CN116777907A - Sheet metal part quality detection method - Google Patents
Sheet metal part quality detection method Download PDFInfo
- Publication number
- CN116777907A CN116777907A CN202311040215.1A CN202311040215A CN116777907A CN 116777907 A CN116777907 A CN 116777907A CN 202311040215 A CN202311040215 A CN 202311040215A CN 116777907 A CN116777907 A CN 116777907A
- Authority
- CN
- China
- Prior art keywords
- pixel point
- illumination
- region
- gray level
- sheet metal
- 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.)
- Withdrawn
Links
- 239000002184 metal Substances 0.000 title claims abstract description 49
- 238000001514 detection method Methods 0.000 title claims abstract description 26
- 238000005286 illumination Methods 0.000 claims abstract description 79
- 230000002159 abnormal effect Effects 0.000 claims abstract description 75
- 238000003708 edge detection Methods 0.000 claims abstract description 16
- 238000000034 method Methods 0.000 claims abstract description 16
- 230000007547 defect Effects 0.000 claims abstract description 14
- 238000012545 processing Methods 0.000 claims abstract description 9
- 230000002708 enhancing effect Effects 0.000 claims abstract description 6
- 238000001914 filtration Methods 0.000 claims abstract description 5
- 230000005856 abnormality Effects 0.000 claims description 28
- 230000003014 reinforcing effect Effects 0.000 claims description 6
- 230000011218 segmentation Effects 0.000 claims description 4
- 238000013528 artificial neural network Methods 0.000 claims description 3
- 238000003672 processing method Methods 0.000 abstract 1
- 238000004458 analytical method Methods 0.000 description 4
- 238000013459 approach Methods 0.000 description 2
- 238000010586 diagram Methods 0.000 description 2
- 238000004519 manufacturing process Methods 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000004891 communication Methods 0.000 description 1
- 238000000605 extraction Methods 0.000 description 1
- 239000007769 metal material Substances 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000007781 pre-processing Methods 0.000 description 1
- 238000006467 substitution reaction Methods 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
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
-
- 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
- G06T7/12—Edge-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
- G06T7/00—Image analysis
- G06T7/60—Analysis of geometric attributes
- G06T7/62—Analysis of geometric attributes of area, perimeter, diameter or volume
-
- 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
- G06T2207/20—Special algorithmic details
- G06T2207/20084—Artificial neural networks [ANN]
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/30—Subject of image; Context of image processing
- G06T2207/30108—Industrial image inspection
- G06T2207/30136—Metal
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Biomedical Technology (AREA)
- Evolutionary Computation (AREA)
- Health & Medical Sciences (AREA)
- Life Sciences & Earth Sciences (AREA)
- Artificial Intelligence (AREA)
- Quality & Reliability (AREA)
- Biophysics (AREA)
- Computational Linguistics (AREA)
- Data Mining & Analysis (AREA)
- Geometry (AREA)
- General Health & Medical Sciences (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- General Engineering & Computer Science (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Investigating Materials By The Use Of Optical Means Adapted For Particular Applications (AREA)
Abstract
The application relates to the technical field of image data processing, in particular to a sheet metal part quality detection method. The method uses an industrial camera to acquire sheet metal part gray image data, and improves an image data processing method: dividing the gray image into areas according to the result obtained by edge detection, then obtaining the illumination component of each pixel point under the illumination condition only by Gaussian filtering, determining the difference between the gray value of each pixel point in each area and the corresponding illumination component, realizing area classification, further obtaining the abnormal degree by combining the position distribution of the connected area and the difference degree of the pixel points, obtaining the enhancement coefficient of the pixel points according to the gray difference and the abnormal degree, enhancing the gray image, and carrying out edge detection on the enhanced gray image to finish sheet metal part quality detection. According to the application, the gray level and the direction characteristic of the scratch defect on the surface of the sheet metal part are combined to strengthen the scratch pixel points, so that more accurate and clear corresponding edge line segments of scratches are obtained, and the judgment accuracy of the scratch defect of the sheet metal part is improved.
Description
Technical Field
The application relates to the technical field of data processing, in particular to a sheet metal part quality detection method.
Background
Sheet metal parts refer to finished products which are manufactured by processing different metal materials according to different requirements, for example: chassis housing, machine tool housing, automotive precision parts, etc. Scratch defects can be generated in the processing and using processes of the sheet metal part, the quality of the sheet metal part is affected, and quality detection is required to be carried out on the sheet metal part after the scratch defects appear. Due to the characteristics of the sheet metal part, reflection of light can occur on the surface, the reflection of light can cause tiny scratches to be unobvious, the integrity extraction of the scratches is affected, and then errors occur in restoration.
In the prior art, when quality detection is carried out on a sheet metal product, edge detection is generally carried out on an image directly according to the gray value difference of a defect part, but due to the existence of reflection of the surface of the sheet metal part, the gray value of a small scratch defect part is not obvious from the difference of a normal area, and scratches cannot be detected well only according to the gray value difference, so that the existing sheet metal part surface defect detection method is inaccurate.
Disclosure of Invention
The application provides a sheet metal part quality detection method, which is used for solving the technical problem that the sheet metal part surface defect detection in the prior art is inaccurate, and the adopted technical scheme is as follows:
the application discloses a sheet metal part quality detection method, which comprises the following steps of:
acquiring a gray level image of the surface of the sheet metal part, and dividing the gray level image into different areas;
acquiring the illumination gray level difference value of the pixel point in each region according to the difference value of the illumination component of the gray level image and the pixel value of each pixel point on the gray level image, determining the illumination gray level difference degree of the region according to the illumination gray level difference value of the pixel point in each region, and determining whether the region belongs to an abnormal region or a possible abnormal region according to the illumination gray level difference degree of each region;
acquiring the degree of abnormality of the pixel points in the possible abnormal area according to the position and direction characteristics of the pixel points in the possible abnormal area;
obtaining an enhancement coefficient of each pixel point in the possible abnormal region according to the illumination gray level difference degree of the possible abnormal region and the abnormal degree of each pixel point in the possible abnormal region;
and reinforcing the gray level image according to the reinforcing coefficient, and carrying out edge detection on the reinforced gray level image to finish the judgment of the surface defect of the sheet metal part.
The beneficial effects of the application are as follows:
according to the method, after a sheet metal part surface gray level image is divided into a plurality of areas, an abnormal area and a possible abnormal area are primarily screened out according to the illumination gray level difference degree of each area, the abnormal degree of a pixel point in the possible abnormal area is obtained according to the position and the direction characteristic of each pixel point in the possible abnormal area, the enhancement coefficient of each pixel point is determined according to the obtained abnormal degree and the illumination gray level difference degree of the possible abnormal area, the enhancement degree of the scratch pixel point by the gray level and the direction characteristic of the sheet metal part surface scratch defect is determined, the determined enhancement coefficient is used for enhancing the gray level image, and then more accurate and clear scratch corresponding edge line segments can be obtained through edge detection, so that the judgment accuracy of the sheet metal part scratch defect is improved.
Further, the method for dividing the gray image into different areas comprises the following steps:
and (3) using an edge detection algorithm to obtain an edge image of the obtained gray image, performing connected domain processing to the obtained edge image, obtaining the length and the width of a minimum circumscribed rectangle of the connected domain, using the smaller value of the length and the width as the length of a window so as to determine the size of the window, and dividing the gray image according to the size of the window to obtain different areas.
Further, the method for obtaining the illumination gray scale difference value of the pixel point in each region according to the difference value between the illumination component of the gray scale image and the pixel value of each pixel point on the gray scale image comprises the following steps:
and carrying out Gaussian filtering treatment on the gray level image to obtain an illumination component of the gray level image, calculating the absolute value of the difference value between the illumination component and the gray level value of each pixel point on the gray level image, and taking the absolute value as the illumination gray level difference value of the corresponding pixel point to obtain the illumination gray level difference value of the pixel point in each region.
Further, the method for determining the illumination gray scale difference degree of each region according to the illumination gray scale difference value of the pixel point in the region comprises the following steps:
in the formula ,indicating the degree of difference of the gray scale of the illumination of the kth region, -, etc.>Representing the variance of the illumination difference of each pixel point in the kth region,/for>Representing the difference of the illumination gray scale of the t pixel point in the k-th area,represents the maximum value in the difference degree of illumination gray scale of each region,/->Representing the minimum value in the degree of difference in the gray level of illumination of the respective areas,/->And the illumination gray scale difference value of the v pixel point in any one area is represented.
Further, the method for determining whether the region belongs to the abnormal region or the possible abnormal region according to the illumination gray scale difference degree of each region comprises the following steps:
and taking the area with the illumination gray level difference degree larger than the threshold value of the illumination gray level difference degree as an abnormal area, and taking the rest areas as possible abnormal areas.
Further, the illumination gray scale difference degree threshold value is 0.7.
Further, the degree of abnormality of the pixel points in the possible abnormal region is:
wherein ,representing the degree of abnormality of each pixel point in the possible abnormality region, +.>Representing the variance of gray level difference between each pixel point and its neighborhood pixel point in the possible abnormal area, +.>Gray value representing any one pixel point in the possible abnormal area, < >>And the gray value of the ith pixel point in the 8-neighbor of any pixel point in the possible abnormal area is represented.
Further, the enhancement coefficient of each pixel point in the possible abnormal area is:
wherein ,for the enhancement coefficient of each pixel point in the possible abnormal area, +.>Representing the degree of abnormality of each pixel point in the possible abnormality region, +.>The degree of abnormality for each connected domain.
Further, the enhancing the gray image according to the enhancement coefficient includes:
enhancement is performed using a linear gray scale enhancement algorithm.
Further, removing the background area by using a neural network semantic segmentation mode to obtain the gray image.
Drawings
FIG. 1 is a flow chart of the sheet metal part quality detection method of the application;
fig. 2 is a schematic diagram of an edge image determined by edge detection of a gray scale image according to the present application.
Detailed Description
The conception of the application is as follows:
the method comprises the steps of firstly preprocessing an image by utilizing an edge detection algorithm, and determining the size of an image dividing window according to a result obtained by edge detection. And further, the gray value of each pixel point in the image is obtained through Gaussian filtering processing on the whole sheet metal image under the condition of only illumination, the difference between the gray value and the corresponding illumination component in each area of the original image is compared, area classification is carried out, the abnormal degree is obtained by combining the position distribution of the connected area and the difference degree of the pixel points, the enhancement coefficient is obtained according to the gray difference and the abnormal degree, the image is adaptively enhanced, the edge detection is carried out on the enhanced image, and the quality detection of the sheet metal part is completed.
The following describes a sheet metal part quality detection method in detail with reference to the drawings and embodiments.
Method embodiment:
the embodiment of the sheet metal part quality detection method disclosed by the application has the following specific processes:
step one, acquiring a gray level image of the surface of a sheet metal part, and dividing the gray level image into different areas.
The collection device is placed above the production line, and the collection device is preferably a high-definition industrial camera to obtain the surface image of the produced sheet metal part. Because of the diversification of the background areas of the production line, the method adopts a neural network semantic segmentation mode to remove the background areas, and the semantic segmentation is the prior art, and is not described in detail here. And carrying out gray weighting treatment on the obtained sheet metal part surface image to obtain a gray level image of the sheet metal part surface.
The edge detection algorithm is used for the obtained gray image to obtain an edge image, and the edge image is obtained after the gray image is subjected to edge detection as shown in fig. 2.
Since different areas of the image need to be divided for subsequent analysis, the suitability of the division window size needs to be ensured. Too large a window does not allow complete analysis, and too small a window increases the amount of redundant computation.
And processing the obtained edge image by using a connected domain to obtain the length and width of the smallest circumscribed rectangle of the connected domain. And recording the minimum value of the length and the width obtained as H, recording H.H as the window size, and dividing the image according to the window size to obtain different areas.
Obtaining the illumination gray level difference value of the pixel point in each region according to the difference value of the illumination component of the gray level image and the pixel value of each pixel point on the gray level image, determining the illumination gray level difference degree of the region according to the illumination gray level difference value of the pixel point in each region, and determining whether the region belongs to an abnormal region or a possible abnormal region according to the illumination gray level difference degree of each region.
Because the scratch area and the normal area have gray value difference under the illumination and normal conditions, the areas can be classified according to the gray value difference.
The obtained sheet metal part gray level diagram is subjected to Gaussian filter processing to obtain an illumination component L, the Gaussian filter kernel is set to 3*3 and is all 1, and the size of the kernel can be adjusted according to specific scenes by an operator. Gaussian filtering is prior art and is not described here too much.
The pixel value of each point in the graph is subtracted from the illumination component obtained above, and the difference is processed in absolute terms because the difference will appear negative.
So far, the illumination gray scale difference value of each pixel point on the gray scale image is obtained.
Because the difference value of the illumination gray scale of the pixel points in the area is larger for the abnormal area, and smaller for the normal area.
The degree of difference in the illumination gray level for each region is:
in the formula ,indicating the degree of difference of the gray scale of the illumination of the kth region, -, etc.>Representing the variance of the illumination difference of each pixel point in the kth region,/for>Representing the difference of the illumination gray scale of the t pixel point in the k-th area,represents the maximum value in the difference degree of illumination gray scale of each region,/->Representing the minimum value in the degree of difference in the gray level of illumination of the respective areas,/->And the illumination gray scale difference value of the v pixel point in any one area is represented.
When an abnormality occurs in the area of the object,the value is larger, the ∈10>Approaching 1, when the area is a normal area,the value is small, and the value is->The value approaches 0.
Setting the threshold value of the difference degree of the illumination gray scale as D, recording the D as 0.7, optionally adjusting, marking the area larger than the threshold value as an abnormal area, and regarding the area smaller than the threshold value as a possible abnormal area, wherein the area possibly is a normal area or an abnormal area with insignificant difference of the illumination gray scale due to reflection of light, and further analysis is needed.
So far, the illumination gray level difference degree of each region is obtained, and the regions are classified.
And thirdly, acquiring the degree of abnormality of the pixel points in the possible abnormal area according to the position and direction characteristics of the pixel points in the possible abnormal area.
For the area smaller than the threshold value, it may be an area where minute scratches occur but the difference in gray value is small, or a normal area due to the influence of reflection of light upon illumination. Therefore, the partial region cannot be directly enhanced, and the position direction of the pixel point in the region needs to be combined with the approximation degree of the normal and abnormal regions for analysis.
Because scratches have unique two-dimensional characteristics, in general, scratches are continuous and distributed in a stripe shape. The images are in continuous straight lines in thin strips, and the directions in local areas are similar.
Firstly, the obtained possible abnormal areas smaller than the threshold value are analyzed, the position information of all abnormal areas in the graph is obtained, and for the tiny undetected scratch areas, the positions of the tiny undetected scratch areas are usually between the abnormal areas, and the directions of the abnormal areas on the two sides are similar.
For grey scale imageThe illumination gray scale difference value of each pixel point is normalized,in the drawingThe point with the maximum value is used as a starting point, 8 neighborhood pixel points are searched, and the 8 neighborhood pixel points are communicated with +.>And traversing all points in the graph to obtain all abnormal connected domains in the graph by the minimum difference value between the value and the starting point.
Calculating the gradient direction of each point on the edge of the connected domain of all the obtained abnormal connected domainsThe vertical direction corresponding to the gradient direction of each point is designated as the attention direction of the point, and is designated as +.>Record->Record->The direction of interest corresponding to the connected domain.
Record the midpoint of each communication domain asCalculating the direction vector of the midpoints of every two connected domains, wherein the vector direction is recorded asFor each connected domain, the degree of abnormality is:
wherein ,for the degree of abnormality of each connected domain +.>For the corresponding direction of interest of the connected domain, +.>Is the direction of the connected domain to the other connected domain. />The data is normalized.
For the possible abnormal region, the directions of the two connected domains near the region are similar, namely the direction of the midpoint of the two connected domains is similar to the attention direction of any connected domain, namelySmaller, P values approach 1.
The abnormal degree threshold of the connected domain is set, the abnormal degree threshold of the connected domain is preferably 0.7, the value of the abnormal degree threshold is optionally adjusted in other embodiments, and the connected domain smaller than the threshold is marked as the connected domain with no undetected area nearby and is not discussed later.
And acquiring all connected domains larger than a threshold value, extracting possible abnormal areas, through which connecting lines of central points of the connected domains pass, as areas to be detected, and further analyzing pixel points in the areas to be detected according to the gray level difference of the pixel points.
For the region to be detected, 8 neighborhood gray values of each point are obtained, and the obtained gray values are stored in a sequenceIn which the pixel value of the point is recorded as +.>The degree of abnormality for each point in the possible abnormality region:
wherein ,representing the degree of abnormality of each pixel point in the possible abnormality region, +.>Representing the variance of gray level difference between each pixel point and its neighborhood pixel point in the possible abnormal region, and describing the difference degree between the pixel point and the neighborhood region, +.>Gray value representing any one pixel point in the possible abnormal area, < >>And the gray value of the ith pixel point in the 8-neighbor of any pixel point in the possible abnormal area is represented. When the point is the point to be enhanced, the gray scale difference exists between the point and the neighborhood thereof,the larger; when the point is a normal point, the difference between the gray level of the point and the gray level of the neighborhood of the point is small, and the gray level of the neighborhood of the point is +.>The smaller.
Thus, the degree of abnormality of each pixel point in the possible abnormality region is obtained.
And step four, obtaining the enhancement coefficient of each pixel point in the possible abnormal region according to the illumination gray level difference degree of the possible abnormal region and the abnormal degree of each pixel point in the possible abnormal region.
The enhancement coefficient of each pixel point in the possible abnormal region is:
wherein ,for the enhancement coefficient of each pixel point in the possible abnormal area, +.>Representing the degree of abnormality of each pixel point in the possible abnormality region, +.>The degree of abnormality for each connected domain.
Thus, the enhancement coefficient of each pixel point in the possible abnormal area is obtained.
And fifthly, reinforcing the gray level image according to the reinforcing coefficient, and carrying out edge detection on the reinforced gray level image to finish the surface defect judgment of the sheet metal part.
Selectively enhancing the image according to the obtained enhancement coefficient, wherein the method uses a linear gray scale enhancement algorithm for enhancing, namelyIn which x refers to the input pixel value and y refers to the output pixel value. The value b is chosen based on conventional empirical values, 10 being taken in one embodiment of the application.
Because the enhanced image has obvious tiny scratches, the influence of illumination on the image is reduced, and therefore, a more complete scratch area can be obtained after the enhanced image is subjected to edge detection. According to the detected scratch area, the surface defect of the sheet metal part can be judged, and the quality detection of the sheet metal part is finished.
The above embodiments are only for illustrating the technical solution of the present application, and not for limiting the same; although the application has been described in detail with reference to the foregoing embodiments, it will be understood by those of ordinary skill in the art that: the technical scheme described in the foregoing embodiments can be modified or some technical features thereof can be replaced by equivalents; such modifications and substitutions do not depart from the spirit of the application and are intended to be included within the scope of the application.
Claims (10)
1. The sheet metal part quality detection method is characterized by comprising the following steps of:
acquiring a gray level image of the surface of the sheet metal part, and dividing the gray level image into different areas;
acquiring the illumination gray level difference value of the pixel point in each region according to the difference value of the illumination component of the gray level image and the pixel value of each pixel point on the gray level image, determining the illumination gray level difference degree of the region according to the illumination gray level difference value of the pixel point in each region, and determining whether the region belongs to an abnormal region or a possible abnormal region according to the illumination gray level difference degree of each region;
acquiring the degree of abnormality of the pixel points in the possible abnormal area according to the position and direction characteristics of the pixel points in the possible abnormal area;
obtaining an enhancement coefficient of each pixel point in the possible abnormal region according to the illumination gray level difference degree of the possible abnormal region and the abnormal degree of each pixel point in the possible abnormal region;
and reinforcing the gray level image according to the reinforcing coefficient, carrying out edge detection on the reinforced gray level image, and judging the surface defect of the sheet metal part.
2. The sheet metal part quality detection method according to claim 1, wherein the method for dividing the gray scale image into different areas is as follows:
and (3) using an edge detection algorithm to obtain an edge image of the obtained gray image, performing connected domain processing to the obtained edge image, obtaining the length and the width of a minimum circumscribed rectangle of the connected domain, using the smaller value of the length and the width as the length of a window so as to determine the size of the window, and dividing the gray image according to the size of the window to obtain different areas.
3. The sheet metal part quality detection method according to claim 1, wherein the method for obtaining the illumination gray scale difference value of the pixel point in each region according to the difference value between the illumination component of the gray scale image and the pixel value of each pixel point on the gray scale image is as follows:
and carrying out Gaussian filtering treatment on the gray level image to obtain an illumination component of the gray level image, calculating the absolute value of the difference value between the illumination component and the gray level value of each pixel point on the gray level image, and taking the absolute value as the illumination gray level difference value of the corresponding pixel point to obtain the illumination gray level difference value of the pixel point in each region.
4. The sheet metal part quality detection method according to claim 1, wherein the method for determining the illumination gray scale difference degree of each region according to the illumination gray scale difference value of the pixel point in the region comprises the following steps:
in the formula ,indicating the degree of difference of the gray scale of the illumination of the kth region, -, etc.>Representing the variance of the illumination difference of each pixel point in the kth region,/for>Representing the difference of the illumination gray scale of the t pixel point in the k region,/for the pixel point>Represents the maximum value in the difference degree of illumination gray scale of each region,/->Representing the minimum value in the degree of difference in the gray level of illumination of the respective areas,/->And the illumination gray scale difference value of the v pixel point in any one area is represented.
5. The sheet metal part quality detection method according to claim 4, wherein the method for determining whether the area belongs to an abnormal area or a possible abnormal area according to the illumination gray scale difference degree of each area is as follows:
and taking the area with the illumination gray level difference degree larger than the threshold value of the illumination gray level difference degree as an abnormal area, and taking the rest areas as possible abnormal areas.
6. The sheet metal part quality detection method according to claim 5, wherein the illumination gray scale difference degree threshold value is 0.7.
7. The sheet metal part quality detection method of claim 1, wherein the degree of abnormality of the pixel points in the possible abnormality region is:
wherein ,representing the degree of abnormality of each pixel point in the possible abnormality region, +.>Representing the variance of gray level difference between each pixel point and its neighborhood pixel point in the possible abnormal area, +.>Gray value representing any one pixel point in the possible abnormal area, < >>And the gray value of the ith pixel point in the 8-neighbor of any pixel point in the possible abnormal area is represented.
8. The sheet metal part quality detection method of claim 1, wherein the enhancement coefficient of each pixel point in the possible anomaly area is:
wherein ,for the enhancement coefficient of each pixel point in the possible abnormal area, +.>Representing the degree of abnormality of each pixel point in the possible abnormality region, +.>The degree of abnormality for each connected domain.
9. The sheet metal part quality detection method of claim 1, wherein the enhancing the grayscale image according to the enhancement factor comprises:
enhancement is performed using a linear gray scale enhancement algorithm.
10. The sheet metal part quality detection method according to claim 1, wherein the gray scale image is obtained by removing a background area by using a neural network semantic segmentation mode.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311040215.1A CN116777907A (en) | 2023-08-18 | 2023-08-18 | Sheet metal part quality detection method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311040215.1A CN116777907A (en) | 2023-08-18 | 2023-08-18 | Sheet metal part quality detection method |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN116777907A true CN116777907A (en) | 2023-09-19 |
Family
ID=88011910
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202311040215.1A Withdrawn CN116777907A (en) | 2023-08-18 | 2023-08-18 | Sheet metal part quality detection method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116777907A (en) |
Cited By (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116993722A (en) * | 2023-09-26 | 2023-11-03 | 山东奥斯登房车有限公司 | Visual detection method for defects of trailer body of trailer |
| CN117036354A (en) * | 2023-10-09 | 2023-11-10 | 深圳市魔样科技有限公司 | Intelligent finger ring display screen detection method |
| CN117095008A (en) * | 2023-10-20 | 2023-11-21 | 深圳市高进实业有限公司 | Intelligent detection method for defects of steel bar pipe of clock |
| CN117132844A (en) * | 2023-10-27 | 2023-11-28 | 江苏惠汕新能源集团有限公司 | Classifying method for cracks and scratches of photovoltaic panel based on image processing |
| CN117314894A (en) * | 2023-11-27 | 2023-12-29 | 深圳市金三维实业有限公司 | Method for rapidly detecting defects of watch bottom cover |
| CN117522863A (en) * | 2023-12-29 | 2024-02-06 | 临沂天耀箱包有限公司 | Integrated box body quality detection method based on image features |
| CN117764893A (en) * | 2024-01-08 | 2024-03-26 | 惠州市惠阳区兆顺益五金加工有限公司 | Numerical control machine tool wear identification method and system |
| CN117808800B (en) * | 2024-02-29 | 2024-05-10 | 深圳市富安娜艺术家居有限公司 | Intelligent assessment method and system for dyeing quality of textile |
-
2023
- 2023-08-18 CN CN202311040215.1A patent/CN116777907A/en not_active Withdrawn
Cited By (14)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116993722A (en) * | 2023-09-26 | 2023-11-03 | 山东奥斯登房车有限公司 | Visual detection method for defects of trailer body of trailer |
| CN116993722B (en) * | 2023-09-26 | 2023-12-08 | 山东奥斯登房车有限公司 | Visual detection method for defects of trailer body of trailer |
| CN117036354A (en) * | 2023-10-09 | 2023-11-10 | 深圳市魔样科技有限公司 | Intelligent finger ring display screen detection method |
| CN117036354B (en) * | 2023-10-09 | 2024-02-02 | 深圳市魔样科技有限公司 | Intelligent finger ring display screen detection method |
| CN117095008A (en) * | 2023-10-20 | 2023-11-21 | 深圳市高进实业有限公司 | Intelligent detection method for defects of steel bar pipe of clock |
| CN117095008B (en) * | 2023-10-20 | 2024-02-06 | 深圳市高进实业有限公司 | Intelligent detection method for defects of steel bar pipe of clock |
| CN117132844B (en) * | 2023-10-27 | 2024-01-23 | 江苏惠汕新能源集团有限公司 | Classifying method for cracks and scratches of photovoltaic panel based on image processing |
| CN117132844A (en) * | 2023-10-27 | 2023-11-28 | 江苏惠汕新能源集团有限公司 | Classifying method for cracks and scratches of photovoltaic panel based on image processing |
| CN117314894A (en) * | 2023-11-27 | 2023-12-29 | 深圳市金三维实业有限公司 | Method for rapidly detecting defects of watch bottom cover |
| CN117314894B (en) * | 2023-11-27 | 2024-03-29 | 深圳市金三维实业有限公司 | Method for rapidly detecting defects of watch bottom cover |
| CN117522863A (en) * | 2023-12-29 | 2024-02-06 | 临沂天耀箱包有限公司 | Integrated box body quality detection method based on image features |
| CN117522863B (en) * | 2023-12-29 | 2024-03-29 | 临沂天耀箱包有限公司 | Integrated box body quality detection method based on image features |
| CN117764893A (en) * | 2024-01-08 | 2024-03-26 | 惠州市惠阳区兆顺益五金加工有限公司 | Numerical control machine tool wear identification method and system |
| CN117808800B (en) * | 2024-02-29 | 2024-05-10 | 深圳市富安娜艺术家居有限公司 | Intelligent assessment method and system for dyeing quality of textile |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN116777907A (en) | Sheet metal part quality detection method | |
| CN106780486B (en) | Steel plate surface defect image extraction method | |
| CN111027546B (en) | Character segmentation method, device and computer readable storage medium | |
| CN115063409B (en) | Method and system for detecting surface material of mechanical cutter | |
| CN109685760B (en) | MATLAB-based SLM powder bed powder laying image convex hull depression defect detection method | |
| CN113298757A (en) | Metal surface defect detection method based on U-NET convolutional neural network | |
| CN115063430B (en) | Electric pipeline crack detection method based on image processing | |
| CN110706224B (en) | Optical element weak scratch detection method, system and device based on dark field image | |
| CN114926441B (en) | Defect detection method and system for injection molding part machining and forming | |
| CN114549981A (en) | Intelligent inspection pointer type instrument recognition and reading method based on deep learning | |
| CN112862770A (en) | Defect analysis and diagnosis system, method and device based on artificial intelligence | |
| CN111489337B (en) | Automatic optical detection pseudo defect removal method and system | |
| CN109239073B (en) | Surface defect detection method for automobile body | |
| CN115953373B (en) | Glass defect detection method, device, electronic equipment and storage medium | |
| CN117094975A (en) | Method and device for detecting surface defects of steel and electronic equipment | |
| CN115719332A (en) | Welding quality detection method | |
| CN114782329A (en) | Bearing defect damage degree evaluation method and system based on image processing | |
| CN114495098B (en) | Diaxing algae cell statistical method and system based on microscope image | |
| CN109886932A (en) | Gear ring of wheel speed sensor detection method of surface flaw based on SVM | |
| CN118279304B (en) | Abnormal recognition method, device and medium for special-shaped metal piece based on image processing | |
| CN116381053A (en) | Ultrasonic detection method and system for welding metal materials | |
| CN116071692A (en) | Morphological image processing-based water gauge water level identification method and system | |
| CN113129265B (en) | Method and device for detecting surface defects of ceramic tiles and storage medium | |
| Zhou et al. | An adaptive clustering method detecting the surface defects on linear guide rails | |
| CN115719326A (en) | PCB defect detection method and device |
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 | ||
| WW01 | Invention patent application withdrawn after publication | ||
| WW01 | Invention patent application withdrawn after publication |
Application publication date: 20230919 |