CN116091499A - Abnormal paint production identification system - Google Patents
Abnormal paint production identification system Download PDFInfo
- Publication number
- CN116091499A CN116091499A CN202310362494.7A CN202310362494A CN116091499A CN 116091499 A CN116091499 A CN 116091499A CN 202310362494 A CN202310362494 A CN 202310362494A CN 116091499 A CN116091499 A CN 116091499A
- Authority
- CN
- China
- Prior art keywords
- gray
- region
- suspected
- characteristic
- area
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
- 239000003973 paint Substances 0.000 title claims abstract description 42
- 238000004519 manufacturing process Methods 0.000 title claims abstract description 40
- 230000002159 abnormal effect Effects 0.000 title description 7
- 230000002776 aggregation Effects 0.000 claims abstract description 36
- 238000004220 aggregation Methods 0.000 claims abstract description 36
- 230000011218 segmentation Effects 0.000 claims abstract description 35
- 239000011248 coating agent Substances 0.000 claims abstract description 24
- 238000000576 coating method Methods 0.000 claims abstract description 24
- 239000011159 matrix material Substances 0.000 claims abstract description 23
- 238000005507 spraying Methods 0.000 claims abstract description 10
- 238000004590 computer program Methods 0.000 claims abstract description 3
- 230000007547 defect Effects 0.000 claims description 68
- 238000000034 method Methods 0.000 claims description 21
- 230000008569 process Effects 0.000 claims description 7
- 230000000694 effects Effects 0.000 claims description 6
- 238000004891 communication Methods 0.000 claims description 3
- 230000000875 corresponding effect Effects 0.000 claims 20
- 230000002596 correlated effect Effects 0.000 claims 2
- 230000005856 abnormality Effects 0.000 abstract description 12
- 238000001514 detection method Methods 0.000 abstract description 7
- 238000012545 processing Methods 0.000 abstract description 7
- 229910000831 Steel Inorganic materials 0.000 description 8
- 230000001419 dependent effect Effects 0.000 description 8
- 239000010959 steel Substances 0.000 description 8
- 238000004364 calculation method Methods 0.000 description 6
- 230000007423 decrease Effects 0.000 description 6
- 238000004422 calculation algorithm Methods 0.000 description 4
- 238000009826 distribution Methods 0.000 description 4
- 230000006870 function Effects 0.000 description 4
- 238000003708 edge detection Methods 0.000 description 3
- 238000005516 engineering process Methods 0.000 description 3
- 238000004458 analytical method Methods 0.000 description 2
- 230000003247 decreasing effect Effects 0.000 description 2
- 239000002184 metal Substances 0.000 description 2
- 229910052751 metal Inorganic materials 0.000 description 2
- 238000010606 normalization Methods 0.000 description 2
- 238000012797 qualification Methods 0.000 description 2
- HCHKCACWOHOZIP-UHFFFAOYSA-N Zinc Chemical compound [Zn] HCHKCACWOHOZIP-UHFFFAOYSA-N 0.000 description 1
- 230000009471 action Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000002146 bilateral effect Effects 0.000 description 1
- 229910010293 ceramic material Inorganic materials 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 238000001914 filtration Methods 0.000 description 1
- 238000009499 grossing Methods 0.000 description 1
- 239000011344 liquid material Substances 0.000 description 1
- 238000005192 partition Methods 0.000 description 1
- 239000000843 powder Substances 0.000 description 1
- 238000012216 screening Methods 0.000 description 1
- 239000011343 solid material Substances 0.000 description 1
Images
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
- G06T7/10—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
- G06T7/10—Segmentation; Edge detection
- G06T7/13—Edge 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
- G06T7/187—Segmentation; Edge detection involving region growing; involving region merging; involving connected component labelling
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/10—Image acquisition modality
- G06T2207/10004—Still image; Photographic image
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/10—Image acquisition modality
- G06T2207/10024—Color image
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/10—Image acquisition modality
- G06T2207/10052—Images from lightfield camera
-
- 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 processing, in particular to a paint production abnormality identification system. The system includes a memory and a processor executing a computer program stored by the memory to perform the steps of: acquiring a gray image of an object to be detected after coating spraying is completed; obtaining corresponding gray characteristic indexes according to the aggregation conditions of the pixel points on each gray level in the suspected central area and the suspected edge area, and obtaining the corresponding segmentation degree of the characteristic area by combining the gradient amplitude values of the edge pixel points of the suspected central area and the edge pixel points on the edge line outside the suspected edge area; obtaining the smooth similarity corresponding to the characteristic region according to the gray region size matrix corresponding to the characteristic region; and determining pit significance based on the segmentation degree and the smooth similarity, and further judging whether to perform early warning. The invention improves the credibility of the paint production detection result.
Description
Technical Field
The invention relates to the technical field of image processing, in particular to a paint production abnormality identification system.
Background
The anticorrosive paint is one kind of liquid or solid material with special protecting, decorating and other functions and is painted onto the surface of article to form film for preventing outer corrosive matter from contacting the article directly. In general, the anticorrosive paint is used for being coated on the steel surface, because the paint contains metal powder which is more active than steel, such as zinc powder, the metal can be prevented from being corroded through electrochemical action, when the anticorrosive paint is sprayed on the steel surface, pit defects appear on the sprayed steel surface due to uneven proportion of the paint, so that the protection effect on the steel is reduced, the steel under the paint is corroded, the service life of the steel is influenced, the abnormal production of the paint is required to be detected, the abnormal production of the paint is detected based on the traditional image processing technology, the defect area in the image of the sprayed steel surface is generally extracted based on the gray threshold value, and then the paint production is evaluated, but the color of the center area of the pit defect is deeper, the gray value is smaller, the color of the edge area of the pit defect is shallower, and the gray value is larger, so that the traditional detection method is not combined with the characteristics of the pit defect, and the reliability of the paint production detection result is lower.
Disclosure of Invention
In order to solve the problem that the reliability of a detection result is low when the existing method is used for detecting the production of the paint, the invention aims to provide a paint production abnormality identification system, and the adopted technical scheme is as follows:
the invention provides a paint production abnormality identification system, which comprises a memory and a processor, wherein the processor executes a computer program stored in the memory so as to realize the following steps:
acquiring a gray image of an object to be detected after coating spraying is completed;
determining a characteristic region according to the gray difference of the pixel points in the gray image; the characteristic region comprises at least two sub-regions; the subareas comprise suspected central areas and suspected edge areas;
obtaining gray characteristic indexes of the corresponding subareas according to the aggregation condition of the pixel points on each gray level in the subareas; obtaining the corresponding segmentation degree of the feature region according to the gray feature index of each sub-region in the feature region, the gradient amplitude of the edge pixel point of the suspected center region and the gradient amplitude of the edge pixel point on the edge line outside the suspected edge region;
obtaining the smooth similarity corresponding to the characteristic region according to the gray region size matrix corresponding to the characteristic region; determining pit significance of the feature area based on the segmentation degree and the smooth similarity; and judging whether to perform early warning or not based on the pit significance.
Preferably, the determining the feature area according to the gray scale difference of the pixel points in the gray scale image includes:
determining pixel points with gray values larger than a preset first gray threshold value in the gray image as suspected defect pixel points, and obtaining a connected domain formed by the suspected defect pixel points; judging whether the number of the pixel points in each connected domain is larger than a preset first number threshold, if so, determining the corresponding connected domain as a suspected defect area, and acquiring the minimum circumscribed rectangle of the suspected defect area;
in the minimum circumscribed rectangle, optionally presetting a second number of pixel points as initial seed points to perform region growth, wherein the growth criterion is as follows: judging whether the gray difference between the growth point and each pixel point in the neighborhood of the growth point is smaller than a preset second gray threshold value, and if so, taking the corresponding neighborhood pixel point as a new growth point; marking a communication domain formed by growing points in the minimum circumscribed rectangle after the growth is completed as a subarea; and determining the minimum circumscribed rectangle with the number of the subareas being more than 2 as a characteristic area.
Preferably, the acquiring process of the suspected center area and the suspected edge area is as follows:
the gray average value of all pixel points in each sub-region in the characteristic region is calculated, the sub-region with the largest gray average value in the characteristic region is used as a suspected central region in the characteristic region, and the sub-region with the smallest gray average value in the characteristic region is used as a suspected edge region in the characteristic region.
Preferably, the obtaining the gray feature index of the corresponding sub-region according to the aggregation condition of the pixel points on each gray level in the sub-region includes:
acquiring a color aggregation vector corresponding to the subarea, and determining the number of aggregation pixel points corresponding to each gray level in the subarea based on the color aggregation vector;
determining the ratio of the number of the aggregation pixel points corresponding to each gray level in the subarea to the total number of the pixel points corresponding to the gray level as the duty ratio of the aggregation pixel points corresponding to the gray level; taking the product of the duty ratio and the corresponding gray level as a first index;
and determining the average value of the first indexes of all gray levels in the subareas as a gray characteristic index of the corresponding subarea.
Preferably, the obtaining the segmentation degree corresponding to the feature region according to the gray feature index of each sub-region in the feature region, the gradient amplitude of the edge pixel point of the suspected center region, and the gradient amplitude of the edge pixel point on the edge line outside the suspected edge region includes:
the average value of the gradient amplitude values of all the edge pixel points in the suspected central area is recorded as a first average value, and the average value of the gradient amplitude values of all the edge pixel points on the outer edge line of the suspected edge area is recorded as a second average value;
and obtaining the segmentation degree corresponding to the feature region according to the gray characteristic index of the suspected center region, the gray characteristic index of the suspected edge region, the first average value and the second average value, wherein the gray characteristic index, the first average value and the second average value corresponding to the suspected edge region are in positive correlation with the segmentation degree, and the gray characteristic index corresponding to the suspected center region is in negative correlation with the segmentation degree.
Preferably, the obtaining the smooth similarity corresponding to the feature area according to the gray area size matrix corresponding to the feature area includes:
in the gray area size matrix, respectively determining differences between the average value of each data in a preset column and all data in the column where the data are located as first differences, wherein all the first differences form a first smooth sequence;
in the gray area size matrix, respectively determining the difference between the average value of each data in a preset column and all data in the column where the data are located as a second difference, wherein all the second differences form a second smooth sequence;
and taking the similarity of the first smooth sequence and the second smooth sequence as the corresponding smooth similarity of the characteristic region.
Preferably, the partition degree and the pit saliency are in positive correlation, and the smooth similarity and the pit saliency are in negative correlation.
Preferably, the determining whether to perform the early warning based on the pit significance includes:
recording a characteristic region with pit saliency larger than a preset saliency threshold as a target region, calculating the area occupation ratio of the target region in the gray level image, judging whether the area occupation ratio is larger than a preset occupation ratio threshold, and if so, carrying out early warning; if the detected value is less than or equal to the preset value, early warning is not carried out.
The invention has at least the following beneficial effects:
according to the method, when the defects occur in the production process of the coating, pit defects appear on the surface of the object to be detected after the coating is sprayed on the surface of the object to be detected, the center area of the pit defects has darker color and smaller gray value, the edge area of the pit defects has lighter color and larger gray value, the traditional method for judging whether defects exist in the gray image of the object to be detected based on the gray threshold value does not consider the characteristics of the pit defects, so that the reliability of the detection result is lower. In addition, the defect area has the characteristics of relatively smaller gray level and relatively smaller number of communicated pixels, and the non-defect area has the characteristics of relatively larger gray level and relatively larger number of communicated pixels, so that the pit saliency of the feature area is determined by combining the characteristics, the greater the pit saliency is, the greater the possibility that pit defects exist in the corresponding feature area is indicated, the coating is evaluated based on the pit saliency, and further, whether early warning is needed is judged, so that workers are reminded of timely checking the production process of the coating, the qualification rate of the coating produced later is improved, and the reliability of the coating production detection result is improved while the recognition accuracy of the coating production abnormality is ensured.
Drawings
In order to more clearly illustrate the embodiments of the invention or the technical solutions and advantages of the prior art, the following description will briefly explain the drawings used in the embodiments or the description of the prior art, and it is obvious that the drawings in the following description are only some embodiments of the invention, and other drawings can be obtained according to the drawings without inventive effort for a person skilled in the art.
FIG. 1 is a flow chart of a method performed by a paint production anomaly identification system according to an embodiment of the present invention;
fig. 2 is a schematic diagram of an edge line of a suspected center region and an outer edge line of a suspected edge region in 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 purpose, the following detailed description is given to a paint production anomaly recognition system according to the present invention with reference to the accompanying drawings and the preferred embodiments.
Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs.
The following specifically describes a specific scheme of a paint production anomaly identification system provided by the invention with reference to the accompanying drawings.
An embodiment of a paint production anomaly identification system:
the specific scene aimed at by this embodiment is: and spraying the surface of the object to be detected by using the produced paint, obtaining a gray level image of the object to be detected after the spraying is finished, determining a characteristic region, a suspected center region in the characteristic region and a suspected edge region in the characteristic region based on gray level differences of pixel points in the gray level image, determining corresponding pit significance according to a color aggregation vector and a gray level region size matrix, judging whether abnormality occurs in the paint production process based on the pit significance, and further judging that abnormality early warning is required.
The embodiment provides a paint production abnormality identification system, which is used for realizing the steps shown in fig. 1, and specifically comprises the following steps:
step S1, acquiring a gray image of the object to be detected after coating spraying is completed.
In the embodiment, a CCD camera is firstly arranged, the CCD camera collects the surface image of an object to be detected after coating spraying is completed at a overlooking angle, the surface image of the object to be detected is an RGB image, the collected surface image of the object to be detected is subjected to gray processing to obtain a corresponding gray image, the gray image is preprocessed, the influence caused by noise is eliminated, the accuracy of a subsequent analysis result is improved, and in order to keep the detailed information of a pit part in the image and remove the noise, the embodiment selects bilateral filtering to carry out the denoising processing on the gray image, and an implementer can also adopt other denoising modes, so that excessive details are not needed. And marking the image after denoising as a gray image of the object to be detected after coating spraying is finished.
So far, the gray level image of the object to be detected after the coating is sprayed is obtained.
Step S2, determining a characteristic region according to the gray scale difference of the pixel points in the gray scale image; the characteristic region comprises at least two sub-regions; the sub-regions include a suspected center region and a suspected edge region.
Considering that when an abnormality occurs in the production process of the coating, after the coating is sprayed on an object to be detected, a pit defect occurs on the surface of the object to be detected, wherein the pit defect mainly comprises two parts, one part is positioned in the center of the pit defect and is similar to a pit hole, the characteristic with darker color and smaller gray value is displayed in a gray image, the other part is positioned at the edge position of the pit defect, the coating is protruded, and the characteristic with lighter color and larger gray value is displayed in the gray image, so that the embodiment firstly determines a characteristic area, a suspected center area in the characteristic area and a suspected edge area in the characteristic area based on the gray difference of pixel points in the gray image of the object to be detected.
Specifically, the gray value of each pixel point in the gray image of the object to be detected is obtained, the Otsu algorithm is used for obtaining the optimal segmentation threshold, the Otsu algorithm is a known technology, the detailed process is not repeated, and the optimal segmentation threshold obtained by the Otsu algorithm is recorded as a preset first gray threshold; as another embodiment, the preset first gray threshold may be manually set in advance according to the specific situation. Determining pixel points with gray values larger than a preset first gray threshold value in a gray image of an object to be detected as suspected defect pixel points, and acquiring a connected domain formed by the suspected defect pixel points; considering that the pit defect is formed by a plurality of pixel points in the gray level image, and single suspected defect pixel points which are scattered in the gray level image do not belong to the pixel points of the pit defect, in order to eliminate the interference of the pixel points to the subsequent analysis result, a preset first quantity threshold value is set, wherein the preset first quantity threshold value is 20 in the embodiment, and in the specific application, an implementer can set according to specific conditions; and judging whether the number of the pixel points in each connected domain is larger than a preset first number threshold, if so, determining the corresponding connected domain as a suspected defect area, and analyzing the suspected defect area later, so that the calculation amount is reduced and the accuracy can be improved. Obtaining minimum circumscribed rectangles of each suspected defect area in a gray level image of an object to be detected, then carrying out area division on each minimum circumscribed rectangle by adopting an area growth method, and optionally presetting a second number of pixel points in each minimum circumscribed rectangle as initial seed points for area growth, wherein the growth criteria are as follows: judging whether the gray difference between the growth point and each pixel point in the neighborhood of the growth point is smaller than a preset second gray threshold value, and if so, taking the corresponding neighborhood pixel point as a new growth point; the preset second number in this embodiment is 10, and the preset second gray level threshold is 5, and in a specific application, the practitioner can set according to a specific situation; marking a communication domain formed by growing points in the minimum circumscribed rectangle after the growth is completed as a subarea; the region growing algorithm is the prior art and will not be described in detail here. Because the pit defect is generally formed by two parts of pits and protrusions, if the pit defect exists on the surface of the object to be detected, the pit defect is formed by at least two sub-areas in the gray level image; based on this, the minimum circumscribed rectangle in which the number of sub-areas is greater than 2 is determined as a feature area, which is more likely to be the area where the pit defect is located, so the present embodiment will analyze each feature area separately next.
In this embodiment, a feature area is taken as an example, and other feature areas can be processed by using the method provided in this embodiment. For any feature region in the gray scale image of the object to be detected: the characteristic region at least comprises two sub-regions, the gray average value of all pixel points in each sub-region in the characteristic region is calculated according to the gray value of each pixel point in each sub-region, the sub-region with the largest gray average value in the characteristic region is used as a suspected central region in the characteristic region, the sub-region with the smallest gray average value in the characteristic region is used as a suspected edge region in the characteristic region, other regions except the suspected central region and the suspected edge region in the characteristic region are used as background regions in the characteristic region, and the background regions are regions with normal coating spraying. If the number of sub-regions in the feature region is 2, the background region does not exist in the feature region.
Step S3, according to the aggregation condition of the pixel points on each gray level in the subarea, gray characteristic indexes of the corresponding subarea are obtained; and obtaining the corresponding segmentation degree of the feature region according to the gray feature index of each sub-region in the feature region, the gradient amplitude of the edge pixel point of the suspected central region and the gradient amplitude of the edge pixel point on the edge line outside the suspected edge region.
In the embodiment, in step S2, a suspected center area, a suspected edge area and a background area in the feature area are determined, the gray values of all the pixel points in the suspected center area are assigned to 0, the gray values of all the pixel points in the suspected edge area are assigned to 1, the gray values of all the pixel points in the background area are assigned to 2, that is, the gray values of all the pixel points in the feature area are reassigned for extracting edge lines in the feature area; and performing edge detection on the assigned characteristic region by using a canny edge detection technology to obtain a plurality of edge lines, namely, obtaining the edge line of the suspected central region and the edge line of the suspected edge region in the characteristic region. The canny edge detection technique is a well-known technique and will not be described in detail here. It should be noted that, in this embodiment, reassigning the gray value of the pixel point in the feature area is to extract the edge line of the suspected center area and the edge line of the suspected edge area, and the feature areas mentioned later are feature areas not subjected to assignment processing.
Considering that when a pit defect exists in the feature area, the pixels in the edge area of the defect have a certain aggregation property, the pixels in the center area of the defect have a certain aggregation degree, and the color aggregation vector can reflect the aggregation property of the pixels, the suspected center area and the suspected edge area are analyzed based on the color aggregation vector in the embodiment. Specifically, a color aggregation vector corresponding to each sub-region is obtained, namely, a color aggregation vector corresponding to a suspected center region and a color aggregation vector corresponding to a suspected edge region are obtained; the process of obtaining the color aggregate vector is a well-known technique, and will not be described in detail here. For any sub-region: acquiring the total number of pixel points corresponding to each gray level in the subarea, acquiring the number of aggregate pixel points corresponding to each gray level in the subarea based on the color aggregate vector corresponding to the subarea, and determining the ratio of the number of the aggregate pixel points corresponding to each gray level in the subarea to the total number of the pixel points corresponding to the gray level corresponding to the subarea as the duty ratio of the aggregate pixel points corresponding to the gray level; it should be noted that: the total number of the pixel points corresponding to a certain gray level represents the total number of the pixel points on the gray level, and the number of the aggregate pixel points corresponding to the certain gray level is the total number of the aggregate pixel points on the gray level; taking the product of the duty ratio and the corresponding gray level as a first index; determining the average value of the first indexes of all gray levels in the subarea as a gray characteristic index of the subarea; the specific calculation formula of the gray characteristic index of the suspected center area is as follows:
wherein,,
is a gray characteristic index corresponding to the suspected central region,
is suspected to beThe total number of gray levels in the center region,
in the case of a gray scale level,
for the number of aggregated pixel points corresponding to gray level k in the suspected center region,
the total number of pixel points corresponding to the gray level k in the suspected center area.
Similarly, the calculation process and calculation formula of the gray characteristic index of the suspected edge area analogize the calculation formula of the gray characteristic index of the suspected center area, and are not repeated here.
Thus, the gray characteristic index of the suspected center area and the gray characteristic index of the suspected edge area are obtained.
The center of the pit defect is similar to a pit hole, the characteristic with darker color and smaller gray value is shown in the gray image, the paint at the edge position of the pit defect is raised, the characteristic with lighter color and larger gray value is shown in the gray image, the suspected center area is the darkest part in the characteristic area, the gray value is the smallest in the whole characteristic area, the suspected edge area is the brightest part in the characteristic area, and the gray value is the biggest in the whole characteristic area; the gradient magnitudes of the pixel points on the edge lines between the suspected center region and the suspected edge region are the largest in all edge lines in the feature region. Since the suspected center region is located at the center of the feature region, the suspected edge region is located outside and adjacent to the suspected center region, and as shown in fig. 2, the figure is a schematic view of the edge line of the suspected center region and the outer edge line of the suspected edge region, 1 in the figure indicates the edge line of the feature region, 2 in the figure indicates the edge line of the suspected center region, and 3 in the figure indicates the outer edge line of the suspected edge region.
In the embodiment, the segmentation effect of the edge line in the feature area is evaluated by combining the gradient amplitude of the pixel point on the edge line of the suspected center area and the gradient amplitude of the pixel point on the outer edge line of the suspected edge area. Specifically, the sobel operator is utilized to obtain the gradient amplitude value of the pixel point on each edge line in the characteristic region, the average value of the gradient amplitude values of all the edge pixel points on each edge line is calculated according to the gradient amplitude value of each pixel point on each edge line, the average value of the gradient amplitude values of all the edge pixel points in the suspected center region is recorded as a first average value, and the average value of the gradient amplitude values of all the edge pixel points on the edge line outside the suspected edge region is recorded as a second average value. Obtaining the segmentation degree corresponding to the feature region according to the gray characteristic index of the suspected center region, the gray characteristic index of the suspected edge region, the first average value and the second average value, wherein the gray characteristic index of the suspected edge region, the first average value and the second average value are in positive correlation with the segmentation degree, and the gray characteristic index of the suspected center region and the segmentation degree are in negative correlation. The positive correlation relationship indicates that the dependent variable increases along with the increase of the independent variable, the dependent variable decreases along with the decrease of the independent variable, and the specific relationship can be multiplication relationship, addition relationship, idempotent of an exponential function and is determined by practical application; the negative correlation indicates that the dependent variable decreases with increasing independent variable, and the dependent variable increases with decreasing independent variable, which may be a subtraction relationship, a division relationship, or the like, and is determined by the actual application. As a specific embodiment, a specific calculation formula of the segmentation degree corresponding to the feature region is:
wherein,,
for the degree of segmentation corresponding to the feature region,
is the average value of the gradient amplitude values of all the edge pixel points of the suspected central region,
for the gradient magnitudes of all edge pixels on the edge line outside the suspected edge region,
is a gray characteristic index of a suspected central area,
is a gray characteristic index of the suspected edge area,
is a natural constant which is used for the production of the high-temperature-resistant ceramic material,
is a logarithmic function based on natural constants.
By adopting the method, the segmentation degree corresponding to the characteristic region can be obtained.
Step S4, obtaining the smooth similarity corresponding to the characteristic region according to the gray region size matrix corresponding to the characteristic region; determining pit significance of the feature area based on the segmentation degree and the smooth similarity; and judging whether to perform early warning or not based on the pit significance.
For a characteristic region in a gray image of an object to be detected: quantizing gray values of pixel points in the characteristic region into p gray levels, and obtaining a gray region size matrix corresponding to the characteristic region; the process of obtaining the gray area size matrix is prior art, and is not excessive hereRedundant description; the value of p is 32 in this embodiment, and in a specific application, an implementer can set according to a specific situation; it should be noted that: the number of rows of the gray area size matrix corresponding to the characteristic area is p, the number of columns is q, if the corresponding gray level has no pixel point, zero is filled in the corresponding position in the gray area size matrix, and q also represents the maximum value of the number of connected pixel points in the characteristic area. In a gray region size matrix corresponding to the characteristic region, respectively determining differences between the average value of each data in a preset column and all data in the column where the data are located as first differences, wherein all the first differences form a first smooth sequence; in a gray region size matrix corresponding to the characteristic region, respectively determining differences between the average value of each data in a preset column and all data in the column where the data are located as second differences, wherein all the second differences form a second smooth sequence; and taking the similarity of the first smooth sequence and the second smooth sequence as the corresponding smooth similarity of the characteristic region. In this embodiment, the preset number of rows and the preset number of columns are both 2, so that the front preset column in the gray area size matrix is the first column and the second column, the rear preset column in the gray area size matrix is the last column and the last but one column, and in a specific application, an implementer can set the preset number of rows and the preset number of columns according to specific situations; according to the method, according to all data in a first column of a gray area size matrix, the average value of all data in the first column is calculated, and the absolute value of the difference value between each data in the first column and the average value of all data in the first column is calculated; calculating the average value of all data in the second column according to all data in the second column in the gray area size matrix, calculating the absolute value of the difference between each data in the second column and the average value of all data in the second column, wherein the absolute value of the difference is used for representing the difference between each data and the average value of all data in the column, so as to obtain the difference between each data in the first column and the second column and the average value of all data in the column respectively, and marking each obtained difference as a first difference, namely obtaining a plurality of first differences, and constructing a first smooth sequence based on all the first differences, namely
Wherein, the method comprises the steps of, wherein,
for the first smooth sequence of the sequence,
for a first difference between the 1 st data in the first column and the mean of all the data in the first column,
for a first difference between the 2 nd data in the first column and the mean of all the data in the first column,
for a first difference between the p-th data in the first column and the mean of all data in the first column,
for a first difference between the 1 st data in the second column and the mean of all the data in the second column,
for a first difference between the 2 nd data in the second column and the mean of all the data in the second column,
is the first difference between the p-th data in the second column and the mean of all the data in the second column. The first smoothing sequence can reflect the distribution condition of the pixel points with fewer connected numbers in the characteristic region on the corresponding gray level. Calculating the average value of all the data in the penultimate column according to all the data in the penultimate column in the gray scale area size matrix, and calculating the absolute value of the difference value between each data in the penultimate column and the average value of all the data in the penultimate column; calculating the average value of all data in the last column according to all data in the last column in the gray area size matrix, and calculating the absolute value of the difference value between each data in the last column and the average value of all data in the last column; the absolute value of the difference is used to characterize the difference between the mean value of each data and all the data in the column, so far, the inverse is obtainedCounting the differences between the average value of each data in the second and last columns and all the data in the columns, and recording each difference obtained at this time as a second difference, i.e. obtaining a plurality of second differences, constructing a second sliding sequence based on all the second differences, i.e.
Wherein, the method comprises the steps of, wherein,
for the second smooth sequence of the sequence,
for the second difference between the 1 st data in the penultimate column and the mean of all the data in the penultimate column,
for the second difference between the 2 nd data in the penultimate column and the mean of all the data in the penultimate column,
for the second difference between the p-th data in the penultimate column and the mean of all data in the penultimate column,
for a second difference between the 1 st data in the last column and the mean of all data in the last column,
for a second difference between the 2 nd data in the last column and the mean of all the data in the last column,
is the second difference between the p-th data in the last column and the mean of all the data in the last column. The second smooth sequence can reflect the distribution condition of the pixel points with more connected numbers in the characteristic region on the corresponding gray level. In the characteristic region, the area of the defect region is smaller, the area of the non-defect region is larger, the gray level of the defect region is smaller, and the pixel point is communicatedThe number of the connected pixel points is relatively small, the gray level of the non-defect area is relatively large, and the number of the connected pixel points is relatively large. If the protrusion degree of the pit defect is larger, the number of connected pixels corresponding to the lower gray level in the first smooth sequence is larger, and the number of connected pixels corresponding to the higher gray level in the second smooth sequence is larger, at this time, the first smooth sequence and the second smooth sequence are more dissimilar. Considering that the similarity between two sequences can reflect the similarity of data in the two sequences, the similarity of the first smooth sequence and the second smooth sequence is used as the corresponding smooth similarity of the feature region, and the cosine similarity is used to characterize the similarity of the two smooth sequences in the embodiment, wherein the larger the cosine similarity of the first smooth sequence and the second smooth sequence is, the more similar the first smooth sequence and the second smooth sequence are. The process of calculating the cosine similarity between two sequences is known in the art and will not be described in detail here. When the similarity between the first smooth sequence and the second smooth sequence is larger, the smaller the difference between the first smooth sequence and the second smooth sequence is, the smaller the possibility that pit defects exist in the characteristic region is, namely the larger the smooth similarity corresponding to the characteristic region is; when the similarity between the first smooth sequence and the second smooth sequence is smaller, the larger the difference between the first smooth sequence and the second smooth sequence is, the more likely the characteristic region has pit defects, namely the smaller the smooth similarity corresponding to the characteristic region is.
So far, the corresponding segmentation degree and smooth similarity of the feature area are obtained, and the larger the segmentation degree is, the more likely pit defects exist in the feature area are indicated; the smaller the smooth similarity, the more likely the feature area is that pit areas exist; when the pit defect exists in the feature area, the greater the significance degree of the feature area, so that the pit significance degree of the feature area is determined by combining the segmentation degree and the smooth similarity corresponding to the feature area, wherein the segmentation degree and the pit significance degree are in positive correlation, and the smooth similarity and the pit significance degree are in negative correlation. The positive correlation relationship indicates that the dependent variable increases along with the increase of the independent variable, the dependent variable decreases along with the decrease of the independent variable, and the specific relationship can be multiplication relationship, addition relationship, idempotent of an exponential function and is determined by practical application; the negative correlation indicates that the dependent variable decreases with increasing independent variable, and the dependent variable increases with decreasing independent variable, which may be a subtraction relationship, a division relationship, or the like, and is determined by the actual application. As a specific embodiment, calculating the ratio of the segmentation degree corresponding to the characteristic region to the smooth similarity degree corresponding to the characteristic region, taking the ratio as a significant index of the characteristic region, carrying out normalization processing on the significant index of the characteristic region, taking the normalization result as the pit significance degree of the characteristic region, and ensuring that the pit significance degree of the characteristic region has the value of [0,1].
By adopting the method provided by the embodiment, the pit saliency of each characteristic area in the gray level image of the object to be detected can be obtained, and the larger the pit saliency is, the more likely the pit defect exists in the corresponding characteristic area is indicated, so that the preset saliency threshold value is set, and in the embodiment, the preset saliency threshold value is 0.7, and in the specific application, an implementer can set according to specific conditions; judging the size relation between the pit saliency of each characteristic area in the gray level image of the object to be detected and a preset saliency threshold value, marking the characteristic area with the saliency larger than the preset saliency threshold value as a target area, wherein the possibility of pit defects in the target area is higher, screening out the characteristic areas with more possibility of pit defects, calculating the sum of areas of all the target areas, taking the ratio of the sum of areas of all the target areas to the area of the gray level image of the object to be detected as the area ratio of the target area in the gray level image, wherein the larger the area ratio is, the more likely the coating is abnormal in the production process, so that the preset duty threshold value is set, the preset duty threshold value in the embodiment is 0.3, and in the specific application, an embodiment can set according to specific conditions; judging whether the area occupation ratio of the target area in the gray level image is larger than a preset occupation ratio threshold value, if so, indicating that the surface coating of the object to be detected is abnormal, wherein the coating is very likely to be abnormal in the production process, so that when the area occupation ratio of the target area in the gray level image is larger than the preset occupation ratio threshold value, early warning is carried out to remind a worker to check the production process of the coating; if the paint is smaller than or equal to the paint, the paint spraying on the surface of the object to be detected is normal, the paint production process is judged to be normal, and no early warning is carried out.
According to the embodiment, when an abnormality occurs in the production process of the coating, after the coating is sprayed on the surface of an object to be detected, pit defects appear on the surface of the object to be detected, the color of the center area of the pit defects is darker, the gray value is smaller, the color of the edge area of the pit defects is lighter, the gray value is larger, according to the characteristic, the characteristic area, the suspected center area in the characteristic area and the suspected edge area in the characteristic area are determined according to the gray difference of pixel points in the gray image of the object to be detected, namely, the gray image of the object to be detected is primarily divided and screened, pit defects are more likely to exist in the characteristic area, and the combination of the aggregation condition of the pixel points on each gray level in the suspected center area and the suspected edge area, the gradient amplitude of the edge pixel points on the edge area outside the suspected center area and the gradient amplitude of the edge pixel points on the edge line of the suspected center area are evaluated, so that the segmentation effect of the edge line in the characteristic area is obtained, and the segmentation degree corresponding to the characteristic area is larger, and the defect is likely to exist in the corresponding characteristic area; considering the characteristics that the gray level of the defect area is relatively smaller and the number of connected pixel points is relatively smaller, and the non-defect area has the characteristics that the gray level is relatively larger and the number of connected pixel points is relatively larger, the embodiment also determines the pit saliency of the feature area by combining the characteristics, and the greater the pit saliency is, the greater the possibility that pit defects exist in the corresponding feature area is indicated, the embodiment evaluates whether the paint has production abnormality or not based on the pit saliency, further judges whether early warning is needed or not, reminds workers to timely check the production process of the paint, improves the qualification rate of the paint produced later, and the system provided by the embodiment improves the reliability of the paint production detection result while guaranteeing the recognition precision of the paint production abnormality.
Claims (8)
1. A paint production anomaly identification system comprising a memory and a processor, wherein the processor executes a computer program stored by the memory to effect the steps of:
acquiring a gray image of an object to be detected after coating spraying is completed;
determining a characteristic region according to the gray difference of the pixel points in the gray image; the characteristic region comprises at least two sub-regions; the subareas comprise suspected central areas and suspected edge areas;
obtaining gray characteristic indexes of the corresponding subareas according to the aggregation condition of the pixel points on each gray level in the subareas; obtaining the corresponding segmentation degree of the feature region according to the gray feature index of each sub-region in the feature region, the gradient amplitude of the edge pixel point of the suspected center region and the gradient amplitude of the edge pixel point on the edge line outside the suspected edge region;
obtaining the smooth similarity corresponding to the characteristic region according to the gray region size matrix corresponding to the characteristic region; determining pit significance of the feature area based on the segmentation degree and the smooth similarity; and judging whether to perform early warning or not based on the pit significance.
2. A paint-production anomaly identification system as claimed in claim 1 wherein the determining of the characteristic region from the gray scale difference of the pixels in the gray scale image comprises:
determining pixel points with gray values larger than a preset first gray threshold value in the gray image as suspected defect pixel points, and obtaining a connected domain formed by the suspected defect pixel points; judging whether the number of the pixel points in each connected domain is larger than a preset first number threshold, if so, determining the corresponding connected domain as a suspected defect area, and acquiring the minimum circumscribed rectangle of the suspected defect area;
in the minimum circumscribed rectangle, optionally presetting a second number of pixel points as initial seed points to perform region growth, wherein the growth criterion is as follows: judging whether the gray difference between the growth point and each pixel point in the neighborhood of the growth point is smaller than a preset second gray threshold value, and if so, taking the corresponding neighborhood pixel point as a new growth point; marking a communication domain formed by growing points in the minimum circumscribed rectangle after the growth is completed as a subarea; and determining the minimum circumscribed rectangle with the number of the subareas being more than 2 as a characteristic area.
3. The paint production anomaly identification system of claim 2, wherein the acquiring process of the suspected center region and the suspected edge region comprises:
the gray average value of all pixel points in each sub-region in the characteristic region is calculated, the sub-region with the largest gray average value in the characteristic region is used as a suspected central region in the characteristic region, and the sub-region with the smallest gray average value in the characteristic region is used as a suspected edge region in the characteristic region.
4. The paint production anomaly identification system of claim 1, wherein the obtaining the gray scale characteristic index of each sub-region according to the aggregation of the pixels on each gray scale in the sub-region comprises:
acquiring a color aggregation vector corresponding to the subarea, and determining the number of aggregation pixel points corresponding to each gray level in the subarea based on the color aggregation vector;
determining the ratio of the number of the aggregation pixel points corresponding to each gray level in the subarea to the total number of the pixel points corresponding to the gray level as the duty ratio of the aggregation pixel points corresponding to the gray level; taking the product of the duty ratio and the corresponding gray level as a first index;
and determining the average value of the first indexes of all gray levels in the subareas as a gray characteristic index of the corresponding subarea.
5. The paint production anomaly identification system according to claim 1, wherein the obtaining the corresponding segmentation degree of the feature region according to the gray feature index of each sub-region in the feature region, the gradient amplitude of the edge pixel point of the suspected center region, and the gradient amplitude of the edge pixel point on the edge line outside the suspected edge region comprises:
the average value of the gradient amplitude values of all the edge pixel points in the suspected central area is recorded as a first average value, and the average value of the gradient amplitude values of all the edge pixel points on the outer edge line of the suspected edge area is recorded as a second average value;
and obtaining the segmentation degree corresponding to the feature region according to the gray characteristic index of the suspected center region, the gray characteristic index of the suspected edge region, the first average value and the second average value, wherein the gray characteristic index, the first average value and the second average value corresponding to the suspected edge region are in positive correlation with the segmentation degree, and the gray characteristic index corresponding to the suspected center region is in negative correlation with the segmentation degree.
6. The paint production anomaly identification system of claim 1, wherein the obtaining the smooth similarity corresponding to the feature region according to the gray scale region size matrix corresponding to the feature region comprises:
in the gray area size matrix, respectively determining differences between the average value of each data in a preset column and all data in the column where the data are located as first differences, wherein all the first differences form a first smooth sequence;
in the gray area size matrix, respectively determining the difference between the average value of each data in a preset column and all data in the column where the data are located as a second difference, wherein all the second differences form a second smooth sequence;
and taking the similarity of the first smooth sequence and the second smooth sequence as the corresponding smooth similarity of the characteristic region.
7. The paint producing anomaly identification system of claim 1, wherein the degree of segmentation is positively correlated with the pit significance and the smooth similarity is negatively correlated with the pit significance.
8. The paint production anomaly identification system of claim 1, wherein the determining whether to perform an early warning based on the pit significance comprises:
recording a characteristic region with pit saliency larger than a preset saliency threshold as a target region, calculating the area occupation ratio of the target region in the gray level image, judging whether the area occupation ratio is larger than a preset occupation ratio threshold, and if so, carrying out early warning; if the detected value is less than or equal to the preset value, early warning is not carried out.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310362494.7A CN116091499B (en) | 2023-04-07 | 2023-04-07 | Abnormal paint production identification system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310362494.7A CN116091499B (en) | 2023-04-07 | 2023-04-07 | Abnormal paint production identification system |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN116091499A true CN116091499A (en) | 2023-05-09 |
| CN116091499B CN116091499B (en) | 2023-06-20 |
Family
ID=86210517
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310362494.7A Active CN116091499B (en) | 2023-04-07 | 2023-04-07 | Abnormal paint production identification system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116091499B (en) |
Cited By (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116645429A (en) * | 2023-07-25 | 2023-08-25 | 山东中胜涂料有限公司 | Visual-aided paint production sample color analysis and detection method |
| CN116703910A (en) * | 2023-08-07 | 2023-09-05 | 威海丰荟建筑工业科技有限公司 | Intelligent detection method for quality of concrete prefabricated bottom plate |
| CN116990993A (en) * | 2023-09-26 | 2023-11-03 | 深圳市柯达科电子科技有限公司 | LCD display panel quality detection method |
| CN117115172A (en) * | 2023-10-25 | 2023-11-24 | 山东鲁润阿胶药业有限公司 | Donkey-hide gelatin quality detection method and system based on machine vision |
| CN117611589A (en) * | 2024-01-24 | 2024-02-27 | 深圳市格瑞邦科技有限公司 | Tablet personal computer quality detection method and system |
| CN117635922A (en) * | 2023-12-06 | 2024-03-01 | 北京薇笑美网络科技有限公司 | Quality identification method based on router network cable interface |
| CN117689552A (en) * | 2024-02-02 | 2024-03-12 | 科普云医疗软件(深圳)有限公司 | Coronary angiography enhancement method for intracardiac interventional therapy |
| CN118470024A (en) * | 2024-07-15 | 2024-08-09 | 济宁医学院 | Food foreign matter rapid detection method and system based on image characteristics |
Citations (19)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101852768A (en) * | 2010-05-05 | 2010-10-06 | 电子科技大学 | Workpiece flaw identification method based on compound characteristics in magnaflux powder inspection environment |
| CN104200521A (en) * | 2014-09-04 | 2014-12-10 | 中国科学院遥感与数字地球研究所 | Three-dimensional reconstruction method of building objects based on model prior high-resolution SAR images |
| JP2016124756A (en) * | 2015-01-05 | 2016-07-11 | グローバルウェーハズ・ジャパン株式会社 | Silicon wafer and method for manufacturing the same |
| CN109815807A (en) * | 2018-12-18 | 2019-05-28 | 浙江大学 | A kind of ship detecting method of pulling in shore based on edge line analysis and converging channels feature |
| CN111693549A (en) * | 2020-05-14 | 2020-09-22 | 西安电子科技大学 | Method for detecting and classifying defects of mobile phone cover plate glass |
| CN111798418A (en) * | 2020-06-22 | 2020-10-20 | 电子科技大学 | Wave-absorbing coating speckle defect detection method based on HOG, LBP and GLCM characteristic fusion |
| CN112465706A (en) * | 2020-12-21 | 2021-03-09 | 天津工业大学 | Automatic gate container residual inspection method |
| CN114359190A (en) * | 2021-12-23 | 2022-04-15 | 武汉金丰塑业有限公司 | Plastic product molding control method based on image processing |
| CN114972343A (en) * | 2022-07-29 | 2022-08-30 | 山东睿鑫激光科技有限公司 | Method for detecting production abnormity of laser welding product |
| CN114994102A (en) * | 2022-08-04 | 2022-09-02 | 武汉钰品研生物科技有限公司 | X-ray-based food foreign matter traceless rapid detection method |
| CN115082482A (en) * | 2022-08-23 | 2022-09-20 | 山东优奭趸泵业科技有限公司 | Metal surface defect detection method |
| CN115294120A (en) * | 2022-10-08 | 2022-11-04 | 誉隆半导体设备(江苏)有限公司 | Valve surface quality detection method based on image recognition |
| CN115311277A (en) * | 2022-10-11 | 2022-11-08 | 南通美乐柯材料科技有限公司 | Pit defect identification method for stainless steel product |
| CN115375688A (en) * | 2022-10-25 | 2022-11-22 | 苏州德斯米尔智能科技有限公司 | Automatic detection method for belt type conveying equipment |
| CN115375676A (en) * | 2022-10-24 | 2022-11-22 | 微山三利特不锈钢有限公司 | Stainless steel product quality detection method based on image recognition |
| CN115496755A (en) * | 2022-11-16 | 2022-12-20 | 南京云创大数据科技股份有限公司 | Method and device for detecting mirror-like surface defects based on structured light and gradient analysis |
| CN115526889A (en) * | 2022-11-24 | 2022-12-27 | 江苏太湖锅炉股份有限公司 | Nondestructive testing method of boiler pressure pipeline based on image processing |
| CN115631173A (en) * | 2022-10-28 | 2023-01-20 | 兰州理工大学 | Composite film defect identification method |
| CN115861320A (en) * | 2023-02-28 | 2023-03-28 | 天津中德应用技术大学 | Intelligent detection method for automobile part machining information |
-
2023
- 2023-04-07 CN CN202310362494.7A patent/CN116091499B/en active Active
Patent Citations (19)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101852768A (en) * | 2010-05-05 | 2010-10-06 | 电子科技大学 | Workpiece flaw identification method based on compound characteristics in magnaflux powder inspection environment |
| CN104200521A (en) * | 2014-09-04 | 2014-12-10 | 中国科学院遥感与数字地球研究所 | Three-dimensional reconstruction method of building objects based on model prior high-resolution SAR images |
| JP2016124756A (en) * | 2015-01-05 | 2016-07-11 | グローバルウェーハズ・ジャパン株式会社 | Silicon wafer and method for manufacturing the same |
| CN109815807A (en) * | 2018-12-18 | 2019-05-28 | 浙江大学 | A kind of ship detecting method of pulling in shore based on edge line analysis and converging channels feature |
| CN111693549A (en) * | 2020-05-14 | 2020-09-22 | 西安电子科技大学 | Method for detecting and classifying defects of mobile phone cover plate glass |
| CN111798418A (en) * | 2020-06-22 | 2020-10-20 | 电子科技大学 | Wave-absorbing coating speckle defect detection method based on HOG, LBP and GLCM characteristic fusion |
| CN112465706A (en) * | 2020-12-21 | 2021-03-09 | 天津工业大学 | Automatic gate container residual inspection method |
| CN114359190A (en) * | 2021-12-23 | 2022-04-15 | 武汉金丰塑业有限公司 | Plastic product molding control method based on image processing |
| CN114972343A (en) * | 2022-07-29 | 2022-08-30 | 山东睿鑫激光科技有限公司 | Method for detecting production abnormity of laser welding product |
| CN114994102A (en) * | 2022-08-04 | 2022-09-02 | 武汉钰品研生物科技有限公司 | X-ray-based food foreign matter traceless rapid detection method |
| CN115082482A (en) * | 2022-08-23 | 2022-09-20 | 山东优奭趸泵业科技有限公司 | Metal surface defect detection method |
| CN115294120A (en) * | 2022-10-08 | 2022-11-04 | 誉隆半导体设备(江苏)有限公司 | Valve surface quality detection method based on image recognition |
| CN115311277A (en) * | 2022-10-11 | 2022-11-08 | 南通美乐柯材料科技有限公司 | Pit defect identification method for stainless steel product |
| CN115375676A (en) * | 2022-10-24 | 2022-11-22 | 微山三利特不锈钢有限公司 | Stainless steel product quality detection method based on image recognition |
| CN115375688A (en) * | 2022-10-25 | 2022-11-22 | 苏州德斯米尔智能科技有限公司 | Automatic detection method for belt type conveying equipment |
| CN115631173A (en) * | 2022-10-28 | 2023-01-20 | 兰州理工大学 | Composite film defect identification method |
| CN115496755A (en) * | 2022-11-16 | 2022-12-20 | 南京云创大数据科技股份有限公司 | Method and device for detecting mirror-like surface defects based on structured light and gradient analysis |
| CN115526889A (en) * | 2022-11-24 | 2022-12-27 | 江苏太湖锅炉股份有限公司 | Nondestructive testing method of boiler pressure pipeline based on image processing |
| CN115861320A (en) * | 2023-02-28 | 2023-03-28 | 天津中德应用技术大学 | Intelligent detection method for automobile part machining information |
Non-Patent Citations (5)
| Title |
|---|
| D. CHAVES ET AL: "A segmentation approach for touching char particles", 《2014 XIX SYMPOSIUM ON IMAGE, SIGNAL PROCESSING AND ARTIFICIAL VISION》, pages 1 - 5 * |
| WU JUNSHENG ET AL: "A algorithm of concave-convex defect recognition based on finite element mesh model structure", 《ICCASM 2010》, pages 482 - 486 * |
| 崔巍等: "基于漏磁图像的焊接接头缺陷识别", 《无损检测》, vol. 36, no. 6, pages 24 - 31 * |
| 杨辉华等: "基于水平集和凹点区域检测的粘连细胞分割方法", 《北京邮电大学学报》, vol. 39, no. 6, pages 11 - 16 * |
| 苑玮琦等: "圆柱型覆膜锂电池圆周面凹坑检测方法研究", 《仪器仪表学报》, vol. 41, no. 2, pages 146 - 156 * |
Cited By (14)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116645429A (en) * | 2023-07-25 | 2023-08-25 | 山东中胜涂料有限公司 | Visual-aided paint production sample color analysis and detection method |
| CN116645429B (en) * | 2023-07-25 | 2023-10-20 | 山东中胜涂料有限公司 | Visual-aided paint production sample color analysis and detection method |
| CN116703910A (en) * | 2023-08-07 | 2023-09-05 | 威海丰荟建筑工业科技有限公司 | Intelligent detection method for quality of concrete prefabricated bottom plate |
| CN116703910B (en) * | 2023-08-07 | 2023-10-17 | 威海丰荟建筑工业科技有限公司 | Intelligent detection method for quality of concrete prefabricated bottom plate |
| CN116990993B (en) * | 2023-09-26 | 2023-12-22 | 深圳市柯达科电子科技有限公司 | LCD display panel quality detection method |
| CN116990993A (en) * | 2023-09-26 | 2023-11-03 | 深圳市柯达科电子科技有限公司 | LCD display panel quality detection method |
| CN117115172A (en) * | 2023-10-25 | 2023-11-24 | 山东鲁润阿胶药业有限公司 | Donkey-hide gelatin quality detection method and system based on machine vision |
| CN117115172B (en) * | 2023-10-25 | 2024-01-12 | 山东鲁润阿胶药业有限公司 | Donkey-hide gelatin quality detection method and system based on machine vision |
| CN117635922A (en) * | 2023-12-06 | 2024-03-01 | 北京薇笑美网络科技有限公司 | Quality identification method based on router network cable interface |
| CN117611589A (en) * | 2024-01-24 | 2024-02-27 | 深圳市格瑞邦科技有限公司 | Tablet personal computer quality detection method and system |
| CN117611589B (en) * | 2024-01-24 | 2024-04-12 | 深圳市格瑞邦科技有限公司 | Tablet personal computer quality detection method and system |
| CN117689552A (en) * | 2024-02-02 | 2024-03-12 | 科普云医疗软件(深圳)有限公司 | Coronary angiography enhancement method for intracardiac interventional therapy |
| CN117689552B (en) * | 2024-02-02 | 2024-04-05 | 科普云医疗软件(深圳)有限公司 | Coronary angiography enhancement method for intracardiac interventional therapy |
| CN118470024A (en) * | 2024-07-15 | 2024-08-09 | 济宁医学院 | Food foreign matter rapid detection method and system based on image characteristics |
Also Published As
| Publication number | Publication date |
|---|---|
| CN116091499B (en) | 2023-06-20 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN116091499B (en) | Abnormal paint production identification system | |
| CN115375676B (en) | Stainless steel product quality detection method based on image recognition | |
| CN116452598B (en) | Axle production quality rapid detection method and system based on computer vision | |
| CN115861320B (en) | Intelligent detection method for automobile part machining information | |
| CN107490582B (en) | Assembly line workpiece detection system | |
| CN113554629A (en) | Strip steel red rust defect detection method based on artificial intelligence | |
| CN115994904B (en) | Garment steamer panel production quality detection method based on computer vision | |
| CN110335246B (en) | License picture definition evaluation method | |
| CN113935666B (en) | Building decoration wall tile abnormity evaluation method based on image processing | |
| CN116883408B (en) | Integrating instrument shell defect detection method based on artificial intelligence | |
| CN118297948B (en) | Water pump impeller quality detection method based on image processing | |
| CN114549441A (en) | Sucker defect detection method based on image processing | |
| CN114972892B (en) | Automobile brake pad defect classification method based on computer vision | |
| CN117495849A (en) | Image feature-based power adapter quality detection method | |
| CN116883412B (en) | Graphene far infrared electric heating equipment fault detection method | |
| CN116342586B (en) | Road surface quality detection method based on machine vision | |
| CN116152242A (en) | Visual detection system of natural leather defect for basketball | |
| CN117876971A (en) | Building construction safety monitoring and early warning method based on machine vision | |
| CN109632811A (en) | Structural steel pattern segregation fault detection based on machine vision quantifies ranking method | |
| CN116109577A (en) | Printing label defect detection system and method | |
| CN114332081B (en) | Textile surface abnormity determination method based on image processing | |
| CN117333825B (en) | Cable bridge monitoring method based on computer vision | |
| CN116071692A (en) | Morphological image processing-based water gauge water level identification method and system | |
| CN117274293B (en) | Accurate bacterial colony dividing method based on image features | |
| CN113610753B (en) | Method, device and storage medium for extracting blogging texture features of tongue image |
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 |