CN114693610A - Welding seam surface defect detection method, equipment and medium based on machine vision - Google Patents
Welding seam surface defect detection method, equipment and medium based on machine vision Download PDFInfo
- Publication number
- CN114693610A CN114693610A CN202210250926.0A CN202210250926A CN114693610A CN 114693610 A CN114693610 A CN 114693610A CN 202210250926 A CN202210250926 A CN 202210250926A CN 114693610 A CN114693610 A CN 114693610A
- Authority
- CN
- China
- Prior art keywords
- image
- welding seam
- weld
- surface defect
- defect
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
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
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/23—Clustering techniques
- G06F18/232—Non-hierarchical techniques
- G06F18/2321—Non-hierarchical techniques using statistics or function optimisation, e.g. modelling of probability density functions
- G06F18/23213—Non-hierarchical techniques using statistics or function optimisation, e.g. modelling of probability density functions with fixed number of clusters, e.g. K-means clustering
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/241—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches
- G06F18/2411—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches based on the proximity to a decision surface, e.g. support vector machines
-
- 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
- G06T2207/30152—Solder
Abstract
The invention discloses a method, equipment and a medium for detecting weld surface defects based on machine vision, wherein the method comprises the following steps: acquiring a welding seam image of the boiler pipeline acquired by an industrial camera, and preprocessing the image: image restoration, image denoising, image enhancement and image rotation correction; carrying out binarization processing on the image obtained by preprocessing, and then carrying out morphological processing; intercepting an ROI (region of interest) in an image obtained after morphological processing by determining the positions of the edge and the central line of a welding seam; intercepting an ROI (region of interest) in an image obtained after morphological processing by determining the positions of the edge and the central line of a welding seam; extracting characteristic information from the ROI area image; inputting the characteristic information into a pre-trained welding seam surface defect detection model for detection, and outputting to obtain the type of the welding seam surface defect; the structure of the weld surface defect detection model adopts a biased binary tree support vector machine. The method improves the accuracy and efficiency of identifying the weld surface defects.
Description
Technical Field
The invention belongs to the technical field of machine vision, and particularly relates to a method, equipment and medium for detecting weld surface defects based on machine vision.
Background
The surface defect detection of the welding line of the thermal pipeline in the furnace in China mainly depends on the most original manual regular detection method, so that the phenomena of false detection, missing detection and the like are often caused, and workers can easily get occupational diseases when working in a dust-diffused environment for a long time. Machine vision is a contactless, non-destructive automatic detection technology, has advantages such as safe and reliable, can work for a long time under adverse circumstances and detection efficiency height, and at present, there are many methods that utilize machine vision to detect weld defect, for example neural network algorithm and support vector machine algorithm, and through the debugging discovery to defect identification model: the effective extraction of the characteristic region directly influences the identification time and accuracy of the model.
Therefore, the region extraction is carried out on the collected welding seam image, and the removal of irrelevant backgrounds is very important.
The patent number is as follows: CN201810709776.9 discloses a weld defect identification method based on an improved convolutional neural network, but the method does not include weld region extraction, only can manually cut the acquired image, and then completes defect identification by using the convolutional neural network, so that the working efficiency is extremely low. CN201711043638.3 discloses a ship plate surface defect detection method based on machine vision, which adopts a global threshold segmentation method to extract an interested region in the preprocessed ship plate image to be detected, so as to obtain a plurality of sub-regions, but the internal light of the boiler is dark, dust is diffused, the pipeline weld region cannot be effectively identified by the global threshold segmentation method, and the weld region is extracted from the image alone, the patent number is: CN201910396121.5 discloses a method and a system for extracting a cylindrical longitudinal weld characteristic region and tracking a weld based on structured light, in which two lasers are used to generate laser stripes on two sides, a line projection operation is performed on pixels of a weld image to find a line with the smallest pixel point, and a line with a fixed width h is selected upwards and downwards to finally realize tracking of the weld region, and the method has two disadvantages: firstly, only local tracking of a welding seam region can be realized, and the whole extraction of the welding seam region cannot be carried out; secondly, the method carries out region extraction through the minimum value of the pixel, and due to more dust in the boiler, when a pipeline image is shot, a plurality of 'snowflakes' appear, so that the region corresponding to the minimum value of the pixel is not the pipeline welding seam region.
In summary, the existing device and method cannot accurately identify and extract the weld characteristic region, and therefore the accuracy of detecting the weld surface defects is limited.
Disclosure of Invention
In order to solve the problem of accuracy of detection of weld surface defects in the prior art, the invention provides a method, equipment and a medium for detecting weld surface defects based on machine vision, wherein the method of dividing and extracting weld areas is realized by utilizing a gray value accumulation and averaging method, then the detection of the defects is completed by extracting the characteristics of the weld areas, the workload of workers for detecting the weld defects of pipelines in a furnace is finally reduced, and the detection efficiency is improved.
In order to achieve the technical purpose, the invention adopts the following technical scheme:
a weld surface defect detection method based on machine vision comprises the following steps:
Further, carrying out image restoration by adopting wiener filtering; carrying out image denoising by adopting self-adaptive mean filtering; and performing image enhancement by adopting a Retinex method.
Further, adopt hough transform to detect the straight line to calculate welding seam inclination angle theta, carry out the rotatory correction of image based on welding seam inclination angle theta, specifically do:
step A1, utilizing hough straight line detection to identify all straight lines in the welding line image;
step A2, calculating the lengths of all straight lines, and eliminating straight lines with the lengths lower than a preset length value;
step A3, calculating slope average values K of all the remaining straight lines, and calculating corresponding inclination angles theta based on the slope average values K;
and step A4, performing rotation correction on the welding seam image according to the inclination angle theta.
Further, the method for intercepting the ROI area comprises the following steps:
step B1, averaging the gray scale of each line of the image obtained by morphological processing;
step B2, at [ delta, L ] th of image2-δ]Traversing and searching the minimum value C of all the gray level average values within the line interval range1Recording the x-th of the corresponding behavior image1A row; wherein L is2The total number of rows of pixel points of the image;
step B3, at the fourth step of the imageAndtraversing and searching places in the range of line intervalMinimum value C of mean value of gray scale2Recording the x-th image of the corresponding behavior2A row; wherein L is0The number of lines of pixel points occupied by the width of the welding line is a preset empirical value;
step B4, determining the action x of the center line of the welding seam0=(x1+x2) And/2, the rows of the upper edge and the lower edge of the welding seam are respectively as follows: [ x ] of0-L0,x0+L0];
Step B5, welding the region [ x ] between the upper edge and the lower edge of the welding seam0-L0,x0+L0]And intercepting the weld image to obtain an ROI area image.
Further, the feature information extracted from the ROI region image includes: image texture features, the location and area of the defect, the perimeter, rectangularity, and circularity of the defect outline.
Further, the defect types include: undercuts, pores and cracks.
Further, the method for identifying the type of the weld surface defect by adopting the weld surface defect detection model with the structure of a partial binary tree support vector machine comprises the following steps:
inputting the texture features of the image into a first support vector machine SVM1 to judge whether the weld has defects;
inputting the positions and areas of the defects into a second support vector machine SVM2, and judging whether the seam has undercut defects;
inputting the circularity and the rectangularity of the defect outline into a third support vector machine SVM3, and judging whether the weld joint has air hole defects or not;
and inputting the position and the area of the defect and the perimeter, the circularity and the rectangularity of the defect outline into a fourth support vector machine SVM4, and judging whether the weld has a crack defect.
An electronic device comprising a memory and a processor, the memory having stored therein a computer program that, when executed by the processor, causes the processor to implement the method of any of the above techniques.
A computer-readable storage medium, on which a computer program is stored which, when executed by a processor, implements the method of any of the above techniques.
Compared with the prior art, the invention has the following advantages:
1. the method has reliable detection results, can identify various defects, and has the accuracy rate of over 90 percent through experiments;
2. the method only needs 40ms from data acquisition to data output, and greatly improves the detection efficiency while ensuring the accuracy of the detection result.
Drawings
FIG. 1 is a flowchart illustrating weld defect identification according to an embodiment of the present invention;
FIG. 2 is a schematic view of a weld joint after reinforcement according to an embodiment of the present invention;
FIG. 3 is a schematic diagram of hough linear detection according to an embodiment of the present invention;
FIG. 4 is a schematic view of a weld after rotation according to an embodiment of the present invention;
FIG. 5 is a schematic diagram of averaging after gray scale accumulation according to an embodiment of the present invention;
FIG. 6 is a schematic view of weld edge location determination according to an embodiment of the present invention;
FIG. 7 is a schematic diagram of the ROI area truncation according to the embodiment of the invention.
FIG. 8 is a diagram illustrating a defect recognition result according to an embodiment of the present invention;
Detailed Description
The following describes embodiments of the present invention in detail, which are developed based on the technical solutions of the present invention, and give detailed implementation manners and specific operation procedures to further explain the technical solutions of the present invention.
In an embodiment of the present invention, a method for detecting weld surface defects based on machine vision is provided, and as shown in fig. 1, the method includes the following steps:
wherein, the wiener filtering is adopted to carry out image restoration; carrying out image denoising by adopting self-adaptive mean filtering; the Retinex method is adopted for image enhancement, and the enhanced weld image is shown in FIG. 2.
Adopt hough transform to detect the straight line to calculate welding seam inclination theta, carry out the rotatory correction of image based on welding seam inclination theta, specifically do:
step A1, identifying all straight lines in the weld image by hough straight line detection, as shown in FIG. 3;
step A2, calculating the lengths of all straight lines, and eliminating straight lines with the lengths lower than a preset length value;
step A3, calculating slope average values K of all the remaining straight lines, and calculating corresponding inclination angles theta based on the slope average values K;
and step A4, taking the circular point as the center, and rotationally correcting the welding seam image according to the inclination angle theta. The rotated weld image is shown in fig. 4.
The preset length value in this embodiment may be preset as an empirical ratio value of the image size width, for example, when the original image length is 200mm, and the experience ratio is 50%, the preset length value is 80 mm.
And 2, carrying out binarization processing on the image obtained by preprocessing, and then carrying out morphological processing.
Specifically, a K-means clustering method can be adopted to classify the foreground and the background in the image so as to realize image binarization; the morphological processing specifically adopts a first-open operation and a second-close operation.
step B1, averaging the gray levels of each line of the image obtained by the morphological processing, the result is shown in fig. 5;
step B2, at [ delta, L ] th of image2-δ]Traversing and searching the minimum value C of all the gray level average values within the line interval range1Recording the x-th of the corresponding behavior image1A row; wherein L is2The total number of rows of pixel points of the image;
step B3, inFirst of the imageAndtraversing and searching the minimum value C of all the gray level average values within the line interval range2Recording the x-th image of the corresponding behavior2Rows, as shown in FIG. 6; wherein L is0The number of the pixel point lines occupied by the width of the welding line is a preset empirical value; where δ is a predetermined estimate of the weld edge background, which in this embodiment is predetermined to be 50.
Step B4, determining the action x of the center line of the welding seam0=(x1+x2) And/2, the rows of the upper edge and the lower edge of the welding seam are respectively as follows: [ x ] of0-L0,x0+L0];
Step B5, welding the region [ x ] between the upper edge and the lower edge of the welding seam0-L0,x0+L0]And (5) intercepting the weld image to obtain an ROI area image, as shown in figure 7.
After the image preprocessing and the ROI extraction, the obtained image pixel points are reduced, and the subsequent defect type identification efficiency can be realized. And an irrelevant background area is removed, only a welding seam area is reserved, and subsequently extracted characteristic information is more accurate, so that the accuracy of defect type identification is improved.
Extracting image texture features by using an LBP operator, extracting the area and the contour perimeter of a weld defect by using ContourAlea and ContourPerimeter functions respectively, extracting defect position information by using an image moment, and extracting the rectangularity D and the circularity C of the defect by using the following formulas:
in the formula, AWThe minimum area of the circumscribed rectangle with the defect is A, and the area of the defect is A; p is the perimeter of the defective region, C is greater than 1 if the defective region is circular, and C is between (0,1) if it is other shapes.
The types of weld surface defects involved in this example include: undercuts, air voids and cracks.
The invention relates to a biased binary tree support vector machine which is composed of a plurality of support vector machines, wherein the implementation of the invention is to arrange 4 support vector machines aiming at the defect type of the surface of a welding seam.
As shown in fig. 1, the input of the first support vector machine is the image texture feature, and the output is the label indicating whether the weld has defects; the input of the second support vector machine is the position and the area of the defect, and the output is a label which indicates whether the seam has undercut defects; the input of the third support vector machine is the circularity and the rectangularity of the defect outline, and the output is a label indicating whether the welding seam has the air hole defect; the input of the fourth support vector machine is the position and the area of the defect and the perimeter, the circularity and the rectangularity of the outline of the defect, and the output is a label which indicates whether the welding seam has a crack defect.
And training the above binary tree support vector machine by using a training sample with a label to obtain a weld surface defect detection model, namely performing defect detection on the weld surface with unknown defect type and unknown defect type. In this embodiment, no matter whether the foregoing support vector machines detect defects or detect corresponding defects, other defect detections are continuously performed according to corresponding feature information using the remaining support vector machines, so that each support vector machine correspondingly outputs 1 detection result, and by combining the 4 detection results, it is finally determined whether some defects exist on the surface of the weld joint at the same time. The example of the recognition result and the real value pair ratio is shown in fig. 8.
The invention further provides an electronic device, which comprises a memory and a processor, wherein the memory stores a computer program, and when the computer program is executed by the processor, the processor is enabled to realize the weld surface defect detection method based on the machine vision.
The present invention also provides a computer-readable storage medium, on which a computer program is stored, which, when executed by a processor, implements the machine vision-based weld surface defect detection method described in the above embodiment.
The above embodiments are preferred embodiments of the present application, and those skilled in the art can make various changes or modifications without departing from the general concept of the present application, and such changes or modifications should fall within the scope of the claims of the present application.
Claims (9)
1. A weld joint surface defect detection method based on machine vision is characterized by comprising the following steps:
step 1, acquiring a welding seam image of a boiler pipeline collected by an industrial camera, and preprocessing the image, wherein the preprocessing mainly comprises the following steps: image restoration, image denoising, image enhancement and image rotation correction;
step 2, carrying out binarization processing on the image obtained by preprocessing, and then carrying out morphological processing; intercepting an ROI (region of interest) in an image obtained after morphological processing by determining the positions of the edge and the central line of a welding seam;
step 3, intercepting an ROI (region of interest) in an image obtained after morphological processing by determining the positions of the edge and the central line of the welding seam;
step 4, extracting characteristic information from the ROI area image;
step 5, inputting the characteristic information into a pre-trained welding seam surface defect detection model for detection, and outputting to obtain the type of the welding seam surface defect; the structure of the pre-trained weld surface defect detection model is obtained by adopting a biased binary tree support vector machine and training by using a characteristic information sample with a defect type label.
2. The method for detecting the defects on the surface of the welding seam based on the machine vision is characterized in that the wiener filtering is adopted for image restoration; carrying out image denoising by adopting self-adaptive mean filtering; and performing image enhancement by adopting a Retinex method.
3. The weld joint surface defect detection method based on machine vision according to claim 1, characterized in that a hough transformation is adopted to detect straight lines, and a weld joint inclination angle θ is calculated, and image rotation correction is performed based on the weld joint inclination angle θ, specifically:
step A1, utilizing hough straight line detection to identify all straight lines in a welding line image;
step A2, calculating the lengths of all straight lines, and eliminating straight lines with the lengths lower than a preset length value;
step A3, calculating slope average values K of all the remaining straight lines, and calculating corresponding inclination angles theta based on the slope average values K;
and step A4, performing rotation correction on the welding seam image according to the inclination angle theta.
4. The weld joint surface defect detection method based on the machine vision is characterized in that the ROI area is intercepted by the method comprising the following steps:
step B1, averaging the gray scale of each line of the image obtained by morphological processing;
step B2, at [ delta, L ] th of image2-δ]In the line interval range, the minimum value C of all the gray level average values is searched in a traversing way1Recording the x-th of the corresponding behavior image1A row; wherein L is2The total number of rows of pixel points of the image;
step B3, at the fourth step of the imageAndtraversing and searching the minimum value C of all the gray level average values within the line interval range2Recording the x-th image of the corresponding behavior2A row; wherein L is0The number of lines of pixel points occupied by the width of the welding line is a preset empirical value;
step B4, determining the action x of the center line of the welding seam0=(x1+x2) And/2, the rows of the upper edge and the lower edge of the welding seam are respectively as follows: [ x ] of0-L0,x0+L0];
Step B5, welding the region [ x ] between the upper edge and the lower edge of the welding seam0-L0,x0+L0]And intercepting the weld image to obtain an ROI area image.
5. The method for detecting the defects of the weld joint surface based on the machine vision according to claim 1, wherein the characteristic information extracted from the ROI area image comprises: image texture features, the position and area of the defect, the perimeter, rectangularity and circularity of the defect outline.
6. The machine vision-based weld surface defect detection method according to claim 1, wherein the defect types include: undercuts, pores and cracks.
7. The weld joint surface defect detection method based on the machine vision is characterized in that a weld joint surface defect detection model with a structure of a biased binary tree support vector machine is adopted, and the method for identifying the type of the weld joint surface defect comprises the following steps:
inputting the texture features of the image into a first support vector machine SVM1 to judge whether the weld has defects;
inputting the positions and areas of the defects into a second support vector machine SVM2, and judging whether the seam has undercut defects;
inputting the circularity and the rectangularity of the defect outline into a third support vector machine SVM3, and judging whether the welding seam has air hole defects or not;
and inputting the position and the area of the defect and the perimeter, the circularity and the rectangularity of the defect outline into a fourth support vector machine SVM4, and judging whether the weld has a crack defect.
8. An electronic device comprising a memory and a processor, the memory having stored therein a computer program, wherein the computer program, when executed by the processor, causes the processor to implement the method of any of claims 1-7.
9. A computer-readable storage medium, on which a computer program is stored, which, when being executed by a processor, carries out the method according to any one of claims 1 to 7.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202210250926.0A CN114693610A (en) | 2022-03-15 | 2022-03-15 | Welding seam surface defect detection method, equipment and medium based on machine vision |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202210250926.0A CN114693610A (en) | 2022-03-15 | 2022-03-15 | Welding seam surface defect detection method, equipment and medium based on machine vision |
Publications (1)
Publication Number | Publication Date |
---|---|
CN114693610A true CN114693610A (en) | 2022-07-01 |
Family
ID=82138683
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202210250926.0A Pending CN114693610A (en) | 2022-03-15 | 2022-03-15 | Welding seam surface defect detection method, equipment and medium based on machine vision |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN114693610A (en) |
Cited By (8)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN115861315A (en) * | 2023-02-27 | 2023-03-28 | 常州微亿智造科技有限公司 | Defect detection method and device |
CN115861357A (en) * | 2023-02-27 | 2023-03-28 | 常州微亿智造科技有限公司 | Workpiece transition edge detection method based on K-means clustering and point location planning |
CN115880280A (en) * | 2023-02-01 | 2023-03-31 | 山东建筑大学鉴定检测中心有限公司 | Detection method for quality of steel structure weld joint |
CN115953366A (en) * | 2022-12-14 | 2023-04-11 | 广州市斯睿特智能科技有限公司 | Weld joint detection method, system and device based on reflection image and storage medium |
CN116777918A (en) * | 2023-08-25 | 2023-09-19 | 苏州科尔珀恩机械科技有限公司 | Visual auxiliary kiln surface defect rapid detection method |
CN116958137A (en) * | 2023-09-19 | 2023-10-27 | 易启科技(吉林省)有限公司 | New energy battery surface defect detection method and device |
CN117058144A (en) * | 2023-10-12 | 2023-11-14 | 南昌理工学院 | Defect identification method and system for weld joint image |
CN117576106A (en) * | 2024-01-17 | 2024-02-20 | 高科建材(咸阳)管道科技有限公司 | Pipeline defect detection method and system |
-
2022
- 2022-03-15 CN CN202210250926.0A patent/CN114693610A/en active Pending
Cited By (13)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN115953366A (en) * | 2022-12-14 | 2023-04-11 | 广州市斯睿特智能科技有限公司 | Weld joint detection method, system and device based on reflection image and storage medium |
CN115880280A (en) * | 2023-02-01 | 2023-03-31 | 山东建筑大学鉴定检测中心有限公司 | Detection method for quality of steel structure weld joint |
CN115861357A (en) * | 2023-02-27 | 2023-03-28 | 常州微亿智造科技有限公司 | Workpiece transition edge detection method based on K-means clustering and point location planning |
CN115861357B (en) * | 2023-02-27 | 2023-06-20 | 常州微亿智造科技有限公司 | Workpiece transition edge detection method based on K-means clustering and point location planning |
CN115861315A (en) * | 2023-02-27 | 2023-03-28 | 常州微亿智造科技有限公司 | Defect detection method and device |
CN116777918B (en) * | 2023-08-25 | 2023-10-31 | 苏州科尔珀恩机械科技有限公司 | Visual auxiliary kiln surface defect rapid detection method |
CN116777918A (en) * | 2023-08-25 | 2023-09-19 | 苏州科尔珀恩机械科技有限公司 | Visual auxiliary kiln surface defect rapid detection method |
CN116958137A (en) * | 2023-09-19 | 2023-10-27 | 易启科技(吉林省)有限公司 | New energy battery surface defect detection method and device |
CN116958137B (en) * | 2023-09-19 | 2023-12-12 | 易启科技(吉林省)有限公司 | New energy battery surface defect detection method and device |
CN117058144A (en) * | 2023-10-12 | 2023-11-14 | 南昌理工学院 | Defect identification method and system for weld joint image |
CN117058144B (en) * | 2023-10-12 | 2023-12-19 | 南昌理工学院 | Defect identification method and system for weld joint image |
CN117576106A (en) * | 2024-01-17 | 2024-02-20 | 高科建材(咸阳)管道科技有限公司 | Pipeline defect detection method and system |
CN117576106B (en) * | 2024-01-17 | 2024-03-29 | 高科建材(咸阳)管道科技有限公司 | Pipeline defect detection method and system |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN114693610A (en) | Welding seam surface defect detection method, equipment and medium based on machine vision | |
CN110163853B (en) | Edge defect detection method | |
CN105784713B (en) | Sealing ring detection method of surface flaw based on machine vision | |
CN109658402B (en) | Automatic detection method for geometric dimension of industrial profile based on computer vision imaging | |
CN110930357B (en) | In-service steel wire rope surface defect detection method and system based on deep learning | |
CN112598001A (en) | Automatic ship water gauge reading identification method based on multi-model fusion | |
CN111369516A (en) | Transformer bushing heating defect detection method based on infrared image recognition | |
CN112233116B (en) | Concave-convex mark visual detection method based on neighborhood decision and gray level co-occurrence matrix description | |
CN112037203A (en) | Side surface defect detection method and system based on complex workpiece outer contour registration | |
CN113077437B (en) | Workpiece quality detection method and system | |
CN109376740A (en) | A kind of water gauge reading detection method based on video | |
CN108509950B (en) | Railway contact net support number plate detection and identification method based on probability feature weighted fusion | |
CN113643276B (en) | Textile texture defect automatic detection method based on statistical analysis | |
CN109584215A (en) | A kind of online vision detection system of circuit board | |
CN113393426A (en) | Method for detecting surface defects of rolled steel plate | |
CN112734729B (en) | Water gauge water level line image detection method and device suitable for night light supplement condition and storage medium | |
CN114495098B (en) | Diaxing algae cell statistical method and system based on microscope image | |
CN115100191A (en) | Metal casting defect identification method based on industrial detection | |
CN116128873A (en) | Bearing retainer detection method, device and medium based on image recognition | |
CN116468687A (en) | Scratch defect detection method and device, storage medium and electronic equipment | |
CN114049316A (en) | Steel wire rope defect detection method based on metallic luster area | |
CN110288040B (en) | Image similarity judging method and device based on topology verification | |
CN113240623A (en) | Pavement disease detection method and device | |
CN110426395B (en) | Method and device for detecting surface of solar EL battery silicon wafer | |
CN115439523A (en) | Method and equipment for detecting pin size of semiconductor device and storage medium |
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 |