuation index, wherein the first defect index and the target fluctuation index are positively correlated with the defect abnormality index.
Optionally, the identifying the crack defect of the aircraft component to be detected according to the defect abnormality index and the color abnormality index corresponding to the pixel point in the target space image includes:
determining a target defect index corresponding to each pixel point according to the defect abnormality index and the color abnormality index corresponding to each pixel point in the target space image, wherein the defect abnormality index and the color abnormality index are positively correlated with the target defect index;
And identifying crack defects of the aircraft component to be detected according to the target defect indexes.
Optionally, the identifying the crack defect of the aircraft component to be detected according to the target defect index includes:
according to the target defect index, carrying out region growth on pixel points in the target space image to obtain a target region set;
determining a target defect index corresponding to each pixel point in each target area in the target area set, and determining a defect representative index corresponding to the target area, wherein the target defect index and the defect representative index are positively correlated;
and when the defect representing index corresponding to the target region in the target region set is larger than a preset defect threshold value, judging the target region as a crack defect region.
The invention has the following beneficial effects:
according to the on-line identification system for the defects of the aircraft parts based on visual detection, the technical problem of low efficiency of identifying the crack defects of the aircraft parts is solved by carrying out data processing on the target space image, and the efficiency of identifying the crack defects of the aircraft parts is improved. Firstly, the target space image often contains surface information of the aircraft component to be detected, so that the image acquisition module is used for acquiring the target space image corresponding to the aircraft component to be detected, and the subsequent identification of crack defects of the aircraft component to be detected can be facilitated. Then, as the color may be changed when the aircraft component is cracked, the color difference analysis module is used for performing color difference analysis processing on each pixel point in the target space image, so that whether the pixel point is a cracked pixel point can be conveniently and subsequently judged. The crack defect pixel may be a pixel in which a crack defect has occurred. Then, color anomaly indexes and preset numbers of preset directions are comprehensively considered, and color contrast analysis processing is carried out on each pixel point in the target space image, so that the accuracy of determining the color contrast index set corresponding to the pixel point can be improved. And then, comprehensively considering the color contrast index set corresponding to each pixel point in the target space image, and improving the accuracy of determining the defect abnormality index corresponding to the pixel point. Finally, based on the defect abnormality index and the color abnormality index corresponding to the pixel points in the target space image, the crack defect identification of the aircraft component to be detected can be realized. Therefore, the invention quantifies a plurality of indexes related to the crack defects, realizes the crack defect identification of the aircraft component to be detected based on visual detection, improves the accuracy of crack defect identification of the aircraft component to be detected, and does not need training of a neural network, thereby improving the efficiency of crack defect identification of the aircraft component to be detected based on visual detection.
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 schematic diagram of an on-line identification system for defects of aircraft parts based on visual inspection according to the present invention;
fig. 2 is a schematic diagram of a plurality of preset directions of a pixel point according to the present invention.
Wherein, the reference numerals include: a first line 201, a second line 202, a third line 203, and a fourth line 204.
Detailed Description
In order to further describe the technical means and effects adopted by the present invention to achieve the preset purpose, the following detailed description is given below of the specific implementation, structure, features and effects of the technical solution according to the present invention with reference to the accompanying drawings and preferred embodiments. 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 an aircraft component defect online identification system based on visual detection, which comprises the following steps:
the image acquisition module is used for acquiring a target space image corresponding to the aircraft component to be detected;
the color difference analysis module is used for carrying out color difference analysis processing on each pixel point in the target space image to obtain a color anomaly index corresponding to the pixel point;
the color contrast analysis module is used for carrying out color contrast analysis processing on each pixel point in the target space image in a preset number of preset directions according to the color anomaly indexes to obtain a color contrast index set corresponding to the pixel point;
the determining module is used for determining defect abnormality indexes corresponding to the pixel points according to the color contrast index set corresponding to each pixel point in the target space image;
and the defect identification module is used for identifying the crack defect of the aircraft component to be detected according to the defect abnormality index and the color abnormality index corresponding to the pixel points in the target space image.
Referring to fig. 1, a schematic structural diagram of an aircraft component defect online identification system based on visual inspection according to the present invention is shown. The aircraft component defect online identification system based on visual detection comprises:
The image acquisition module 101 is used for acquiring a target space image corresponding to the aircraft component to be detected.
In some embodiments, a target aerial image corresponding to the aircraft component to be inspected may be acquired.
The object space image may be a LAB (color space) image. The L channel included in the LAB image may represent pixel brightness, and the range of values may be [0, 100]. The LAB image may include an a-channel that may represent the green to red component of the pixel, and may range in value from-127, 128. The LAB image may include B-channels representing blue to yellow components of the pixel, and may range in value from-127, 128. The aircraft component to be inspected may be an aircraft component to be subjected to crack defect detection. The color of a normal aircraft component (an aircraft component that is not subject to crack defects) may be solid. For example, the aircraft component to be inspected may be an aircraft skin. The target spatial image may be a LAB image of the aircraft component to be detected.
It should be noted that, because the target space image often includes the surface information of the aircraft component to be detected, the image acquisition module is used to acquire the target space image corresponding to the aircraft component to be detected, so that the subsequent identification of the crack defect of the aircraft component to be detected can be facilitated.
As an example, this step may include the steps of:
first, an initial surface image corresponding to the aircraft component to be inspected is acquired by a CCD (Charge Coupled Device, charge-coupled device) camera.
Wherein the initial surface image may be a surface image of the aircraft component to be inspected. The initial surface image may be an RGB (Red Green Blue) image.
It should be noted that, the initial surface image is acquired by the CCD camera, so that abnormal noise in the image acquisition process can be avoided to a certain extent, and thus adverse effects caused by the abnormal noise can be reduced.
And secondly, denoising the initial surface image to obtain a reference image.
The reference image may be an initial surface image after denoising.
For example, a gaussian filtering method may be used to denoise the initial surface image to obtain the reference image.
It should be noted that, the denoising processing is performed on the initial surface image, so that adverse effects of noise in the image acquisition environment on subsequent crack defect recognition can be reduced to a certain extent.
Third, converting the reference image into a LAB image, and determining the converted image as a target spatial image.
It should be noted that, because the LAB image often accords with the color perception feature of human eyes more than the RGB image, the initial surface image (RGB image) is converted into the target space image (LAB image), so that the target space image more accords with the color perception feature of human eyes can be obtained, and the subsequent processing can be facilitated.
The color difference analysis module 102 is configured to perform color difference analysis processing on each pixel point in the target spatial image, so as to obtain a color anomaly index corresponding to the pixel point.
In some embodiments, color difference analysis may be performed on each pixel in the target spatial image to obtain a color anomaly indicator corresponding to the pixel.
It should be noted that, because the color may be changed when the aircraft component has a crack defect, the color difference analysis module performs color difference analysis processing on each pixel point in the target space image, so that it is convenient to determine whether the pixel point is a crack defect pixel point or not. The crack defect pixel may be a pixel in which a crack defect has occurred.
As an example, this step may include the steps of:
the first step is to normalize the L value, the A value and the B value included in the LAB value corresponding to the pixel point to obtain a first channel value, a second channel value and a third channel value corresponding to the pixel point.
The LAB values corresponding to the pixel points may include: l value, a value, and B value. The L value may be an L channel value. The a value may be an a-channel value. The B value may be a B channel value. The first channel value may be a normalized L value. The second channel value may be a normalized value of a. The third channel value may be a normalized B value.
For example, determining the first channel value, the second channel value, and the third channel value corresponding to the pixel point may include the sub-steps of:
in the first substep, a range normalization algorithm may be adopted to normalize the L values included in the LAB values corresponding to each pixel point in the target spatial image, so as to obtain first channel values corresponding to each pixel point.
And a second sub-step, normalizing the A value included in the LAB value corresponding to each pixel point in the target space image by adopting a range normalization algorithm to obtain a second channel value corresponding to each pixel point.
And a third sub-step, normalizing the B value included in the LAB value corresponding to each pixel point in the target space image by adopting a range normalization algorithm to obtain a third channel value corresponding to each pixel point.
It should be noted that, normalization is performed on the L value, the a value and the B value included in the LAB value corresponding to the pixel point, so that the influence between different scales corresponding to the L value, the a value and the B value can be eliminated, and the subsequent processing can be facilitated.
And a second step of determining the sum of the first channel value, the second channel value and the third channel value corresponding to the pixel point as the whole channel value corresponding to the pixel point.
And thirdly, determining the duty ratio of the first channel value, the second channel value and the third channel value corresponding to the pixel point in the whole channel value as the first channel duty ratio, the second channel duty ratio and the third channel duty ratio corresponding to the pixel point.
Wherein the first channel duty cycle may be a duty cycle of the first channel value in the overall channel value. The second channel duty cycle may be a duty cycle of the second channel value in the overall channel value. The third channel duty cycle may be a duty cycle of the third channel value in the overall channel value.
For example, the formula for determining the first channel duty ratio, the second channel duty ratio, and the third channel duty ratio corresponding to the pixel point may be:
wherein, the liquid crystal display device comprises a liquid crystal display device,is the first channel duty ratio corresponding to the ith pixel point in the target space image.Is the second channel duty ratio corresponding to the ith pixel point in the target space image.Is the third channel duty ratio corresponding to the ith pixel point in the target space image.The value is a first channel value corresponding to the ith pixel point in the target space image, namely a value obtained by normalizing an L value included in the LAB value corresponding to the ith pixel point. The value is a second channel value corresponding to the ith pixel point in the target space image, namely a value obtained by normalizing the A value included in the LAB value corresponding to the ith pixel point.The third channel value corresponding to the ith pixel point in the target space image is the value obtained by normalizing the B value included in the LAB value corresponding to the ith pixel point.Is the integral channel value corresponding to the ith pixel point in the target space image. i is the orderAnd marking the serial numbers of the pixel points in the space image.
It should be noted that the number of the substrates,the duty cycle of the first channel value corresponding to the ith pixel point in the overall channel value may be characterized.The duty cycle of the second channel value corresponding to the ith pixel point in the overall channel value may be characterized.The duty cycle of the third channel value corresponding to the ith pixel point in the overall channel value can be characterized. If it is0, then can be set to、Andis 0.
And step four, determining color feature entropy corresponding to the pixel point according to the first channel duty ratio, the second channel duty ratio and the third channel duty ratio corresponding to the pixel point.
For example, the formula for determining the color feature entropy corresponding to the pixel point may be:
wherein, the liquid crystal display device comprises a liquid crystal display device,is the color characteristic entropy corresponding to the ith pixel point in the target space image. Is corresponding to the ith pixel point in the target space imageThe first channel duty cycle.Is the second channel duty ratio corresponding to the ith pixel point in the target space image.Is the third channel duty ratio corresponding to the ith pixel point in the target space image.Is based on 2Logarithmic (log).Is based on 2Logarithmic (log).Is based on 2Logarithmic (log). i is the sequence number of the pixel point in the target space image.
It should be noted that the number of the substrates,the distribution condition among the L channel value, the A channel value and the B channel value corresponding to the ith pixel point can be represented, so that the color distribution corresponding to the ith pixel point can be represented.
And fifthly, determining color feature entropy corresponding to each neighborhood pixel point in a preset neighborhood corresponding to the pixel point, and obtaining a first feature entropy set corresponding to the pixel point by taking the color feature entropy corresponding to the neighborhood pixel point as a first feature entropy corresponding to the neighborhood pixel point.
The preset neighborhood may be a preset neighborhood. The neighborhood pixel may be a pixel within a preset neighborhood. The first feature entropy set corresponding to the pixel point may include: color feature entropy corresponding to each neighborhood pixel point in the preset neighborhood corresponding to the pixel point.
For example, the color feature entropy corresponding to the neighboring pixel point may be determined by referring to the above manner of determining the color feature entropy corresponding to the pixel point, and the specific manner may be: the first step to the fourth step, which are implemented by the color difference analysis module 102 and included as examples, may be performed with the neighboring pixel point as a pixel point, and the obtained color feature entropy is the color feature entropy corresponding to the neighboring pixel point and is also the first feature entropy corresponding to the neighboring pixel point.
And sixthly, determining the absolute value of the difference value between the color characteristic entropy corresponding to the pixel point and each first characteristic entropy in the first characteristic entropy set as a color difference index, and obtaining a color difference index set corresponding to the pixel point.
The color difference index set corresponding to the pixel point may include: the absolute value of the difference value between the color characteristic entropy corresponding to the pixel point and each first characteristic entropy in the first characteristic entropy set.
And seventh, determining color anomaly indexes corresponding to the pixel points according to the color difference index sets corresponding to the pixel points.
Wherein, the color difference index in the color difference index set may be positively correlated with the color anomaly index.
For example, the formula for determining the color anomaly index corresponding to the pixel point may be:
wherein, the liquid crystal display device comprises a liquid crystal display device,is the color anomaly index corresponding to the ith pixel point in the target space image.Is the color characteristic entropy corresponding to the ith pixel point in the target space image.Is the first in the target space imageColor feature entropy corresponding to the r-th neighborhood pixel point in the preset neighborhood corresponding to the i pixel points, namely, the first feature entropy corresponding to the r-th neighborhood pixel point.Is the number of neighborhood pixels in the preset neighborhood corresponding to the ith pixel in the target space image. For example, if the preset neighborhood is eight neighbors, then May be 8.Is thatIs the absolute value of (c).And (3) withAnd shows positive correlation. i is the sequence number of the pixel point in the target space image. r is the sequence number of the neighbor pixel point in the preset neighbor corresponding to the ith pixel point.
It should be noted that, since the color of the normal aircraft component is often solid, the color distribution between the normal pixels is often relatively similar. The normal pixel may be a pixel where no crack defect occurs. Since the degree of crack defect occurring in each pixel point in the crack defect region tends to be different, the color distribution among the crack defect pixels tends to be different. When (when)The smaller the channel value between the ith pixel point and the r neighborhood pixel point, the more similar the channel value between the ith pixel point and the r neighborhood pixel point is, and the more similar the channel value distribution between the ith pixel point and the r neighborhood pixel point is, the more similar the color distribution between the ith pixel point and the r neighborhood pixel point is. Thus whenThe larger the pixel is, the more different the color distribution between the ith pixel and each neighboring pixel in the preset neighborhood is, and the more likely the ith pixel is a crack defect pixel is.
The color contrast analysis module 103 is configured to perform color contrast analysis processing on each pixel point in the target spatial image in a preset number of preset directions according to the color anomaly index, so as to obtain a color contrast index set corresponding to the pixel point.
In some embodiments, color contrast analysis processing may be performed on each pixel point in the target spatial image in a preset number of preset directions according to the color anomaly index, so as to obtain a color contrast index set corresponding to the pixel point.
The preset number may be a preset number. The preset number may be an even number. For example, the preset number may be 4. The preset direction may be a preset direction. The range of the angle corresponding to the preset direction may be [0 °,180 °), for example, the preset direction may be, but is not limited to: horizontal, vertical, 45 ° and 135 ° directions. The angle corresponding to the horizontal direction may be 0 °. The angle corresponding to the vertical direction may be 90 °. The 45 ° direction may correspond to an angle of 45 °. The 135 ° direction may correspond to an angle of 135 °. As shown in fig. 2, the solid black dots may represent a pixel, and the direction in which the first straight line 201 is located may be the horizontal direction of the pixel. The direction in which the second line 202 is located may be a 45 ° direction of the pixel point. The angle between the second line 202 and the first line 201 may be 45 °. The direction in which the third straight line 203 is located may be the vertical direction of the pixel point. The angle between the third line 203 and the first line 201 may be 90 °. The direction in which the fourth straight line 204 is located may be a 135 ° direction of the pixel point. The angle between the fourth line 204 and the first line 201 may be 135 °.
It should be noted that, color anomaly indexes and preset numbers of preset directions are comprehensively considered, and color contrast analysis processing is performed on each pixel point in the target space image, so that accuracy of determining a color contrast index set corresponding to the pixel point can be improved.
As an example, this step may include the steps of:
the first step, grouping a preset number of preset directions to obtain a preset direction group set.
For example, each two preset directions in the preset number of preset directions may be divided into one preset direction group, to obtain a preset direction group set.
As another example, if the preset number is 4, the 4 preset directions include: the horizontal direction, the vertical direction, the 45 ° direction, and the 135 ° direction may be divided into one preset direction group, and the 45 ° direction and the 135 ° direction may be divided into one preset direction group.
And secondly, determining the color contrast index corresponding to the pixel point in the preset direction group according to the color anomaly index and each preset direction group in the preset direction group set to obtain the color contrast index set corresponding to the pixel point.
The color contrast index set corresponding to the pixel point may include: the pixel points correspond to the color contrast indexes in each preset direction group. The color contrast indexes in the color contrast index set corresponding to the pixel points can be in one-to-one correspondence with the preset direction groups in the preset direction group set.
For example, according to the color anomaly index and each preset direction group in the preset direction group set, determining the color contrast index corresponding to the pixel point in the preset direction group, and obtaining the color contrast index set corresponding to the pixel point may include the following substeps:
and a first sub-step of screening out the reference pixel point groups corresponding to each preset direction in the preset direction groups from the target space image to obtain a reference pixel point group set.
The reference pixel point group corresponding to the pixel point in the preset direction may include: and filtering out the pixel points from the preset direction of the pixel points. The reference pixel point group set may include: the pixel points are in reference pixel point groups corresponding to all preset directions.
For example, all the pixels in the preset direction of the pixel may be formed into a reference pixel group corresponding to the pixel in the preset direction. If the preset direction is the horizontal direction, all the pixels in the horizontal direction of the pixel can be combined into a reference pixel group corresponding to the pixel in the horizontal direction, i.e. the row where the pixel is located can be used as the reference pixel group corresponding to the pixel in the horizontal direction.
For another example, a preset number of pixels closest to the pixel may be selected from the preset direction of the pixel, and the selected preset number of pixels may be formed into a reference pixel group corresponding to the pixel in the preset direction. Wherein the preset number may be a preset number. For example, the preset number may be 20.
And a second sub-step of determining an abnormality representative index corresponding to each reference pixel point group according to the color abnormality index corresponding to each reference pixel point in each reference pixel point group in the reference pixel point group set to obtain an abnormality representative index set.
Wherein the color anomaly index and the anomaly representative index are positively correlated.
And a third sub-step of determining the ratio of the first representative index to the second representative index as a color contrast index corresponding to the pixel point in the preset direction group.
The first representative index may be the largest abnormality representative index in the abnormality representative index set. The second representative index may be the smallest abnormality representative index in the abnormality representative index set.
For example, if the preset direction group includes two preset directions, the formula corresponding to the color contrast index corresponding to the pixel point in the preset direction group may be:
Wherein, the liquid crystal display device comprises a liquid crystal display device,is that the ith pixel point in the target space image is in a preset direction groupColor contrast indexes corresponding to the j-th preset direction group in the set. The j-th preset direction group may include two preset directions.Is an abnormal representative index corresponding to a reference pixel point group corresponding to the 1 st preset direction included in the j preset direction group of the ith pixel point in the target space image.Is an abnormal representative index corresponding to a reference pixel point group corresponding to the 2 nd preset direction included in the j preset direction group of the ith pixel point in the target space image.Is a factor greater than 0, which is preset, mainly to preventThe denominator is 0. As an example of the presence of a metal such as,may be 0.01.The color anomaly index corresponding to the t-th reference pixel point in the reference pixel point group corresponding to the 1 st preset direction included in the j-th preset direction group of the ith pixel point in the target space image.The color anomaly index corresponding to the a-th reference pixel point in the reference pixel point group corresponding to the 2 nd preset direction included in the j-th preset direction group of the ith pixel point in the target space image.Is the number of reference pixel points in the reference pixel point group corresponding to the 1 st preset direction included in the j preset direction group of the ith pixel point in the target space image. Is the number of reference pixel points in the reference pixel point group corresponding to the 2 nd preset direction included in the j preset direction group of the ith pixel point in the target space image.Is thatAndand the maximum value of the index is also the first representative index corresponding to the ith pixel point in the target space image in the jth preset direction group.Is thatAndand the minimum value of the (b) is also a second representative index corresponding to the ith pixel point in the target space image in the jth preset direction group.And (3) withAnd shows positive correlation.And (3) withAnd shows positive correlation. i is the sequence number of the pixel point in the target space image. t is the serial number of the reference pixel point in the reference pixel point group corresponding to the 1 st preset direction included in the j preset direction group. a is the serial number of the reference pixel point in the reference pixel point group corresponding to the 2 nd preset direction included in the j preset direction group of the ith pixel point.
It should be noted that, if the pixel points in the reference pixel point group corresponding to the two preset directions included in the jth preset direction group are normalPixel point, thenAndthe corresponding values tend to be relatively similar, resulting inAndcorresponding values are relatively similar, resulting inOften close to 1. If the reference pixel point group corresponding to the two preset directions included in the jth preset direction group has the crack defect pixel points, color anomaly indexes are often different, thereby causing Often greater than 1. Thus whenThe larger the reference pixel point group is, the more likely the crack defect pixel point exists in the reference pixel point group corresponding to the two preset directions included in the jth preset direction group. Since the crack defect pixel points are not isolated, when the ith pixel point is more likely to exist in the reference pixel point group corresponding to the two preset directions included in the jth preset direction group, the ith pixel point is more likely to be the crack defect pixel point. Thus whenThe larger the i-th pixel, the more likely it is a crack defect pixel.
The determining module 104 is configured to determine a defect anomaly index corresponding to each pixel point according to the color contrast index set corresponding to each pixel point in the target spatial image.
In some embodiments, the defect anomaly index corresponding to each pixel point in the target aerial image may be determined according to a set of color contrast indexes corresponding to each pixel point.
It should be noted that, by comprehensively considering the color contrast index set corresponding to each pixel point in the target space image, the accuracy of determining the defect abnormality index corresponding to the pixel point can be improved.
As an example, this step may include the steps of:
The first step, determining a first defect index according to a color contrast index set corresponding to each window pixel point in a preset window corresponding to the pixel point.
The color contrast index may be positively correlated with the first defect index. The preset window may be a preset window. For example, the preset window may be a 3×3 window. The window pixel points may be pixel points in a preset window. The pixel point may be located at a center position of a preset window corresponding to the pixel point.
And secondly, combining the color contrast index set corresponding to each window pixel point in the preset window into the target feature coordinates corresponding to the window pixel points.
For example, if the set of color contrast indicators includes: the first color contrast index and the second color contrast index may be used as an abscissa included in the target feature coordinates, and the second color contrast index may be used as an ordinate included in the target feature coordinates, where the obtained target feature coordinates may be (first color contrast index, second color contrast index).
And thirdly, determining the Euclidean distance between the target feature coordinates corresponding to the pixel points and the target feature coordinates corresponding to each window pixel point in a preset window as a color fluctuation index corresponding to the window pixel points, and obtaining a color fluctuation index set.
The color fluctuation index set corresponding to the pixel point may include: color fluctuation indexes corresponding to all window pixel points in a preset window corresponding to the pixel points.
And fourthly, screening out the maximum color fluctuation index from the color fluctuation index set, and taking the maximum color fluctuation index as a target fluctuation index.
For example, the maximum color fluctuation index may be selected from the color fluctuation index set corresponding to the pixel point, and used as the target fluctuation index corresponding to the pixel point.
It should be noted that, if the window pixel points in the preset window are normal pixel points, the target feature coordinates corresponding to the window pixel points are often similar, and the color fluctuation index in the obtained color fluctuation index set is often smaller. Therefore, the larger the target fluctuation index corresponding to the pixel point is, the more likely the pixel point is a crack defect pixel point.
Fifthly, determining defect abnormality indexes corresponding to the pixel points according to the first defect indexes and the target fluctuation indexes.
Wherein, the first defect index and the target fluctuation index can be positively correlated with the defect abnormality index.
For example, the formula for determining the defect abnormality index corresponding to the pixel point may be:
Wherein, the liquid crystal display device comprises a liquid crystal display device,is a defect abnormality index corresponding to the ith pixel point in the target space image.Is a first defect index corresponding to the ith pixel point in the target space image.Is the mth color contrast index in the color contrast index set corresponding to the pixel point of the x window in the preset window corresponding to the ith pixel point in the target space image. N is the number of window pixels in the preset window. For example, if the preset window is a 3×3 window, N is 9.Is the preset corresponding to the ith pixel pointThe number of color contrast indexes in the color contrast index set corresponding to the x-th window pixel point in the window.Is a target fluctuation index corresponding to the ith pixel point in the target space image. i is the sequence number of the pixel point in the target space image. And x is the serial number of the pixel point of the window in the preset window corresponding to the ith pixel point. m is the serial number of the color contrast index in the color contrast index set corresponding to the x-th window pixel point.Andare all in contact withAnd shows positive correlation.And (3) withAnd shows positive correlation.
It should be noted that the number of the substrates,the larger it is often stated that the x-th window pixel is more likely to be a crack defect pixel. When (when)And when the pixel is larger, the window pixel in the preset window corresponding to the ith pixel is more likely to be a crack defect pixel. Since the crack defect pixel points are not isolated, when the window pixel point in the preset window corresponding to the ith pixel point is more likely to be the crack defect pixel point, the ith pixel point is more likely to be the crack defect pixel point. Thus when The larger the i-th pixel, the more likely it is a crack defect pixel.The larger it tends to indicate that the i-th pixel is more likely to be a crack defect pixel. Thus (2)The larger it tends to indicate that the i-th pixel is more likely to be a crack defect pixel.
The defect identifying module 105 is configured to determine whether the aircraft component to be detected has a crack defect according to a defect abnormality index and a color abnormality index corresponding to the pixel points in the target space image.
In some embodiments, whether the aircraft component to be detected has a crack defect may be determined according to a defect abnormality index and a color abnormality index corresponding to the pixel point in the target spatial image.
It should be noted that, based on the defect abnormality index and the color abnormality index corresponding to the pixel points in the target space image, the crack defect identification of the aircraft component to be detected can be realized. Therefore, the invention quantifies a plurality of indexes related to the crack defects, realizes the crack defect identification of the aircraft component to be detected based on visual detection, improves the accuracy of crack defect identification of the aircraft component to be detected, and does not need training of a neural network, thereby improving the efficiency of crack defect identification of the aircraft component to be detected.
As an example, this step may include the steps of:
the first step, determining a target defect index corresponding to each pixel point according to the defect abnormality index and the color abnormality index corresponding to each pixel point in the target space image.
Wherein, the defect abnormal index and the color abnormal index can be positively correlated with the target defect index.
For example, the formula for determining the target defect index corresponding to the pixel point may be:
wherein, the liquid crystal display device comprises a liquid crystal display device,is a target defect index corresponding to the ith pixel point in the target space image.Is a defect abnormality index corresponding to the ith pixel point in the target space image.Is the color anomaly index corresponding to the ith pixel point in the target space image. i is the sequence number of the pixel point in the target space image.Andare all in contact withAnd shows positive correlation.Is thatIs included in the above formula (c).
And secondly, identifying crack defects of the aircraft component to be detected according to the target defect indexes.
It should be noted that the number of the substrates,the larger it tends to indicate that the i-th pixel is more likely to be a crack defect pixel. When (when)The larger the pixel is, the more different the color distribution between the ith pixel and each neighboring pixel in the preset neighborhood is, and the more likely the ith pixel is a crack defect pixel is. Thus (2) The largerWhen this is the case, it is often explained that the i-th pixel is more likely to be a crack defect pixel. And is also provided withRealize the alignment ofCan facilitate subsequent processing.
For example, according to the target defect index, determining whether the aircraft component to be detected has a crack defect may include the following sub-steps:
and a first sub-step, carrying out region growth on the pixel points in the target space image according to the target defect index to obtain a target region set.
The target region in the target region set may be a region obtained after region growth.
For example, a pixel point with the largest target defect index in the target space image can be used as an initial seed pixel point, the growth threshold can be set to be 0.2, and the pixel points in the target space image are subjected to region growth according to the target defect indexes corresponding to the pixel points in the target space image, so that a target region set is obtained.
And a second sub-step of determining a defect representative index corresponding to the target region by using target defect indexes corresponding to each pixel point in each target region in the target region set.
Wherein the target defect indicator may be positively correlated with the defect representative indicator.
For example, the average value of the target defect indexes corresponding to all the pixel points in the target area can be determined as the defect representative index corresponding to the target area, and the target defect indexes and the defect representative index are positively correlated.
And a third sub-step of judging the target area as a crack defect area when the defect representing index corresponding to the target area in the target area set is larger than a preset defect threshold.
The crack defect region may be a region where a crack defect exists. The preset defect threshold may be a maximum defect representing index set when the target area is considered to be a normal area, which is set in advance. For example, the preset defect threshold may be 0.7. The normal region may be a region where a crack defect does not occur.
When the target area set includes a crack defect area, it may be determined that the aircraft component to be inspected includes a crack defect. And when the target area set does not have the crack defect area, judging that the aircraft component to be detected does not have the crack defect.
Optionally, determining whether the aircraft component to be detected has a crack defect according to the target defect index may include the following substeps:
And a first sub-step, when a target defect index corresponding to the pixel point in the target space image is larger than a preset pixel defect threshold value, judging the pixel point as a crack defect pixel point.
The preset pixel defect threshold may be a preset maximum target defect index when the pixel point is considered to be a normal pixel point. For example, the preset pixel defect threshold may be 0.7. The normal pixel may be a pixel where a crack defect does not occur.
The crack defect pixel may be a pixel in which a crack defect has occurred.
And a second sub-step, when the crack defect pixel points exist in the target space image, judging that the aircraft component to be detected has crack defects.
And a third sub-step, when no crack defect pixel point exists in the target space image, judging that the aircraft component to be detected does not have a crack defect.
In conclusion, the target space image corresponding to the aircraft component to be detected is firstly obtained, so that crack defect recognition can be conveniently carried out on the aircraft component to be detected subsequently. Then, because color change may be caused when the aircraft component generates crack defect, color difference analysis processing is performed on each pixel point in the target space image, so that whether the pixel point is a crack defect pixel point can be conveniently judged subsequently. Then, color anomaly indexes and preset numbers of preset directions are comprehensively considered, and color contrast analysis processing is carried out on each pixel point in the target space image, so that the accuracy of determining the color contrast index set corresponding to the pixel point can be improved. And then, comprehensively considering the color contrast index set corresponding to each pixel point in the target space image, and improving the accuracy of determining the defect abnormality index corresponding to the pixel point. Finally, based on the defect abnormal index and the color abnormal index corresponding to the pixel points in the target space image, the identification of the crack defects of the aircraft component based on visual detection can be realized, and the training of a neural network is not needed, so that the efficiency of identifying the crack defects of the aircraft component to be detected can be improved.
The above embodiments are only for illustrating the technical solution of the present invention, and are not limiting; although the invention has been described in detail with reference to the foregoing embodiments, it will be understood by those of ordinary skill in the art that: the technical scheme described in the foregoing embodiments can be modified or some technical features thereof can be replaced by equivalents; such modifications and substitutions do not depart from the spirit of the invention and are intended to be included within the scope of the invention.

Claims (5)

1. An on-line identification system for defects of aircraft components based on visual inspection, the system comprising:
the image acquisition module is used for acquiring a target space image corresponding to the aircraft component to be detected;
the color difference analysis module is used for carrying out color difference analysis processing on each pixel point in the target space image to obtain a color anomaly index corresponding to the pixel point;
the color contrast analysis module is used for carrying out color contrast analysis processing on each pixel point in the target space image in a preset number of preset directions according to the color anomaly indexes to obtain a color contrast index set corresponding to the pixel point;
The determining module is used for determining defect abnormality indexes corresponding to the pixel points according to the color contrast index set corresponding to each pixel point in the target space image;
the defect identification module is used for identifying the crack defect of the aircraft component to be detected according to the defect abnormality index and the color abnormality index corresponding to the pixel points in the target space image;
the determining, according to the color contrast index set corresponding to each pixel point in the target spatial image, a defect abnormality index corresponding to the pixel point includes:
determining a first defect index according to a color contrast index set corresponding to each window pixel point in a preset window corresponding to the pixel point, wherein the color contrast index and the first defect index are positively correlated;
combining color contrast index sets corresponding to each window pixel point in a preset window into target feature coordinates corresponding to the window pixel points;
determining the Euclidean distance between the target feature coordinates corresponding to the pixel points and the target feature coordinates corresponding to each window pixel point in a preset window as a color fluctuation index corresponding to the window pixel points to obtain a color fluctuation index set;
Screening out the maximum color fluctuation index from the color fluctuation index set, and taking the maximum color fluctuation index as a target fluctuation index;
determining a defect abnormality index corresponding to the pixel point according to the first defect index and the target fluctuation index, wherein the first defect index and the target fluctuation index are positively correlated with the defect abnormality index;
the target space image is an LAB image;
performing color difference analysis processing on each pixel point in the target space image to obtain a color anomaly index corresponding to the pixel point, wherein the color anomaly index comprises:
respectively normalizing an L value, an A value and a B value which are included in the LAB value corresponding to the pixel point to obtain a first channel value, a second channel value and a third channel value corresponding to the pixel point;
determining the sum of the first channel value, the second channel value and the third channel value corresponding to the pixel point as an integral channel value corresponding to the pixel point;
the duty ratios of the first channel value, the second channel value and the third channel value corresponding to the pixel point in the whole channel value are respectively determined to be the first channel duty ratio, the second channel duty ratio and the third channel duty ratio corresponding to the pixel point;
determining color feature entropy corresponding to the pixel point according to the first channel duty ratio, the second channel duty ratio and the third channel duty ratio corresponding to the pixel point;
Determining color feature entropy corresponding to each neighborhood pixel point in a preset neighborhood corresponding to the pixel point, and obtaining a first feature entropy set corresponding to the pixel point by taking the color feature entropy corresponding to each neighborhood pixel point as a first feature entropy corresponding to the neighborhood pixel point;
determining the absolute value of the difference value between the color characteristic entropy corresponding to the pixel point and each first characteristic entropy in the first characteristic entropy set as a color difference index to obtain a color difference index set corresponding to the pixel point;
and determining color anomaly indexes corresponding to the pixel points according to the color difference index sets corresponding to the pixel points, wherein the color difference indexes in the color difference index sets are positively correlated with the color anomaly indexes.
2. The on-line identification system for defects of aircraft parts based on visual inspection according to claim 1, wherein the performing color contrast analysis processing on each pixel point in the target space image in a preset number of preset directions according to color anomaly indexes to obtain a color contrast index set corresponding to the pixel point comprises:
grouping a preset number of preset directions to obtain a preset direction group set;
and determining the color contrast index corresponding to the pixel point in the preset direction group according to the color anomaly index and each preset direction group in the preset direction group set, and obtaining the color contrast index set corresponding to the pixel point.
3. The system for on-line identification of defects of aircraft parts based on visual inspection according to claim 2, wherein the determining the color contrast index of the pixel point corresponding to the preset direction group according to the color anomaly index and each preset direction group in the preset direction group set comprises:
screening out a reference pixel point group corresponding to each preset direction of the pixel points in the preset direction groups from the target space image to obtain a reference pixel point group set;
determining an abnormal representing index corresponding to each reference pixel point in each reference pixel point set in the reference pixel point set according to the color abnormal index corresponding to each reference pixel point in the reference pixel point set, and obtaining an abnormal representing index set, wherein the color abnormal index and the abnormal representing index are positively correlated;
and determining the ratio of the first representative index to the second representative index as the color comparison index corresponding to the pixel point in the preset direction group, wherein the first representative index is the largest abnormal representative index in the abnormal representative index set, and the second representative index is the smallest abnormal representative index in the abnormal representative index set.
4. The online identification system for defects of an aircraft component based on visual inspection according to claim 1, wherein the identifying the crack defects of the aircraft component to be inspected according to the defect abnormality index and the color abnormality index corresponding to the pixel points in the target space image comprises:
determining a target defect index corresponding to each pixel point according to the defect abnormality index and the color abnormality index corresponding to each pixel point in the target space image, wherein the defect abnormality index and the color abnormality index are positively correlated with the target defect index;
and identifying crack defects of the aircraft component to be detected according to the target defect indexes.
5. The system for on-line identification of defects of aircraft parts based on visual inspection according to claim 4, wherein the identifying of crack defects of the aircraft parts to be inspected according to the target defect index comprises:
according to the target defect index, carrying out region growth on pixel points in the target space image to obtain a target region set;
determining a target defect index corresponding to each pixel point in each target area in the target area set, and determining a defect representative index corresponding to the target area, wherein the target defect index and the defect representative index are positively correlated;
And when the defect representing index corresponding to the target region in the target region set is larger than a preset defect threshold value, judging the target region as a crack defect region.
CN202310695760.8A 2023-06-13 2023-06-13 Aircraft component defect online identification system based on visual detection Active CN116433665B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202310695760.8A CN116433665B (en) 2023-06-13 2023-06-13 Aircraft component defect online identification system based on visual detection

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202310695760.8A CN116433665B (en) 2023-06-13 2023-06-13 Aircraft component defect online identification system based on visual detection

Publications (2)

Publication Number Publication Date
CN116433665A CN116433665A (en) 2023-07-14
CN116433665B true CN116433665B (en) 2023-08-22

Family

ID=87080091

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202310695760.8A Active CN116433665B (en) 2023-06-13 2023-06-13 Aircraft component defect online identification system based on visual detection

Country Status (1)

Country Link
CN (1) CN116433665B (en)

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN107052086A (en) * 2017-06-01 2017-08-18 扬州苏星机器人科技有限公司 Stamping parts surface defect detection apparatus and detection method based on 3D vision
CN111460931A (en) * 2020-03-17 2020-07-28 华南理工大学 Face spoofing detection method and system based on color channel difference image characteristics
CN114359275A (en) * 2022-03-16 2022-04-15 南通俊朗智能科技有限公司 Hydraulic gear pump defect detection method and system based on artificial intelligence
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

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN107052086A (en) * 2017-06-01 2017-08-18 扬州苏星机器人科技有限公司 Stamping parts surface defect detection apparatus and detection method based on 3D vision
CN111460931A (en) * 2020-03-17 2020-07-28 华南理工大学 Face spoofing detection method and system based on color channel difference image characteristics
CN114359275A (en) * 2022-03-16 2022-04-15 南通俊朗智能科技有限公司 Hydraulic gear pump defect detection method and system based on artificial intelligence
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

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
"Surface Defect Detection Methods for Industrial Products A Review";Yajun Chen etc.;《applied sciences》;全文 *

Also Published As

Publication number Publication date
CN116433665A (en) 2023-07-14

Similar Documents

Publication Publication Date Title
CN114937055B (en) Image self-adaptive segmentation method and system based on artificial intelligence
CN111563896B (en) Image processing method for detecting abnormality of overhead line system
CN116721107B (en) Intelligent monitoring system for cable production quality
CN112200160A (en) Deep learning-based direct-reading water meter reading identification method
CN116433669B (en) Machine vision-based quality detection method for weld joints of steel frame of anti-seismic structure
CN115526889B (en) Nondestructive testing method of boiler pressure pipeline based on image processing
CN112258490A (en) Low-emissivity coating intelligent damage detection method based on optical and infrared image fusion
CN111199194A (en) Automobile intelligent cabin instrument testing method based on machine vision and deep learning
CN117455917B (en) Establishment of false alarm library of etched lead frame and false alarm on-line judging and screening method
CN113487563B (en) EL image-based self-adaptive detection method for hidden cracks of photovoltaic module
CN112750113B (en) Glass bottle defect detection method and device based on deep learning and linear detection
CN116630332B (en) PVC plastic pipe orifice defect detection method based on image processing
CN116433665B (en) Aircraft component defect online identification system based on visual detection
CN114820597B (en) Smelting product defect detection method, device and system based on artificial intelligence
CN114998346B (en) Waterproof cloth quality data processing and identifying method
CN114155246B (en) Deformable convolution-based power transmission tower pin defect detection method
CN108830834B (en) Automatic extraction method for video defect information of cable climbing robot
CN115954291A (en) Crack monitoring system of TOPCon structure photovoltaic cell panel
CN114511851B (en) Hairspring algae cell statistical method based on microscope image
CN115546141A (en) Small sample Mini LED defect detection method and system based on multi-dimensional measurement
CN114612333A (en) Microscopic image-based microencapsulated algae cell statistical method
CN113947563A (en) Cable process quality dynamic defect detection method based on deep learning
CN117351021B (en) Intelligent detection method for production quality of photovoltaic connecting wire
CN117197112B (en) Industrial screen printing defect detection method and system based on sample comparison
CN117152687B (en) Communication line state monitoring system

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