CN116309608B - Coating defect detection method using ultrasonic image - Google Patents
Coating defect detection method using ultrasonic image Download PDFInfo
- Publication number
- CN116309608B CN116309608B CN202310594608.0A CN202310594608A CN116309608B CN 116309608 B CN116309608 B CN 116309608B CN 202310594608 A CN202310594608 A CN 202310594608A CN 116309608 B CN116309608 B CN 116309608B
- Authority
- CN
- China
- Prior art keywords
- value
- column
- image
- internal image
- defect
- 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
- 230000007547 defect Effects 0.000 title claims abstract description 124
- 238000000576 coating method Methods 0.000 title claims abstract description 50
- 239000011248 coating agent Substances 0.000 title claims abstract description 47
- 238000001514 detection method Methods 0.000 title claims abstract description 17
- 238000009825 accumulation Methods 0.000 claims abstract description 102
- 238000009826 distribution Methods 0.000 claims abstract description 31
- 230000008859 change Effects 0.000 claims abstract description 20
- 238000012545 processing Methods 0.000 claims abstract description 9
- 238000000034 method Methods 0.000 claims description 30
- 238000005520 cutting process Methods 0.000 claims description 28
- 238000013507 mapping Methods 0.000 claims description 7
- 238000002604 ultrasonography Methods 0.000 claims description 4
- 238000012163 sequencing technique Methods 0.000 claims description 3
- 238000012935 Averaging Methods 0.000 claims description 2
- 239000012535 impurity Substances 0.000 description 15
- 239000010410 layer Substances 0.000 description 13
- 238000003384 imaging method Methods 0.000 description 7
- 239000002356 single layer Substances 0.000 description 7
- 239000006185 dispersion Substances 0.000 description 6
- 230000002950 deficient Effects 0.000 description 5
- 238000004458 analytical method Methods 0.000 description 4
- 239000000523 sample Substances 0.000 description 4
- 230000000903 blocking effect Effects 0.000 description 3
- 238000004364 calculation method Methods 0.000 description 3
- 239000011247 coating layer Substances 0.000 description 3
- 230000000694 effects Effects 0.000 description 3
- 238000005516 engineering process Methods 0.000 description 2
- 238000009776 industrial production Methods 0.000 description 2
- 238000007781 pre-processing Methods 0.000 description 2
- 230000008569 process Effects 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000007405 data analysis Methods 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 238000011156 evaluation Methods 0.000 description 1
- 239000006115 industrial coating Substances 0.000 description 1
- 239000000463 material Substances 0.000 description 1
- 239000002184 metal Substances 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000000750 progressive effect Effects 0.000 description 1
- 230000008961 swelling Effects 0.000 description 1
- 239000004753 textile Substances 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/0002—Inspection of images, e.g. flaw detection
- G06T7/0004—Industrial image inspection
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/60—Analysis of geometric attributes
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/10—Image acquisition modality
- G06T2207/10132—Ultrasound image
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02P—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN THE PRODUCTION OR PROCESSING OF GOODS
- Y02P90/00—Enabling technologies with a potential contribution to greenhouse gas [GHG] emissions mitigation
- Y02P90/30—Computing systems specially adapted for manufacturing
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Quality & Reliability (AREA)
- Geometry (AREA)
- Investigating Or Analyzing Materials By The Use Of Ultrasonic Waves (AREA)
Abstract
The invention relates to the technical field of data processing, in particular to a coating defect detection method using an ultrasonic image, which comprises the steps of collecting an internal image of a coating, and carrying out reverse processing on the internal image to obtain a corresponding reverse gray image; constructing an accumulation sequence curve and a column accumulation sequence curve; obtaining a defect area in the reversed gray level image according to the row accumulation sequence curve and the column accumulation sequence curve; acquiring an internal image set, and constructing a gray level distribution curve of a defect middle line in a target internal image to calculate a peak distance discrete value and a symmetric value of the defect middle line; and calculating a fault ultrasonic attenuation discrete parameter and an average gray level change rate of the internal image of the target to obtain a defect type index, and determining the defect type of the coating according to the defect type index. The invention improves the efficiency of defect identification under the coating.
Description
Technical Field
The invention relates to the technical field of data processing, in particular to a coating defect detection method applying an ultrasonic image.
Background
The current industrial coating technology is widely applied in various fields of electronics, aviation, automobiles, medical treatment, textile and the like, and has a huge share in the international market. Correspondingly, the detection flow after the coating process in the industrial production plays a very important role in the whole industrial production and the product quality guarantee.
Some defects can not be avoided in the coating process, wherein in the coating process of the surface, the most common is the bulge problem, the bulge can be generated on the surface of the coating due to two specific defects, namely impurities and bubbles, and the accuracy of the traditional defect detection method for identifying the defects is low due to the similarity of the two defects, namely the traditional defect detection method based on machine vision of the surface texture features of the defects has high false detection rate on the impurities and the bubbles in the bulge of the coating.
Disclosure of Invention
In order to solve the technical problems, the invention aims to provide a coating defect detection method applying an ultrasonic image, and the adopted technical scheme is as follows:
acquiring an internal image of the coating, and acquiring the internal image; performing inversion processing on the internal image to obtain a corresponding inverted gray level image; respectively constructing a row accumulation sequence curve and a column accumulation sequence curve according to the row accumulation sequence of each row and the column accumulation sequence of each column in the reversed-phase gray level image;
respectively counting the wave crest number and the wave trough number in the row accumulation sequence curve and the column accumulation sequence curve; respectively acquiring a row range end cutting average value of a row accumulation sequence curve and a column range end cutting average value of a column accumulation sequence curve based on the number of wave crests and the number of wave troughs; taking the numerical values of the line range end-cutting average value and the column range end-cutting average value as standard lines, and respectively calculating the duty ratio of the sectional line accumulation sequence curve and the sectional column accumulation sequence curve in the whole accumulation sequence curve to which the sectional line accumulation sequence curve belongs for the sectional line accumulation sequence curve corresponding to the numerical value larger than the line range end-cutting average value and the sectional column accumulation sequence curve corresponding to the numerical value larger than the column range end-cutting average value; acquiring a defect area in the reverse gray image based on the duty ratio;
sequencing the internal images corresponding to the defect area from near to far by adjusting the acquisition distance of the internal images to obtain a corresponding internal image set; taking the last internal image in the internal image set as a target internal image, constructing a gray level distribution curve of a defect middle line in the target internal image according to the gray level value, and obtaining a peak point of the gray level distribution curve to calculate a symmetrical value; calculating a peak distance discrete value of the defect middle row based on a column coordinate sequence of the peak points;
acquiring a peak distance discrete value corresponding to each internal image in the internal image set, and calculating a fault ultrasonic attenuation discrete parameter of the target internal image by combining the symmetrical value and all the peak distance discrete values; calculating the average gray scale change rate of the internal image of the target; taking the product of the fault ultrasonic attenuation discrete parameter and the average gray level change rate as a defect type index of the target internal image, and confirming the defect type of the coating according to the defect type index.
Further, the row accumulation sequence refers to the gray value sum of all the pixels of the corresponding row, and the column accumulation sequence refers to the gray value sum of all the pixels of the corresponding column.
Further, the method for obtaining the line range end-cutting average value of the line accumulation sequence curve comprises the following steps:
wherein ,the line range end cutting average value; />The number of peaks; />Is a preset value; />The sequence is accumulated for row a.
Further, the method for obtaining the column-range tail-end average value of the column accumulation sequence curve comprises the following steps:
wherein ,the column range end average value; />Is the number of wave troughs; />Is a preset value; />The sequence is accumulated for column b.
Further, the method for acquiring the defect area in the reverse gray scale image based on the duty ratio comprises the following steps:
when the duty ratio is larger than a preset duty ratio threshold value, mapping an area surrounded by the corresponding segmented row accumulation sequence curve and the segmented column accumulation sequence curve in the reverse gray level image as a defect area.
Further, the method for obtaining the peak point of the gray level distribution curve to calculate the symmetry value includes:
and obtaining peak points and maximum points of the gray level distribution curve, taking the column coordinates of the maximum points as symmetry axes, calculating absolute differences of corresponding gray levels between a first peak point on the left side of the symmetry axes and a last peak point on the right side of the symmetry axes, then calculating absolute differences of corresponding gray levels between a second peak point on the left side of the symmetry axes and a last-to-last peak point on the right side of the symmetry axes, and so on, and taking the result of accumulating, summing and averaging all the calculated absolute differences as symmetry values.
Further, the method for obtaining the discrete value of the peak distance of the defect middle row based on the column coordinate sequence of the peak point comprises the following steps:
and forming a column coordinate sequence by column coordinates of all the wave peak points, calculating a backward differential sequence of the column coordinate sequence, obtaining average distances among the wave peak points, calculating a difference square between each element in the backward differential sequence and the average distances, and calculating a mean value of the difference squares as a peak distance discrete value.
Further, the method for calculating the fault ultrasonic attenuation discrete parameter of the internal image of the target by combining the symmetry value and all the peak distance discrete values comprises the following steps:
calculating the average value of all the peak distance discrete values, obtaining a first difference square between each peak distance discrete value and the average value, obtaining a first difference square sum, taking the first difference square sum as a denominator, taking the product of the symmetrical value and the number of images in the internal image set as the denominator to obtain a corresponding ratio, and taking the result of the secondary evolution of the ratio as a fault ultrasonic attenuation discrete parameter of the target internal image.
Further, the method for calculating the average gray scale change rate of the internal image of the target comprises the following steps:
calculating a first average gray value of the target internal image and a second average gray value of the first internal image in the internal image set; obtaining a difference value between the first average gray value and the second average gray value, obtaining a corresponding ratio by taking the difference value as a numerator and taking the difference value of the image value between the target internal image and the first internal image as a denominator, and taking the ratio as the average gray change rate of the target internal image; the image value refers to an A value of an A-th internal image in the internal image set, and A is a positive integer.
The invention has the following beneficial effects: the invention can accurately and rapidly identify the defects under the coating by utilizing the symmetry and the discreteness of the gray peak on the single line of the defect part in the ultrasonic imaging by using the impurities and the bubbles under the defect of the coating, and can provide parameter reference for the defect degree under the coating according to the number of the obtained layers and the positioned defect size.
Drawings
In order to more clearly illustrate the embodiments of the invention or the technical solutions and advantages of the prior art, the following description will briefly explain the drawings used in the embodiments or the description of the prior art, and it is obvious that the drawings in the following description are only some embodiments of the invention, and other drawings can be obtained according to the drawings without inventive effort for a person skilled in the art.
FIG. 1 is a flow chart of steps of a method for detecting defects of a coating using an ultrasonic image according to an embodiment of the present invention;
fig. 2 is a schematic diagram of a correspondence relationship between a row accumulation sequence curve and a column accumulation sequence curve and an inverse gray scale image according to an embodiment of the present invention.
Detailed Description
In order to further describe the technical means and effects adopted by the present invention to achieve the preset purposes, the following detailed description refers to specific embodiments, structures, features and effects of a coating defect detection method using ultrasonic images 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 aims at the following scenes: the method combines the imaging characteristics of an ultrasonic C-scan imaging system (C ultrasonic) and the line gray scale characteristics of images to accurately detect and identify the defect type under the coating.
The following specifically describes a specific scheme of a coating defect detection method using an ultrasonic image provided by the invention with reference to the accompanying drawings.
Referring to fig. 1, a flowchart of a method for detecting a coating defect using an ultrasonic image according to an embodiment of the invention is shown, where the method includes:
s001, collecting an internal image of the coating; performing inversion processing on the internal image to obtain a corresponding inverted gray level image; and respectively constructing a row accumulation sequence curve and a column accumulation sequence curve according to the row accumulation sequence of each row and the column accumulation sequence of each column in the reversed-phase gray level image.
Specifically, an ultrasonic instrument is utilized to scan the surface of the coating, an internal image is obtained, the internal image is a gray level image, and the obtained gray level image is subjected to image preprocessing operation, so that the image quality is improved.
During ultrasonic detection, ultrasonic waves can be absorbed and bounced on the surface of the internal cavity of the cavity to be detected to different degrees due to different material densities of objects in the cavity to be detected, the larger the density value is, the larger the gray value is, the brighter the image is, the darker the image is, then the probe is used for transmitting the ultrasonic waves which are absorbed and bounced, the ultrasonic waves are received back to form a gray image, and the gray value and the distribution of pixel points in the image are determined according to the intensity of the ultrasonic waves received by the ultrasonic probe. Thus, according to this principle, the internal image of the coating is obtained in combination with the echo of the impurity and bubble surfaces in the coating to different degrees of ultrasound. And solving a sum value sequence of gray values of pixel points of rows and columns for the high-quality image after the image preprocessing, and establishing a coordinate system to draw a change curve of the sum value sequence in the row and column direction.
The invention adopts ultrasonic-C scanning, wherein C ultrasonic is a two-dimensional scanning imaging technology for extracting echo information vertical to a designated section of an acoustic beam, and layered image information of defects can be extracted by using the C ultrasonic so as to analyze an internal three-dimensional structure, wherein the total number of layers in scanning is L.
When the swelling phenomenon exists in the coating on the surface of the object with the higher metal equal density, the normal part of the ultrasonic image is brighter, and the defect part is darker, so that the magnitude of the sum can be reduced by carrying out the inverse treatment on the internal image. Summing the gray values of the obtained inverted gray images row by row and column by column:
wherein ,row accumulation sequence representing the ith row in an inverted gray scale image,/>A column accumulation sequence representing an ith column in the inverted gray scale image, i representing an ith row in the inverted gray scale image, n representing a total number of rows of the inverted gray scale image, j representing a jth column in the inverted gray scale image, m representing a total number of columns of the inverted gray scale image, and>and the gray value of the pixel of the ith row and the jth column in the inverted gray image is represented.
Counting the row accumulation sequence and the column accumulation sequence in the reversed gray level image, and respectively drawing curves to obtain a row accumulation sequence curve and a column accumulation sequence curveLines, wherein the correspondence between the row accumulation sequence curve and the column accumulation sequence curve and the inverse gray scale image is shown in figure 2,for the first defective area in the inverted gray image, is>For the second defective area in the inverted gray image, is>Points on the column accumulation sequence curve, +.>The points on the line accumulation sequence curve, respectively.
Step S002, respectively counting the wave crest number and the wave trough number in the row accumulation sequence curve and the column accumulation sequence curve; respectively acquiring a row range end cutting average value of a row accumulation sequence curve and a column range end cutting average value of a column accumulation sequence curve based on the number of wave crests and the number of wave troughs; taking the numerical values of the line range end-cutting average value and the column range end-cutting average value as standard lines, and respectively calculating the duty ratio of the sectional line accumulation sequence curve and the sectional column accumulation sequence curve in the whole accumulation sequence curve to which the sectional line accumulation sequence curve belongs for the sectional line accumulation sequence curve corresponding to the numerical value larger than the line range end-cutting average value and the sectional column accumulation sequence curve corresponding to the numerical value larger than the column range end-cutting average value; a defective region in the inverse gray scale image is acquired based on the duty ratio.
Specifically, the extremum in the row accumulation sequence curve and the column accumulation sequence curve is J, wherein the extremum isMinimum value->The maxima correspond to peaks in the curve and the minima correspond to valleys. And counting the wave peaks and wave troughs in the row accumulation sequence curve and the column accumulation sequence curve to obtain the wave peak quantity f and the wave trough quantity g.
In fact, the row and column accumulation sequence curves are not smooth curves, but rather are formed of discrete points, so that data processing of the row and column accumulation sequence curves is required to obtain effective information at the location of the defect in the image.
Setting the line range end-cutting average value of the line accumulation sequence curve as H and the column range end-cutting average value of the column accumulation sequence curve as L, and setting the line accumulation sequence as H and the column accumulation sequence as L, wherein the range end-cutting average value is the specification of the end-cutting average value, i.e. selecting a range with the size of e=15, and carrying out end-cutting average value on the extreme value sequence in the range, and the removed part of the line accumulation sequence is、/>The part removed from the column accumulation sequence is +.>、/>The following calculation formulas are respectively given by the row range end-cut average value H and the column range end-cut average value L:
in the formula ,the line range end cutting average value; />The number of peaks; />Is preset toA value; />Accumulating the sequence for the row of row a; />The column range end average value; />Is the number of wave troughs.
By using a range-cut average, it is possible to avoid that subsequent data analysis is affected by some oversized extremum, resulting in neglecting small defects.
Calculating the duty ratio of each sectional row accumulation sequence curve and sectional column accumulation sequence curve in the whole accumulation sequence curve for the continuous accumulation sequence larger than H, L by taking the numerical value of H, L as a benchmark、/>：
wherein The duty cycle of the sequence curve is accumulated for the segmented rows on the row accumulation sequence, +.>Accumulating the duty cycle of the sequence curve for the segmented columns over the column accumulation sequence; />For a two-dimensional summation over a segmented row accumulation sequence curve of a row accumulation sequence,summing the two dimensions of the whole row accumulation sequence curve; />For a two-dimensional summation over a segmented column accumulation sequence curve of the column accumulation sequence, +.>Summing the two dimensions of the whole column accumulated sequence curve; x and y are the horizontal axis coordinate positions of the points on the row accumulation sequence curve and the column accumulation sequence curve which are equal to the range end cutting average value H, L respectively; />、/>Representing a segmented row accumulation sequence curve and a segmented column accumulation sequence curve, respectively.
When the duty ratio is larger than a preset duty ratio threshold value, mapping an area surrounded by the corresponding segmented row accumulation sequence curve and the segmented column accumulation sequence curve in the reverse gray level image as a defect area.
Preferably, the preset duty cycle threshold is 0.3.
Step S003, sorting the internal images corresponding to the defect areas from near to far by adjusting the acquisition distance of the internal images to obtain corresponding internal image sets; taking the last internal image in the internal image set as a target internal image, constructing a gray level distribution curve of a defect middle line in the target internal image according to the gray level value, and obtaining a peak point of the gray level distribution curve to calculate a symmetrical value; and calculating the peak distance discrete value of the defect middle row based on the column coordinate sequence of the peak points.
Specifically, the imaging principle of ultrasonic waves is based on different attenuation degrees of energy in different media (when bulges in the coating are bubbles, only a single-layer coating on the outermost layer affects ultrasonic attenuation, the width of the single-layer coating is the same, and the blocking capability of the single-layer coating on different positions of the same layer at the defect is the same, so that the single-layer gray level at the defect position of the bubble in the image of the same layer is low in discrete degree, the average gray level of each layer in the image is low in discrete degree and slow in change, when the bulges are impurities, the effect of blocking attenuation on the ultrasonic waves is also achieved, and because the density of the impurities is uneven inside, the blocking attenuation degree of the ultrasonic waves by different layers is different for each layer, so that the gray level of the single-layer image at the defect position after ultrasonic imaging is high in discrete degree, the average gray level of each layer in the image is high and changeable, the gray level consistency of the corresponding region is high and the gray level value of the defect region of different sections is similar in the ultrasonic image of the coating; the gray level consistency of the impurity regions is generally smaller, the gray level distribution of the defect regions with different sections is different, and the detection and identification are realized by utilizing the difference of the two defects in the ultrasonic image.
The method comprises the steps of (1) sequencing an internal image set of a defect area from near to far according to the distance between the internal image set and an ultrasonic probe, counting the gray values of pixel points in the defect area, and performing the following calculation and analysis by using the obtained gray values:
taking the last internal image in the internal image set as a target internal image, and constructing a gray level distribution curve of a defect middle line in the target internal image according to the gray level value, wherein the gray level distribution curve of the defect middle line has a mapping corresponding relation with the position of the line and the gray level change of the pixel pointThe method comprises the following steps:
formula logic is middle row = mapping length + compensation length;
the method reflects that the gray level distribution curve of a certain line at the defect position has a corresponding relation with the line number of the line, and because the gray level information carried by the middle line at the defect position is more, the middle line is selected as an analysis basis, and the following analysis is carried out on the mutual position relation (the peak corresponds to the attenuation abrupt change part of the ultrasonic wave in the coating) between the gray level distribution peaks of the certain line.
The mapping length YS is a fixed parameter value, and the compensation length BC is the difference between the gray value of the pixel corresponding to the peak in the gray distribution curve of the middle line at the defect position and the gray value of the first pixel of the gray distribution curve.
Acquiring the peak point and the maximum point corresponding to the defect in the gray level distribution curve of the defect middle line, and acquiring the column coordinates of the peak point and the maximum point on the defect middle line according to the mapping corresponding relation, wherein the column coordinates of the peak point are as follows、/>、/>、…、/>The column coordinate of the maximum point is ZD.
Calculating symmetry D of a defect middle line gray level distribution curve in the target internal image at the defect position: because the noise obstructs and attenuates the ultrasonic wave, in the target internal image, the complicated attenuation process is already carried out, because the degree of attenuation of the ultrasonic wave is the same at each place in the bubble, and because the imaging image has space symmetry, the imaged image has symmetry on a certain row or a certain column, and the internal density of the noise is complex, and the characteristics are not provided, so that the symmetry of the defect intermediate gray distribution curve in the target internal image can be analyzed as a subsequent judgment index.
wherein ,indicated at defect middle row ZH, at +.>Gray values of column pixel points; />Indicated at defect middle row ZH, at +.>Column (i.e. about maximum ZD as symmetry axis +.>On the other side of the symmetry axis), i.e., the pixel point gray value, x is the number of peaks of the middle line gray distribution curve.
And calculating the absolute difference value of the peak value between the first peak point on the left side and the last peak point on the right side (namely, the gray value in the corresponding target internal image) by taking the column coordinate ZD of the maximum value point as a symmetry axis, calculating the absolute difference value of the peak value between the second peak point on the left side and the last peak point on the right side according to the method, analogically, and finally accumulating and summing all the obtained absolute difference values to obtain an average value as a symmetry value D, wherein the larger the D value is, the larger the gray difference on the two sides is, the lower the symmetry is, and the more likely the impurity is.
The symmetry value D is the corresponding degree of the peak values of the two sides of the gray level distribution curve of the defect middle line, and the higher the corresponding degree is, the smaller the symmetry value D is.
In addition, obtaining a defect middle line gray level distribution curve at each layer defect and obtaining a peak sequence according to the column coordinate sequence of the peak、/>、/>、…、/>Calculating the backward differential sequence of the sequence, which is marked as +.>、/>、/>、…、/>, wherein />The backward differential sequence of column coordinate sequences of peaks here represents the distance between each peak and the previous one, referred to herein as the peak distance sequence.
Calculating the peak distance discrete value of the defect middle line in each internal image in the internal image set：
in the formula ,is at the +.>Backward differential sequence elements of column coordinates of the peak points; />Concentration of the internal image->Average distance between peak points of defect middle line in the internal image; />Is the number of elements of the peak distance sequence.
Peak distance discreteness of the defect middle lineReflecting the +.>Whether the distance of the wave crest in the gray level distribution curve of the defect middle line in the internal image is nearly uniform or not, < >>The larger the value, the greater the degree of distance dispersion, i.e. the less likely the defect is to be an impurity; on the contrary, let(s)>The smaller the value, the closer to coincidence, the more likely the defect is a bubble.
Step S004, obtaining a peak distance discrete value corresponding to each internal image in the internal image set, and calculating fault ultrasonic attenuation discrete parameters of the target internal image by combining the symmetrical values and all the peak distance discrete values; calculating the average gray scale change rate of the internal image of the target; taking the product of the fault ultrasonic attenuation discrete parameter and the average gray level change rate as a defect type index of the target internal image, and confirming the defect type of the coating according to the defect type index.
Specifically, the symmetry value D and the single-layer peak distance discrete value of the gray distribution curve of the defect middle line of the target internal image obtained by step S002Obtaining the peak distance discrete value of the gray level distribution curve of the defect middle line of all layers in the internal image set +.>Obtaining a set of discrete peak distance values corresponding to each layer in the internal image set, namely +.>、、/>、…、/>Calculating a fault ultrasonic attenuation discrete parameter of an internal image of the target according to the peak distance discrete value set>。
The lower the degree of symmetry, the higher the peak distance discreteness, and the more likely to be impurities, so that the fault ultrasonic attenuation discreteness parameterThe calculation formula of (2) is as follows:
wherein D is the symmetry value of the gray level distribution curve of the defect middle line of the target internal image,concentration of the internal image->Peak distance discrete value of internal image, +.>For all mathematical expectations of the peak distance dispersion values, i.e. mean +.>The total number of images that are the internal image set.
Fault ultrasonic attenuation discrete parameterThe internal images are concentrated in the same defect area, the average distance of the peaks of the gray level distribution curve of the defect middle line of each internal image is taken as sample data, and the discrete degree of standard deviation analysis data is calculated>The larger the value, the greater the degree of dispersion of the average distance value of each internal image in the internal image set, the more likely the defect is an impurity; on the contrary, let(s)>The smaller the value, the closer to coincidence, the more likely the defect is a bubble.
The fault ultrasonic attenuation discreteness parameter is symmetry and discreteness of defect intermediate lines in all internal images in the internal image set, and the parameter is thatThe greater the value, the greater the degree of dispersion.
Because of the average gray level discreteness in the atlas in the aboveIs the evaluation of the dispersion of the whole of the atlas in the average gray value, so that the gray dispersion in the single-layer image of the atlas is possible for the complexity of adjacent internal density due to impurities>And average gray level discreteness in the graphic set +.>Since the defect type is likely to be erroneously judged, the average gradation change rate at the defective area is calculated>：
Average gray scale rate of changeThe rate of attenuation of the ultrasonic wave in the coating layer is expressed as the gray scale change rate between the last coating layer and the first coating layer, because the ultrasonic wave is slow in the bubble and fast in the impurity,/th>The larger the value, the faster the decay, the more likely to be an impurity; and conversely, bubbles.
in the formula ,for the average gray value of the defect area of the target internal image, i.e. the last layer image, the internal image set has A internal images in total,/>Is the average gray value at the defective area of the 1 st internal image.
According to the defect region obtained in the above, the relevant gray scale characteristics of the ultrasonic image, namely the fault ultrasonic attenuation discrete parameterAverage gray scale change rate->The following index of defect type is derived>：
Defect type indexThe larger the defect, the more likely it is an impurity, and the smaller the defect, the more likely it is a bubble defect.
in the formula ,for the discrete parameter of fault ultrasonic attenuation +.>Is the average gray scale rate of change.
And obtaining a large amount of related data of the related gray features by using a defect database, and obtaining a classification threshold value of each feature value by using an Ojin algorithm to realize defect type identification under the ultrasonic coating.
In summary, the embodiment of the invention collects the internal image of the coating, and performs the inverse processing on the internal image to obtain the corresponding inverse gray level image; constructing a row accumulation sequence curve and a column accumulation sequence curve; obtaining a defect area in the reversed gray level image according to the row accumulation sequence curve and the column accumulation sequence curve; acquiring an internal image set, and constructing a gray level distribution curve of a defect middle line in a target internal image to calculate a peak distance discrete value and a symmetric value of the defect middle line; and calculating a fault ultrasonic attenuation discrete parameter and an average gray level change rate of the internal image of the target to obtain a defect type index, and determining the defect type of the coating according to the defect type index. The invention improves the efficiency of defect identification under the coating.
It should be noted that: the sequence of the embodiments of the present invention is only for description, and does not represent the advantages and disadvantages of the embodiments. 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 are also possible or may be advantageous.
In this specification, each embodiment is described in a progressive manner, and identical and similar parts of each embodiment are all referred to each other, and each embodiment mainly describes differences from other embodiments.
The foregoing description of the preferred embodiments of the present invention is not intended to be limiting, but rather, any modifications, equivalents, improvements, etc. that fall within the principles of the present invention are intended to be included within the scope of the present invention.
Claims (9)
1. A method for detecting defects in a coating using an ultrasound image, the method comprising the steps of:
collecting an internal image of the coating, and carrying out reverse processing on the internal image to obtain a corresponding reverse gray level image; respectively constructing a row accumulation sequence curve and a column accumulation sequence curve according to the row accumulation sequence of each row and the column accumulation sequence of each column in the reversed-phase gray level image;
respectively counting the wave crest number and the wave trough number in the row accumulation sequence curve and the column accumulation sequence curve; respectively acquiring a row range end cutting average value of a row accumulation sequence curve and a column range end cutting average value of a column accumulation sequence curve based on the number of wave crests and the number of wave troughs; taking the numerical values of the line range end-cutting average value and the column range end-cutting average value as standard lines, and respectively calculating the duty ratio of the sectional line accumulation sequence curve and the sectional column accumulation sequence curve in the whole accumulation sequence curve to which the sectional line accumulation sequence curve belongs for the sectional line accumulation sequence curve corresponding to the numerical value larger than the line range end-cutting average value and the sectional column accumulation sequence curve corresponding to the numerical value larger than the column range end-cutting average value; acquiring a defect area in the reverse gray image based on the duty ratio;
sequencing the internal images corresponding to the defect area from near to far by adjusting the acquisition distance of the internal images to obtain a corresponding internal image set; taking the last internal image in the internal image set as a target internal image, constructing a gray level distribution curve of a defect middle line in the target internal image according to the gray level value, and obtaining a peak point of the gray level distribution curve to calculate a symmetrical value; calculating a peak distance discrete value of the defect middle row based on a column coordinate sequence of the peak points;
acquiring a peak distance discrete value corresponding to each internal image in the internal image set, and calculating a fault ultrasonic attenuation discrete parameter of the target internal image by combining the symmetrical value and all the peak distance discrete values; calculating the average gray scale change rate of the internal image of the target; taking the product of the fault ultrasonic attenuation discrete parameter and the average gray level change rate as a defect type index of the target internal image, and confirming the defect type of the coating according to the defect type index.
2. A coating defect detection method using an ultrasound image as set forth in claim 1, wherein the row accumulation sequence refers to a gray value sum of all pixels of the corresponding row, and the column accumulation sequence refers to a gray value sum of all pixels of the corresponding column.
3. The method for detecting defects in a coating using an ultrasonic image according to claim 1, wherein the method for obtaining the line-range end-of-line average value of the line accumulation sequence curve comprises the steps of:
wherein ,the line range end cutting average value; />The number of peaks; />Is a preset value; />The sequence is accumulated for row a.
4. The method for detecting defects in a coating using an ultrasonic image according to claim 1, wherein the method for obtaining a column-range end-of-line average value of the column-cumulative sequence curve comprises:
wherein ,the column range end average value; />Is the number of wave troughs; />Is a preset value; />The sequence is accumulated for column b.
5. A coating defect detection method using ultrasonic images according to claim 1, wherein the method for acquiring defect areas in inverse gray scale images based on the duty ratio comprises:
when the duty ratio is larger than a preset duty ratio threshold value, mapping an area surrounded by the corresponding segmented row accumulation sequence curve and the segmented column accumulation sequence curve in the reverse gray level image as a defect area.
6. The method for detecting coating defects using ultrasonic images according to claim 1, wherein the method for obtaining peak points of gray scale distribution curves to calculate symmetry values comprises:
and obtaining peak points and maximum points of the gray level distribution curve, taking the column coordinates of the maximum points as symmetry axes, calculating absolute differences of corresponding gray levels between a first peak point on the left side of the symmetry axes and a last peak point on the right side of the symmetry axes, then calculating absolute differences of corresponding gray levels between a second peak point on the left side of the symmetry axes and a last-to-last peak point on the right side of the symmetry axes, and so on, and taking the result of accumulating, summing and averaging all the calculated absolute differences as symmetry values.
7. The method for detecting defects in a coating using an ultrasonic image according to claim 1, wherein the method for obtaining the discrete value of the peak distance of the intermediate line of the defect calculated based on the column coordinate sequence of the peak point comprises the steps of:
and forming a column coordinate sequence by column coordinates of all the wave peak points, calculating a backward differential sequence of the column coordinate sequence, obtaining average distances among the wave peak points, calculating a difference square between each element in the backward differential sequence and the average distances, and calculating a mean value of the difference squares as a peak distance discrete value.
8. A coating defect detection method using ultrasonic images as set forth in claim 1, wherein the method for calculating tomographic ultrasound attenuation discrete parameters of the internal image of the object by combining the symmetry value and all the peak distance discrete values comprises:
calculating the average value of all the peak distance discrete values, obtaining a first difference square between each peak distance discrete value and the average value, obtaining a first difference square sum, taking the first difference square sum as a denominator, taking the product of the symmetrical value and the number of images in the internal image set as the denominator to obtain a corresponding ratio, and taking the result of the secondary evolution of the ratio as a fault ultrasonic attenuation discrete parameter of the target internal image.
9. A coating defect detection method using ultrasonic images according to claim 1, wherein the method of calculating the average gray scale change rate of the internal image of the object comprises:
calculating a first average gray value of the target internal image and a second average gray value of the first internal image in the internal image set; obtaining a difference value between the first average gray value and the second average gray value, obtaining a corresponding ratio by taking the difference value as a numerator and taking the difference value of the image value between the target internal image and the first internal image as a denominator, and taking the ratio as the average gray change rate of the target internal image; the image value refers to an A value of an A-th internal image in the internal image set, and A is a positive integer.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310594608.0A CN116309608B (en) | 2023-05-25 | 2023-05-25 | Coating defect detection method using ultrasonic image |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310594608.0A CN116309608B (en) | 2023-05-25 | 2023-05-25 | Coating defect detection method using ultrasonic image |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN116309608A CN116309608A (en) | 2023-06-23 |
| CN116309608B true CN116309608B (en) | 2023-08-04 |
Family
ID=86794572
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310594608.0A Active CN116309608B (en) | 2023-05-25 | 2023-05-25 | Coating defect detection method using ultrasonic image |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116309608B (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116958136B (en) * | 2023-09-19 | 2024-01-30 | 惠州市金箭精密部件有限公司 | Lead screw thread production defect detection method based on image processing |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102393422A (en) * | 2011-08-22 | 2012-03-28 | 江苏省产品质量监督检验研究院 | Ultrasonic time of flight diffraction (TOFD)-based offline defect judgment method |
| CN114881969A (en) * | 2022-05-06 | 2022-08-09 | 蔡志刚 | Wood knot detection method and detection system based on computer vision on wood board surface |
| CN115187590A (en) * | 2022-09-08 | 2022-10-14 | 山东艾克赛尔机械制造有限公司 | Automobile part defect detection method based on machine vision |
| CN115272334A (en) * | 2022-09-29 | 2022-11-01 | 江苏美克美斯自动化科技有限责任公司 | Method for detecting micro defects on surface of steel rail under complex background |
| CN115457031A (en) * | 2022-10-27 | 2022-12-09 | 江苏集宿智能装备有限公司 | Method for identifying internal defects of integrated box based on X-ray |
Family Cites Families (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP4233397B2 (en) * | 2002-10-01 | 2009-03-04 | 株式会社東京精密 | Image defect inspection method, image defect inspection apparatus, and appearance inspection apparatus |
| TWI603074B (en) * | 2015-07-03 | 2017-10-21 | 元智大學 | Optical film defect detection method and system thereof |
-
2023
- 2023-05-25 CN CN202310594608.0A patent/CN116309608B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102393422A (en) * | 2011-08-22 | 2012-03-28 | 江苏省产品质量监督检验研究院 | Ultrasonic time of flight diffraction (TOFD)-based offline defect judgment method |
| CN114881969A (en) * | 2022-05-06 | 2022-08-09 | 蔡志刚 | Wood knot detection method and detection system based on computer vision on wood board surface |
| CN115187590A (en) * | 2022-09-08 | 2022-10-14 | 山东艾克赛尔机械制造有限公司 | Automobile part defect detection method based on machine vision |
| CN115272334A (en) * | 2022-09-29 | 2022-11-01 | 江苏美克美斯自动化科技有限责任公司 | Method for detecting micro defects on surface of steel rail under complex background |
| CN115457031A (en) * | 2022-10-27 | 2022-12-09 | 江苏集宿智能装备有限公司 | Method for identifying internal defects of integrated box based on X-ray |
Non-Patent Citations (1)
| Title |
|---|
| CRACK-LIKE DEFECT DETECTION AND SIZING FROM IMAGE SEGMENTATION THROUGH COOCCURRENCE MATRIX ANALYSIS;MOYSAN, J et al.;ULTRASONICS;第359-363页 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN116309608A (en) | 2023-06-23 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN112581457B (en) | Pipeline inner surface detection method and device based on three-dimensional point cloud | |
| Merazi-Meksen et al. | Mathematical morphology for TOFD image analysis and automatic crack detection | |
| CN116309608B (en) | Coating defect detection method using ultrasonic image | |
| CN116664559B (en) | Machine vision-based memory bank damage rapid detection method | |
| CN113643275B (en) | Ultrasonic defect detection method based on unsupervised manifold segmentation | |
| CN114943736B (en) | Production quality detection method and system for automobile radiating fins | |
| CN115797354B (en) | Method for detecting appearance defects of laser welding seam | |
| CN116758084B (en) | Intelligent detection method for welding defects of sheet metal parts based on image data | |
| CN110348459B (en) | Sonar image fractal feature extraction method based on multi-scale rapid carpet covering method | |
| CN110544233A (en) | Depth image quality evaluation method based on face recognition application | |
| CN113066064A (en) | Cone beam CT image biological structure identification and three-dimensional reconstruction system based on artificial intelligence | |
| CN113689415A (en) | Steel pipe wall thickness online detection method based on machine vision | |
| CN116862917B (en) | Textile surface quality detection method and system | |
| CN112465743A (en) | Periodic structure quality detection method | |
| CN103093474B (en) | Three-dimensional mammary gland ultrasound image partition method based on homoplasmon and partial energy | |
| CN117011292A (en) | Method for rapidly detecting surface quality of composite board | |
| CN107657620A (en) | A kind of method and system of textured metal freezing region recognition | |
| CN114881969A (en) | Wood knot detection method and detection system based on computer vision on wood board surface | |
| CN113160301B (en) | Method and system for measuring bending radius of bent pipe | |
| CN117474873B (en) | Surface treatment system before brazing of high-chromium wear-resistant castings | |
| CN113850914A (en) | Matrix conversion method for linear laser three-dimensional scanning point cloud data | |
| CN107478656A (en) | Paper pulp mixing effect method of determination and evaluation based on machine vision, device, system | |
| CN111915630B (en) | Ceramic material grain segmentation algorithm based on data and model combined driving | |
| CN114494165A (en) | Clustering-based light bar extraction method and device | |
| CN110634117B (en) | Method for quantitatively evaluating image by utilizing three-dimensional parameters |
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 |