CN113077416A - Welding spot welding defect detection method and system based on image processing - Google Patents
Welding spot welding defect detection method and system based on image processing Download PDFInfo
- Publication number
- CN113077416A CN113077416A CN202110268941.3A CN202110268941A CN113077416A CN 113077416 A CN113077416 A CN 113077416A CN 202110268941 A CN202110268941 A CN 202110268941A CN 113077416 A CN113077416 A CN 113077416A
- Authority
- CN
- China
- Prior art keywords
- welding
- defect
- image
- acquiring
- cnn
- 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- Withdrawn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
- 230000007547 defect Effects 0.000 title claims abstract description 121
- 238000003466 welding Methods 0.000 title claims abstract description 121
- 238000001514 detection method Methods 0.000 title claims abstract description 45
- 238000012545 processing Methods 0.000 title claims abstract description 22
- 238000013527 convolutional neural network Methods 0.000 claims abstract description 68
- 238000007781 pre-processing Methods 0.000 claims abstract description 21
- 238000000034 method Methods 0.000 claims abstract description 17
- 238000010276 construction Methods 0.000 claims description 5
- 238000002372 labelling Methods 0.000 claims description 3
- 238000012549 training Methods 0.000 abstract description 7
- 238000010586 diagram Methods 0.000 description 2
- 238000007790 scraping Methods 0.000 description 2
- 238000011179 visual inspection Methods 0.000 description 2
- 240000007651 Rubus glaucus Species 0.000 description 1
- 235000011034 Rubus glaucus Nutrition 0.000 description 1
- 235000009122 Rubus idaeus Nutrition 0.000 description 1
- 238000007792 addition Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000008094 contradictory effect Effects 0.000 description 1
- 230000002950 deficient Effects 0.000 description 1
- 238000000605 extraction Methods 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 238000005457 optimization Methods 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
- 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
- G06N3/045—Combinations of networks
-
- 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
- G06N3/08—Learning methods
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/60—Analysis of geometric attributes
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/10—Image acquisition modality
- G06T2207/10004—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
- G06T2207/20—Special algorithmic details
- G06T2207/20212—Image combination
-
- 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
- G06T2207/30141—Printed circuit board [PCB]
Abstract
The invention discloses a welding spot welding defect detection method and system based on image processing, which relate to the field of electronic detection and mainly comprise the following steps: acquiring and preprocessing an image sample set of a circuit board with welding spot welding defects; selecting and extracting welding defect characteristics in an image sample set through a convolutional neural network, and generating a detection model; acquiring a complete image of the circuit board, and calling a detection model to acquire a welding defect boundary frame of the complete image; and generating and outputting a marking picture according to the welding defect boundary frame. The method performs convolutional neural network training on the sample set by utilizing the CascadeR-CNN algorithm, so that the extracted features are more consistent with the features of actual welding spot defects by utilizing the algorithm, and meanwhile, the method can detect welding defects of more types of patch elements.
Description
Technical Field
The invention relates to the field of electronic detection, in particular to a method and a system for detecting welding spot welding defects based on image processing.
Background
The optical automatic detection steps of the quality of the circuit board element put out in the market at present are complex, mainly based on a defect detection method based on reference, statistical modeling needs to be carried out again every time a new product model is detected, a template is built by carrying out statistical learning on a sample which is artificially inspected to be qualified, and then automatic detection is carried out. Meanwhile, the related software design is not complete enough (for example, automatic detection cannot clearly distinguish electrode needle marks, scraping and pollution). Currently, automatic detection is only used for reducing the workload of visual inspection, and cannot completely replace manual visual inspection. Therefore, developing a set of low-cost and high-precision automatic circuit board welding spot detection method is an important link for realizing the industrial automation of electronic manufacturing, and has great market value and social value.
Disclosure of Invention
In order to solve the problems, improve the universality of the detection of the welding points of the circuit board and reduce the labor cost, the invention provides a welding point welding defect detection method based on image processing, which comprises the following steps:
s1: acquiring and preprocessing an image sample set of a circuit board with welding spot welding defects;
s2: selecting and extracting welding defect characteristics in an image sample set through a convolutional neural network, and generating a detection model;
s3: acquiring a complete image of the circuit board, and calling a detection model to acquire a welding defect boundary frame of the complete image;
s4: and generating and outputting a marking picture according to the welding defect boundary frame.
Further, in step S1, the preprocessing includes: adjusting the size of the image sample to a preset size, and randomly turning and normalizing.
Further, in step S2, the convolutional neural network selects the Cascade R-CNN algorithm to train the image sample set, and the specific steps are as follows:
and acquiring an output result of suspected welding defect characteristics in each layer of R-CNN network of the image sample, taking the output result as the input of the next-level R-CNN network, and finally acquiring a boundary box of element welding characteristics, wherein the R-CNN networks at all levels are cascaded based on respective IoU threshold values.
Further, in step S3, the detection model is a model that determines the welding defects in two stages by using a Cascade R-CNN algorithm, where the two stages are specifically:
acquiring a defect preliminary judgment area in the complete image by using a Cascade R-CNN algorithm according to the welding defect characteristics;
and substituting the defect primary judgment area into a Cascade R-CNN algorithm again, and carrying out secondary judgment on the defect primary judgment area according to the welding defect characteristics to obtain a welding defect boundary frame.
Further, in step 3, the acquiring of the complete image specifically includes the steps of:
and acquiring local images with preset sizes of all parts on the circuit board, and splicing the local images into a complete image according to the sequence of all parts on the circuit board.
The invention also provides a welding spot welding defect detection system based on image processing, which comprises:
the preprocessing unit is used for acquiring and preprocessing an image sample set of the circuit board with the welding defect of the welding spot;
the model construction unit is used for selecting and extracting welding defect characteristics in the image sample set through a convolutional neural network and generating a detection model;
the defect judging unit is used for calling the detection model to extract a welding defect boundary frame in the complete image;
and the output unit is used for generating and outputting a labeling picture according to the welding defect boundary frame.
Further, the preprocessing unit, preprocessing, includes: adjusting the size of the image sample to a preset size, and randomly turning and normalizing.
Further, in the model construction unit, the convolutional neural network selects a Cascade R-CNN algorithm to train the image sample set, and the specific mode is as follows:
and acquiring an output result of suspected welding defect characteristics in each layer of R-CNN network of the image sample, taking the output result as the input of the next-level R-CNN network, and finally acquiring a boundary box of element welding characteristics, wherein the R-CNN networks at all levels are cascaded based on respective IoU threshold values.
Further, the defect judgment unit judges the welding defect in two stages by the detection model through a Cascade R-CNN algorithm, and the specific stages are as follows:
acquiring a defect preliminary judgment area in the complete image by using a Cascade R-CNN algorithm according to the welding defect characteristics;
and substituting the defect primary judgment area into a Cascade R-CNN algorithm again, and carrying out secondary judgment on the defect primary judgment area according to the welding defect characteristics to obtain a welding defect boundary frame.
The device further comprises an image acquisition unit, which is used for acquiring partial images with preset sizes of all parts on the circuit board and splicing the partial images into a complete image according to the sequence of all parts on the circuit board.
Compared with the prior art, the invention at least has the following beneficial effects:
(1) according to the method and the system for detecting the welding spot welding defects based on the image processing, the Cascade R-CNN algorithm is utilized to carry out convolutional neural network training on the sample set, the extracted characteristics are enabled to be more consistent with the characteristics of the actual welding spot defects by utilizing the algorithm, and meanwhile, the welding defects of more types of patch elements can be detected;
(2) in the process of extracting the defect boundary frame, the Cascade R-CNN algorithm is used again to obtain a defect primary judgment area, secondary judgment is carried out to reduce the area, so that the finally obtained defect boundary frame is more accurate in position and smaller in range, and a user is helped to accurately judge a fault occurrence point;
(3) the complete circuit board image is obtained in an image splicing mode, manual coordinate calibration is not needed, and meanwhile, the original data set is expanded through image preprocessing, so that defect characteristics obtained during convolutional neural network training are more sufficient.
Drawings
FIG. 1 is a method step diagram of a method and system for detecting weld defects of a weld spot based on image processing;
FIG. 2 is a system diagram of a method and system for detecting welding defects of a welding spot based on image processing.
Detailed Description
The following are specific embodiments of the present invention and are further described with reference to the drawings, but the present invention is not limited to these embodiments.
Example one
In order to improve the universality of the detection of the welding spot of the circuit board, reduce the labor cost, better identify the welding spot defect, distinguish electrode pin marks, scraping and pollution and reduce the dependence on human judgment, as shown in figure 1, the invention provides a welding spot welding defect detection method based on image processing, which mainly comprises the following steps:
s1: acquiring and preprocessing an image sample set of a circuit board with welding spot welding defects;
s2: selecting and extracting welding defect characteristics in an image sample set through a convolutional neural network, and generating a detection model;
s3: acquiring a complete image of the circuit board, and calling a detection model to acquire a welding defect boundary frame of the complete image;
s4: and generating and outputting a marking picture according to the welding defect boundary frame.
Before the circuit board is detected, a model needs to be established, and the accuracy of the model reaches the standard through training. It should be noted that Cascade R-CNN is based on fast-RCNN, and multiple networks of R-CNN are cascaded based on different IoU thresholds, so as to achieve continuous optimization of detection results. Different from the common cascade, the output result of the former R-CNN network can be used as the input of the latter R-CNN network, and the IoU threshold values for defining positive and negative samples are higher in the detection model in the future. The invention has the following concrete characteristics:
and acquiring an output result of suspected welding defect characteristics in each layer of R-CNN network of the image sample, taking the output result as the input of the next-level R-CNN network, and finally acquiring a boundary box of element welding characteristics, wherein the R-CNN networks at all levels are cascaded based on respective IoU threshold values.
Besides reducing the requirement on the number of samples through a Cascade R-CNN algorithm, the method reasonably expands the number of samples through a preprocessing means by adjusting the size of the image sample to a preset size (self-set), randomly turning and normalizing, so that the trained model can adapt to more types of feature extraction.
After the model is built, the model needs to be verified, and considering that the traditional method obtains images of each part in a form of manually calibrating coordinates and judges the images of each part one by one, in order to improve the judging speed, the invention selects to judge the welding defects of the whole image, and in order to realize the judgment, the concrete steps of obtaining the complete image are as follows:
and acquiring local images with preset sizes of all parts on the circuit board, and splicing the local images into a complete image according to the sequence of all parts on the circuit board.
Therefore, the image of the whole circuit board can be processed in the process of judging the welding defects once, and the judging efficiency is greatly improved.
And the subsequent processing of the image by using the detection model utilizes the characteristics of the Cascade R-CNN algorithm: training architecture for a two-stage system. For an original image, the first stage is to extract sub-nets (H0) from candidate areas (possibly defective spots) of the entire image to generate an initial detection hypothesis, called a goal proposal. In the second phase, these assume that the input to the region of interest detection subnetwork (H1) (i.e. bringing the subnetwork into the next phase), a final classification number (Cls) and a bounding box (B) are assigned to each goal proposal. This is expressed in the present invention as follows:
acquiring a defect preliminary judgment area in the complete image by using a Cascade R-CNN algorithm according to the welding defect characteristics;
and substituting the defect primary judgment area into a Cascade R-CNN algorithm again, and carrying out secondary judgment on the defect primary judgment area according to the welding defect characteristics to obtain a welding defect boundary frame.
In a specific embodiment, the invention is deployed in opencv4.1.0 on raspberry pi 4B, and a UI interface is designed in the Qt5 environment, so as to perform defect detection on a circuit board. A500-ten-thousand-pixel full-color high-speed digital CCD camera is used as a camera, local images of a circuit board are scanned for multiple times, and the local images are spliced according to the sequence of each part on the circuit board to obtain a complete image. The multiple detection results are compared with the manual calibration results, and the actual application accuracy rate can reach more than 95%.
Example two
In order to better explain the technical content of the present invention, the present embodiment explains the present invention by the form of a system structure, as shown in fig. 2, a welding spot welding defect detecting system based on image processing includes:
the preprocessing unit is used for acquiring and preprocessing an image sample set of the circuit board with the welding defect of the welding spot;
the model construction unit is used for selecting and extracting welding defect characteristics in the image sample set through a convolutional neural network and generating a detection model;
the defect judging unit is used for calling the detection model to extract a welding defect boundary frame in the complete image;
and the output unit is used for generating and outputting a labeling picture according to the welding defect boundary frame.
Further, the preprocessing unit, preprocessing, includes: adjusting the size of the image sample to a preset size, and randomly turning and normalizing.
Further, in the model construction unit, the convolutional neural network selects a Cascade R-CNN algorithm to train the image sample set, and the specific mode is as follows:
and acquiring an output result of suspected welding defect characteristics in each layer of R-CNN network of the image sample, taking the output result as the input of the next-level R-CNN network, and finally acquiring a boundary box of element welding characteristics, wherein the R-CNN networks at all levels are cascaded based on respective IoU threshold values.
Further, the defect judgment unit judges the welding defect in two stages by the detection model through a Cascade R-CNN algorithm, and the specific stages are as follows:
acquiring a defect preliminary judgment area in the complete image by using a Cascade R-CNN algorithm according to the welding defect characteristics;
and substituting the defect primary judgment area into a Cascade R-CNN algorithm again, and carrying out secondary judgment on the defect primary judgment area according to the welding defect characteristics to obtain a welding defect boundary frame.
The device further comprises an image acquisition unit, which is used for acquiring partial images with preset sizes of all parts on the circuit board and splicing the partial images into a complete image according to the sequence of all parts on the circuit board.
In summary, the method and the system for detecting welding spot welding defects based on image processing provided by the invention perform convolutional neural network training on a sample set by using the Cascade R-CNN algorithm, so that the extracted features better conform to the features of actual welding spot defects by using the algorithm, and meanwhile, the welding defects of more types of patch elements can be detected.
And in the process of extracting the defect boundary frame, the Cascade R-CNN algorithm is used again to obtain the defect primary judgment region, and secondary judgment is carried out to reduce the region, so that the finally obtained position of the defect boundary frame is more accurate, the range is smaller, and a user is helped to accurately judge the fault occurrence point.
The complete circuit board image is obtained in an image splicing mode, manual coordinate calibration is not needed, and meanwhile, the original data set is expanded through image preprocessing, so that defect characteristics obtained during convolutional neural network training are more sufficient.
It should be noted that all the directional indicators (such as up, down, left, right, front, and rear … …) in the embodiment of the present invention are only used to explain the relative position relationship between the components, the movement situation, etc. in a specific posture (as shown in the drawing), and if the specific posture is changed, the directional indicator is changed accordingly.
Moreover, descriptions of the present invention as relating to "first," "second," "a," etc. are for descriptive purposes only and are not to be construed as indicating or implying relative importance or implicit ly indicating a number of technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include at least one such feature. In the description of the present invention, "a plurality" means at least two, e.g., two, three, etc., unless specifically limited otherwise.
In the present invention, unless otherwise expressly stated or limited, the terms "connected," "secured," and the like are to be construed broadly, and for example, "secured" may be a fixed connection, a removable connection, or an integral part; can be mechanically or electrically connected; they may be directly connected or indirectly connected through intervening media, or they may be connected internally or in any other suitable relationship, unless expressly stated otherwise. The specific meanings of the above terms in the present invention can be understood by those skilled in the art according to specific situations.
In addition, the technical solutions in the embodiments of the present invention may be combined with each other, but it must be based on the realization of those skilled in the art, and when the technical solutions are contradictory or cannot be realized, such a combination of technical solutions should not be considered to exist, and is not within the protection scope of the present invention.
The specific embodiments described herein are merely illustrative of the spirit of the invention. Various modifications or additions may be made to the described embodiments or alternatives may be employed by those skilled in the art without departing from the spirit or ambit of the invention as defined in the appended claims.
Claims (10)
1. A welding spot welding defect detection method based on image processing is characterized by comprising the following steps:
s1: acquiring and preprocessing an image sample set of a circuit board with welding spot welding defects;
s2: selecting and extracting welding defect characteristics in an image sample set through a convolutional neural network, and generating a detection model;
s3: acquiring a complete image of the circuit board, and calling a detection model to acquire a welding defect boundary frame of the complete image;
s4: and generating and outputting a marking picture according to the welding defect boundary frame.
2. The method for detecting welding spot welding defects based on image processing as claimed in claim 1, wherein in said step S1, the preprocessing includes: adjusting the size of the image sample to a preset size, and randomly turning and normalizing.
3. The method for detecting welding spot welding defects based on image processing as claimed in claim 1, wherein in said step S2, the convolutional neural network selects a Cascade R-CNN algorithm to train the image sample set, and the specific steps are as follows:
and acquiring an output result of suspected welding defect characteristics in each layer of R-CNN network of the image sample, taking the output result as the input of the next-level R-CNN network, and finally acquiring a boundary box of element welding characteristics, wherein the R-CNN networks at all levels are cascaded based on respective IoU threshold values.
4. The method for detecting welding spot welding defects based on image processing as claimed in claim 3, wherein in step S3, the detection model is a two-stage judgment of welding defects through Cascade R-CNN algorithm, and the specific stages are as follows:
acquiring a defect preliminary judgment area in the complete image by using a Cascade R-CNN algorithm according to the welding defect characteristics;
and substituting the defect primary judgment area into a Cascade R-CNN algorithm again, and carrying out secondary judgment on the defect primary judgment area according to the welding defect characteristics to obtain a welding defect boundary frame.
5. The method for detecting welding spot welding defects based on image processing as claimed in claim 1, wherein in said step 3, the obtaining of the complete image specifically comprises the steps of:
and acquiring local images with preset sizes of all parts on the circuit board, and splicing the local images into a complete image according to the sequence of all parts on the circuit board.
6. A welding spot welding defect detection system based on image processing is characterized by comprising:
the preprocessing unit is used for acquiring and preprocessing an image sample set of the circuit board with the welding defect of the welding spot;
the model construction unit is used for selecting and extracting welding defect characteristics in the image sample set through a convolutional neural network and generating a detection model;
the defect judging unit is used for calling the detection model to extract a welding defect boundary frame in the complete image;
and the output unit is used for generating and outputting a labeling picture according to the welding defect boundary frame.
7. The welding spot welding defect detection system based on image processing as claimed in claim 6, wherein said preprocessing unit, preprocessing, comprises: adjusting the size of the image sample to a preset size, and randomly turning and normalizing.
8. The system for detecting welding spot welding defects based on image processing as claimed in claim 6, wherein in said model building unit, the convolutional neural network selects Cascade R-CNN algorithm to train the image sample set, the concrete way is:
and acquiring an output result of suspected welding defect characteristics in each layer of R-CNN network of the image sample, taking the output result as the input of the next-level R-CNN network, and finally acquiring a boundary frame of element welding characteristics, wherein the R-CNN networks at all levels are cascaded based on respective IoU threshold values.
9. The system according to claim 8, wherein the defect determining unit determines the weld defect in two stages by using a detection model according to a Cascade R-CNN algorithm, the two stages being:
acquiring a defect preliminary judgment area in the complete image by using a Cascade R-CNN algorithm according to the welding defect characteristics;
and substituting the defect primary judgment area into a Cascade R-CNN algorithm again, and carrying out secondary judgment on the defect primary judgment area according to the welding defect characteristics to obtain a welding defect boundary frame.
10. The system for detecting welding spot welding defects based on image processing as claimed in claim 6, further comprising an image obtaining unit for obtaining partial images with preset sizes of the parts on the circuit board and splicing the partial images into a complete image according to the sequence of the parts on the circuit board.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202110268941.3A CN113077416A (en) | 2021-03-12 | 2021-03-12 | Welding spot welding defect detection method and system based on image processing |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202110268941.3A CN113077416A (en) | 2021-03-12 | 2021-03-12 | Welding spot welding defect detection method and system based on image processing |
Publications (1)
Publication Number | Publication Date |
---|---|
CN113077416A true CN113077416A (en) | 2021-07-06 |
Family
ID=76612230
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202110268941.3A Withdrawn CN113077416A (en) | 2021-03-12 | 2021-03-12 | Welding spot welding defect detection method and system based on image processing |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN113077416A (en) |
Cited By (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN113570571A (en) * | 2021-07-27 | 2021-10-29 | 深圳大学 | Industrial edge end power battery defect detection method and system |
CN114742832A (en) * | 2022-06-13 | 2022-07-12 | 惠州威尔高电子有限公司 | Welding defect detection method for MiniLED thin plate |
CN115423811A (en) * | 2022-11-04 | 2022-12-02 | 长春光华微电子设备工程中心有限公司 | Method and device for registering welding points on chip |
-
2021
- 2021-03-12 CN CN202110268941.3A patent/CN113077416A/en not_active Withdrawn
Cited By (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN113570571A (en) * | 2021-07-27 | 2021-10-29 | 深圳大学 | Industrial edge end power battery defect detection method and system |
CN114742832A (en) * | 2022-06-13 | 2022-07-12 | 惠州威尔高电子有限公司 | Welding defect detection method for MiniLED thin plate |
CN115423811A (en) * | 2022-11-04 | 2022-12-02 | 长春光华微电子设备工程中心有限公司 | Method and device for registering welding points on chip |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN111179251B (en) | Defect detection system and method based on twin neural network and by utilizing template comparison | |
JP7004145B2 (en) | Defect inspection equipment, defect inspection methods, and their programs | |
CN113077416A (en) | Welding spot welding defect detection method and system based on image processing | |
JP6869490B2 (en) | Defect inspection equipment, defect inspection methods, and their programs | |
CN111507958A (en) | Target detection method, training method of detection model and electronic equipment | |
JP5010207B2 (en) | Pattern inspection apparatus and semiconductor inspection system | |
CN111597941B (en) | Target detection method for dam defect image | |
WO2017071406A1 (en) | Method and system for detecting pin of gold needle element | |
CN108871185B (en) | Method, device and equipment for detecting parts and computer readable storage medium | |
TW201930908A (en) | Board defect filtering method and device thereof and computer-readabel recording medium | |
Chavan et al. | Quality control of PCB using image processing | |
CN108428247B (en) | Method and system for detecting direction of soldering tin point | |
JP2010008159A (en) | Visual inspection processing method | |
CN117351472A (en) | Tobacco leaf information detection method and device and electronic equipment | |
CN116934737A (en) | Weld joint combination defect identification and classification method | |
JP2768344B2 (en) | Pattern inspection apparatus and inspection method thereof | |
CN115937555A (en) | Industrial defect detection algorithm based on standardized flow model | |
CN116071335A (en) | Wall surface acceptance method, device, equipment and storage medium | |
CN112750113B (en) | Glass bottle defect detection method and device based on deep learning and linear detection | |
KR20230036650A (en) | Defect detection method and system based on image patch | |
CN117455917B (en) | Establishment of false alarm library of etched lead frame and false alarm on-line judging and screening method | |
CN112183311A (en) | Transformer substation equipment fault detection method based on bimodal data fusion | |
CN112465002B (en) | Intelligent recognition method and device for copper wire weld mark on fire scene | |
CN117095316B (en) | Road surface inspection method, device, equipment and readable storage medium | |
CN117495846B (en) | Image detection method, device, electronic equipment 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 | ||
WW01 | Invention patent application withdrawn after publication |
Application publication date: 20210706 |
|
WW01 | Invention patent application withdrawn after publication |