CN113537301B - Defect detection method based on template self-adaptive matching of bottle body labels - Google Patents
Defect detection method based on template self-adaptive matching of bottle body labels Download PDFInfo
- Publication number
- CN113537301B CN113537301B CN202110698309.2A CN202110698309A CN113537301B CN 113537301 B CN113537301 B CN 113537301B CN 202110698309 A CN202110698309 A CN 202110698309A CN 113537301 B CN113537301 B CN 113537301B
- Authority
- CN
- China
- Prior art keywords
- label
- image
- detection
- template
- bottle body
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Images
Classifications
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02P—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN THE PRODUCTION OR PROCESSING OF GOODS
- Y02P90/00—Enabling technologies with a potential contribution to greenhouse gas [GHG] emissions mitigation
- Y02P90/30—Computing systems specially adapted for manufacturing
Landscapes
- Image Analysis (AREA)
- Image Processing (AREA)
Abstract
The invention discloses a defect detection method based on template self-adaptive matching of bottle body labels, which comprises the following steps: dividing the target label according to the coordinate attribute to form a plurality of label bodies, and manufacturing corresponding detection templates for the plurality of label bodies; preprocessing the acquired bottle body image to form a label image, wherein the preprocessing comprises the segmentation of a bottle body label, the position correction and the recording of coordinates; and calling the detection template by using a detection function to correspondingly detect the label image, and outputting a detection result. The invention can effectively detect a plurality of defect items such as bottle body label loss, label up-down and left-right deviation, label incomplete, label wrinkles, label scratches, label bubbles, label deflection and the like, and can carry out online real-time detection on a production line without influencing normal production rhythm.
Description
Technical Field
The invention relates to the technical field of defect counting, in particular to a defect detection method based on self-adaptive matching of bottle body labels on the basis of a template.
Background
In the bottle manufacturing plant, the production line generally all is the operation of not stopping for a long time, and the labeller also needs long-time operation production promptly, needs staff's whole journey to the quality problem of label to detect, but system bottle, subsides mark have accomplished mechanization, and output is big, and is fast, and this makes label measurement personnel's work load grow, and long-time work consumes a large amount of efforts, leaks to examine in normal production, the wrong detection etc. causes product quality to descend, and manufacturing cost increases.
Disclosure of Invention
The invention aims to provide a defect detection method based on self-adaptive template matching bottle body label, which is used for shooting a bottle body with a label to acquire an image for detection, judging whether the label is a qualified product or not, and feeding back a detection result to a preorder process, so that a labeling machine can be adjusted in time, and loss is reduced.
In order to achieve the purpose, the technical scheme adopted by the invention is as follows:
a defect detection method based on template self-adaptive matching of bottle body labels comprises the following steps:
s1, segmenting a target label according to coordinate attributes to form a plurality of label bodies, and manufacturing corresponding detection templates for the plurality of label bodies;
s2, preprocessing the acquired bottle body image to form a label image, wherein the preprocessing comprises the segmentation of a bottle body label, the position correction and the recording of coordinates;
and S3, calling the detection template by using a detection function to correspondingly detect the label image, and outputting a detection result.
Preferably, before preprocessing the collected bottle body image to form a label image, the method further comprises the step of judging whether the label exists or not, recording the label-free bottle body as unqualified, and preprocessing the label-contained bottle body image to form the label image.
Preferably, in step S3, when the detection template is used to correspondingly detect the label image, different regions of the label are searched and matched by the detection templates of different parts of the label, the edge gradient of the search region is calculated by using corresponding detection operators from the transverse direction and the longitudinal direction respectively through edge gradient calculation, the edge of the region to be matched is obtained, and then the matched image is matched with the edge of the detection template, and after the matched image is obtained, the label defect detection is performed by using a corresponding detection algorithm in the matched image.
Preferably, when the matching is performed with the edge of the detection template, the image to be matched is firstly sampled up and down by using pyramid level, the proportion relationship between the edge of the image to be matched and the edge of the detection template is calculated again, the proportion in the two directions of length and width is adaptively scaled, certain scaling adjustment is performed, and finally the scaling proportion in the two directions of length and width of the image to be matched is set, so that the matched image is obtained.
Preferably, in step S1, the detection template is formed by:
firstly, pyramid hierarchical calculation is carried out on a label image formed after segmentation so as to deal with the scale change of the label, then the edge gradient and the direction information of the label are calculated, and the detection template manufacturing is completed.
Preferably, the dividing is performed according to a set ROI region for the label of the bottle.
Preferably, before the position correction is performed on the divided label, the method further comprises the step of judging the vertical and horizontal offset and/or skew of the divided label.
Preferably, the step of judging the up-down, left-right offset and/or skew of the label is as follows:
mapping the coordinates of the divided labels, searching a plurality of characteristic points of the bottle body, recording the coordinates and angles of the labels formed after division in an original image relative to the characteristic points of the bottle body, calculating the relative positions of the bottle body and the coordinates, and judging the vertical and horizontal offsets of the labels; the formula is as follows:
wherein N is the number of label partitions, N (x,y) Is a coordinate value, L, of a characteristic point of the bottle body (xi,yi) Is the coordinate of the ith label, diffi i Comparing a range value of the ith qualified label coordinate with the standard difference value of the bottle body coordinate according to the set label fluctuation with Diff, and judging the up-and-down deviation of the label;
judging the label skew, and the formula is as follows:
wherein N is θ Is the angle of the bottle body,
and the angle of the ith label is theta, the relative angle difference between the angle of the bottle body characteristic point and the angle of the label is theta, and the inclination of the label is judged by comparing the range value which can be fluctuated according to the set label with the theta.
Preferably, in step S3, the segmented label image is matched with the detection template, and defects of the label, including missing, scratch, bubble, and wrinkle, are detected according to different algorithms.
Preferably, each defect position is found by using a difference method, and the difference is expressed as:
D(x,y)＝|T(m,n)-G(m′,n′)|
wherein, T (m, n) is a pre-established template image, G (m ', n') is a test label image of the wine bottle, D (x, y) is a differential image, m, n are transverse coordinates and longitudinal coordinates of pixels of the template image, m ', n' are transverse coordinates and longitudinal coordinates of pixels of the test label image, and x, y are transverse coordinates and longitudinal coordinates of pixels of the differential image.
The invention detects the labels of different types and different styles by a segmentation method, can quickly detect the problem labels, does not need manual intervention, can be suitable for various application scenes, reduces the development time and can realize quick transplantation.
The method has the advantages that the better matching result can be obtained at the radian position of the edge of the bottle body, the template matching precision is higher, the effect is better, the subsequent difference precision is improved, the misjudgment rate of the result is reduced, and the accuracy, the robustness and the usability of obstacle recognition are effectively improved.
The defect detection method based on the template self-adaptive matching bottle body label, provided by the invention, has the advantages that the detection template can be adjusted and adapted according to labels of different types and different styles, and can be adjusted and used aiming at different production lines, so that the problem label can be quickly detected by using different templates for matching according to different production lines in the following step.
The invention can quickly detect the quality of the bottle label production line, effectively improves the production efficiency and reduces the false alarm rate of the detection system, and can be applied to various bottle labeling production lines.
Drawings
FIG. 1 is a flow chart of a method for defect detection based on template adaptive matching of bottle labels according to the present invention;
FIGS. 2 a-2 b are graphs comparing the matching results of the template matching method of the present invention with the matching results of the conventional matching method;
3 a-3 c are schematic diagrams of a plurality of labels formed by segmenting a label according to the present invention;
fig. 4 a-4 e are graphs showing the detection results of the present invention for the defects of the label.
Detailed Description
In order to make the objects, technical solutions and advantages of the present invention more apparent, the present invention is further described in detail below with reference to the following detailed examples and the accompanying drawings.
The invention provides a defect detection method based on template self-adaptive matching bottle body labels, a flow chart of which is shown in figure 1, and the method comprises the following steps:
step S1: manufacturing detection templates of different label parts;
for different types and styles of bottle bodies, labels attached to the bottle bodies are segmented by utilizing a coordinate range, as shown in fig. 3 a-3 ac, a plurality of label part images formed after label segmentation are obtained, pyramid hierarchical calculation is carried out on the segmented collected images so as to deal with the scale change of the labels, edge gradient and direction information of the labels are calculated, and a matching template is manufactured.
As shown in fig. 2a, the graph is a matching result of the conventional template matching, where there is a matching deviation in the edge portion of the small font, and fig. 2b is a matching result of the template matching method used in the present invention, where the matching method in the edge portion where the deviation is likely to occur in the conventional matching can perform matching well, and the matching accuracy and the detection accuracy can be improved to a greater extent.
Step S2: and preprocessing the acquired image of the bottle or the bottle body, including operations of preliminary detection, label segmentation, label position correction, label coordinate recording and the like.
The method comprises the following specific steps: the method comprises the steps of firstly carrying out preliminary detection on collected images, judging whether a bottle body of a bottle body is printed with a label or not, and directly rejecting unqualified products and shooting a next image if the bottle body of the bottle body is not printed with the label.
If the bottle body has a label, the position of the bottle body is corrected, the label of the bottle body is divided according to a set ROI (region of interest), the divided label is subjected to coordinate mapping, the coordinates and the angles of the labels formed after division in an original drawing are recorded, the characteristic points of the bottle body are searched, the relative position between the bottle body and each coordinate is calculated, and the vertical deviation of the label is judged, wherein the formula is as follows:
wherein N is the number of label partitions, N (x,y) Is a coordinate value, L, of a characteristic point of the bottle body (xi,yi) Is the coordinate of the ith label, diffi j And comparing the standard difference value of the ith qualified label coordinate and the bottle body coordinate with a Diff (Diff) according to a fluctuable range value of a set label, and judging the up-and-down deviation of the label.
Judging whether the skew exists or not, wherein the formula is as follows:
wherein N is θ Is the angle of the bottle body,
the angle of the ith label is theta, the relative angle difference between the angle of the bottle body characteristic point of the bottle body and the angle of the label is theta, and the inclination of the label is judged by comparing the fluctuated range value of the set label with the theta.
Step S3: respectively inputting the divided images into the detection function of each label area, and judging whether the labels of the part are qualified or not;
the method comprises the following specific steps: respectively inputting the image preprocessed in the step S2 into a corresponding detection function or algorithm, using the template adaptive matching of the present invention in the corresponding detection function, performing search matching on different regions through different partial templates, and calculating the edge gradient thereof, wherein operators in two directions, i.e. in the transverse direction and the longitudinal direction: operator a and operator B, as follows:
calculating the edge gradient of the search area through the operator of the formula to obtain the edge of the area to be matched, matching the edge of the area to be matched with the edge of the template, utilizing pyramid level up-down sampling, calculating the proportion relation between the matched image edge and the edge of the template to adaptively scale the proportion in the length direction and the width direction, carrying out certain proportion scaling adjustment, setting the scaling proportion of the matched image in the length direction and the width direction, and finally obtaining the matched image. Through detecting defects such as label breakage, scratches, bubbles, wrinkles and the like in the matched images, as shown in fig. 4 a-4 e, fig. 4a is a certain detected image and can see that an obvious defect exists, fig. 4b is a step of detecting the image and outputting and marking the defective image, and fig. 4c is a detected image and can see that a certain scratch defect exists, and the scratch defect can be well marked and removed through a detection function. Fig. 4e illustrates the wrinkle defect detection by the wrinkle detection algorithm, and the wrinkle defect is marked by the wrinkle detection algorithm and the result is recorded.
Step S4: outputting the detection results of the steps S2 and S3;
the method specifically comprises the following steps: and (4) comprehensively outputting the results of the bottle body labels detected by the S2 and the S3, judging whether the bottle body labels are qualified products or not, inputting the qualified products to the next station, rejecting unqualified products at the station, and recording and feeding back the results to the staff.
The practical application of utilizing machine vision to detect the printing quality of the bottle body label has a great challenge, especially, the bottle bodies in the market have various styles and especially have some irregular-shaped bottles, which brings certain difficulty to the image acquisition and pretreatment of the step S2 and directly influences the accuracy of the detection result.
The existing traditional matching method is based on gray scale matching and characteristic matching, the gray scale matching has higher requirements on the field acquisition environment and the precision is difficult to meet the industrial requirements, the characteristic matching effect is best, but the algorithm is complex, the time consumption is long, and the real-time performance is difficult to realize.
As shown in fig. 2a, it can be seen that a large matching error occurs at an edge portion of a font by using a conventional matching method, which causes an error in a detection result and reduces accuracy, and fig. 2b shows that a matching method used in the present invention can also have a good matching effect at the edge portion, improve subsequent detection accuracy, and reduce loss in actual industrial production.
The invention can be used in an automatic production line for detecting the labels of the bottles and in a production line for labeling the bottles. At present, the label of wine is mainly detected manually, the detection result is directly or indirectly influenced by factors such as the specificity of the production line, the working time is long, detection is required day and night, the environment in the production line and the like, so that the yield of production is reduced, the product damaged by the label flows into the market, and the influence on the product is large. The lamp inspection worker works under the irradiation of high-intensity light for a long time, fatigue is easy to generate, missing inspection is easier to generate, and meanwhile, the current factory also faces the difficulty of people increasingly lacking in workers.
The invention can detect various defects based on the template self-adaptive matching detection algorithm, can realize the quick detection in the production line, can deal with various environments in the production line, is far superior to the operation of workers for monotonous repeated action machines without error operation, can greatly improve the product yield of the production line and reduce the production cost.
The method can be used for carrying out quality detection on the labeling production line of the white spirit bottles with high speed and large quantity, effectively improves the production efficiency and reduces the false alarm rate of a detection system, can be applied to various bottle labeling production lines, and is mainly used for solving the problem of product quality control in automatic production.
The invention can be applied to the detection of the label defect and the defect detection of the bottle body. Because the production of the bottle body is influenced by the production process, a certain amount of defective bottles are inevitably existed in the firing process of the boiler, for example, the bottle mouth has impurities and defects, and the bottle body has the defects of burst, damage, crack and the like, and the defective bottles can be removed in advance through the invention.
The automatic control system can efficiently realize automation and standardization of the production line, solve the technical problem of bottle body high-reflection transparent material visual image detection, timely feed back problems in the production line to reduce loss, perform accurate, intelligent and stable automatic control on bottle body production quality, and replace manual quality inspectors in the existing production line.
The above-mentioned embodiments are intended to illustrate the objects, technical solutions and advantages of the present invention in further detail, and it should be understood that the above-mentioned embodiments are only exemplary embodiments of the present invention, and are not intended to limit the present invention, and any modifications, equivalents, improvements and the like made within the spirit and principle of the present invention should be included in the protection scope of the present invention.
Claims (6)
1. The defect detection method based on template self-adaptive matching of bottle body labels is characterized by comprising the following steps:
s1, dividing a target label according to coordinate attributes to form a plurality of label bodies, and manufacturing corresponding detection templates for the plurality of label bodies;
s2, preprocessing the collected bottle body image to form a label image, wherein the preprocessing comprises the segmentation of a bottle body label, the position correction and the coordinate recording;
s3, calling the detection template by using a detection function to correspondingly detect the label image, and outputting a detection result;
in step S3, matching the segmented label image with a detection template, and detecting the defects of the label, including deletion, scratch, bubble and wrinkle, according to different algorithms;
finding out the positions of the defects by using a difference method, wherein the difference is expressed as:
D(x,y)＝|T(m,n)-G(m′,n′)|
wherein, T (m, n) is a pre-established template image, G (m ', n') is a test label image of the wine bottle, D (x, y) is a difference image, m, n are transverse coordinates and longitudinal coordinates of pixels of the template image, m ', n' are transverse coordinates and longitudinal coordinates of pixels of the test label image, and x, y are transverse coordinates and longitudinal coordinates of pixels of the difference image;
before the position correction is carried out on the segmented label, the method also comprises the step of judging the vertical and horizontal offset and/or the skew of the segmented label;
the steps of judging the up-down, left-right offset and/or skew of the label are as follows:
mapping the coordinates of the divided labels, searching a plurality of characteristic points of the bottle body, recording the coordinates and angles of the labels formed after division in an original image relative to the characteristic points of the bottle body, calculating the relative positions of the bottle body and the coordinates, and judging the vertical and horizontal offsets of the labels; the formula is as follows:
wherein N is the number of label partitions, N (x,y) Is a coordinate value of the bottle body characteristic point,
is the coordinate of the ith label, diffi i Comparing a range value of the ith qualified label coordinate with the standard difference value of the bottle body coordinate according to the set label fluctuation with Diff, and judging the up-and-down deviation of the label;
judging the label skew, and the formula is as follows:
wherein N is θ Is the angle of the bottle body,
and the angle of the ith label is theta, the relative angle difference between the angle of the bottle body characteristic point and the angle of the label is theta, and the inclination of the label is judged by comparing the range value which can be fluctuated according to the set label with the theta.
2. The method of claim 1, further comprising the step of determining the presence or absence of a label before preprocessing the collected bottle image to form a label image, recording the non-labeled bottle as non-conforming, and preprocessing the labeled bottle image to form a label image.
3. The method for detecting the defect of the self-adaptive matching bottle label based on the template as claimed in claim 1, wherein in step S3, when the detection template is used to correspondingly detect the label image, the detection template of different parts of the label is used to search and match different areas of the label, the edge gradient of the search area is calculated by using corresponding detection operators from the transverse direction and the longitudinal direction respectively through the edge gradient calculation, the edge of the area to be matched is obtained, then the edge of the detection template is matched, and after the matched image is obtained, the label defect detection is performed by using a corresponding detection algorithm in the matched image.
4. The method for detecting the defects of the self-adaptive matching bottle labels based on the template as claimed in claim 3, wherein when the edges of the detection template are matched, the image to be matched is firstly sampled up and down by pyramid level, the proportion relation between the edges of the image to be matched and the edges of the detection template is calculated again, the proportion between the edges of the image to be matched and the edges of the detection template is adaptively scaled in the length and width directions, a certain proportion scaling adjustment is performed, and finally the scaling proportion between the edges of the image to be matched and the detection template in the length and width directions is set, so that the matched image is obtained.
5. The method for detecting defects of bottle labels based on template adaptive matching as claimed in claim 1, wherein in step S1, the detection template is formed by the following steps:
firstly, pyramid hierarchical calculation is carried out on a label image formed after segmentation so as to deal with the scale change of the label, then the edge gradient and the direction information of the label are calculated, and the detection template manufacturing is completed.
6. The method for detecting defects of bottle labels based on template adaptive matching according to claim 1, wherein the segmentation is to segment the labels of the bottles according to a set ROI.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110698309.2A CN113537301B (en) | 2021-06-23 | 2021-06-23 | Defect detection method based on template self-adaptive matching of bottle body labels |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110698309.2A CN113537301B (en) | 2021-06-23 | 2021-06-23 | Defect detection method based on template self-adaptive matching of bottle body labels |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN113537301A CN113537301A (en) | 2021-10-22 |
| CN113537301B true CN113537301B (en) | 2022-08-16 |
Family
ID=78096484
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110698309.2A Active CN113537301B (en) | 2021-06-23 | 2021-06-23 | Defect detection method based on template self-adaptive matching of bottle body labels |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113537301B (en) |
Families Citing this family (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114972157A (en) * | 2022-01-10 | 2022-08-30 | 华为技术有限公司 | Edge defect detection method, device and storage medium |
| CN114581407B (en) * | 2022-03-06 | 2023-04-14 | 扬州晶华新能源科技有限公司 | Self-adaptive defect detection method for photovoltaic module |
| CN115791817B (en) * | 2023-02-06 | 2023-04-28 | 泸州老窖股份有限公司 | Transparent wine bottle quality detection method |
| CN116563289B (en) * | 2023-07-11 | 2023-09-29 | 凯德技术长沙股份有限公司 | Labeling quality detection method and system based on machine vision |
| CN116777912B (en) * | 2023-08-22 | 2023-11-03 | 山东明佳科技有限公司 | Method, system, equipment and storage medium for identifying label folds of milk glass bottle body |
| CN117474924B (en) * | 2023-12-28 | 2024-03-15 | 山东鲁抗医药集团赛特有限责任公司 | Label defect detection method based on machine vision |
| CN117975168B (en) * | 2024-03-28 | 2024-05-31 | 千里眼(广州)人工智能科技有限公司 | Method and system for detecting quality of printed label |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103679702A (en) * | 2013-11-20 | 2014-03-26 | 华中科技大学 | Matching method based on image edge vectors |
| CN106952257A (en) * | 2017-03-21 | 2017-07-14 | 南京大学 | A kind of curved surface label open defect detection method based on template matches and Similarity Measure |
| CN109724990A (en) * | 2019-01-08 | 2019-05-07 | 上海大学 | The quick positioning and detection method in coding region in a kind of packing box label |
| CN112837257A (en) * | 2019-11-06 | 2021-05-25 | 广州达普绅智能设备有限公司 | Curved surface label splicing detection method based on machine vision |
| CN112902843A (en) * | 2021-02-04 | 2021-06-04 | 北京创源微致软件有限公司 | Label attaching effect detection method |
Family Cites Families (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107561087A (en) * | 2017-08-31 | 2018-01-09 | 广东工业大学 | A kind of mouse logo positioning and defect inspection method based on machine vision |
| CN110070090B (en) * | 2019-04-25 | 2021-06-25 | 上海大学 | Logistics label information detection method and system based on handwritten character recognition |
-
2021
- 2021-06-23 CN CN202110698309.2A patent/CN113537301B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103679702A (en) * | 2013-11-20 | 2014-03-26 | 华中科技大学 | Matching method based on image edge vectors |
| CN106952257A (en) * | 2017-03-21 | 2017-07-14 | 南京大学 | A kind of curved surface label open defect detection method based on template matches and Similarity Measure |
| CN109724990A (en) * | 2019-01-08 | 2019-05-07 | 上海大学 | The quick positioning and detection method in coding region in a kind of packing box label |
| CN112837257A (en) * | 2019-11-06 | 2021-05-25 | 广州达普绅智能设备有限公司 | Curved surface label splicing detection method based on machine vision |
| CN112902843A (en) * | 2021-02-04 | 2021-06-04 | 北京创源微致软件有限公司 | Label attaching effect detection method |
Also Published As
| Publication number | Publication date |
|---|---|
| CN113537301A (en) | 2021-10-22 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN113537301B (en) | Defect detection method based on template self-adaptive matching of bottle body labels | |
| CN109900706B (en) | Weld joint based on deep learning and weld joint defect detection method | |
| CN107341802B (en) | Corner sub-pixel positioning method based on curvature and gray scale compounding | |
| CN106370659B (en) | A kind of bottle body package quality visible detection method | |
| CN102162797B (en) | Algorithm for detecting glass bottle neck damage and bottle bottom dirt | |
| CN109308700A (en) | A kind of visual identity defect inspection method based on printed matter character | |
| CN107561087A (en) | A kind of mouse logo positioning and defect inspection method based on machine vision | |
| CN115439481A (en) | Deaerator welding quality detection method based on image processing | |
| CN112200776A (en) | Chip packaging defect detection method and detection device | |
| CN114565848B (en) | Liquid medicine level detection method and system in complex scene | |
| CN114549504A (en) | Appearance quality detection method based on machine vision | |
| CN114022441A (en) | Defect detection method for irregular hardware | |
| Xie et al. | A rapid inspection method for encapsulating quality of PET bottles based on machine vision | |
| Gong et al. | Adaptive visual inspection method for transparent label defect detection of curved glass bottle | |
| CN113702393A (en) | Intrinsic safety type mining conveyor belt surface damage detection system and detection method | |
| CN108416790A (en) | A kind of detection method for workpiece breakage rate | |
| CN111539951B (en) | Visual detection method for outline size of ceramic grinding wheel head | |
| CN113160220A (en) | Door handle homing and bending detection method based on deep learning | |
| CN115753791B (en) | Defect detection method, device and system based on machine vision | |
| CN113192061B (en) | Extraction method and device of LED package appearance detection image, electronic equipment and storage medium | |
| CN111192261A (en) | Method for identifying lens defect types | |
| CN115713485A (en) | Defect detection method for insufficient glass bottle opening | |
| CN111612769A (en) | Method for detecting surface concave-convex defect of triode packaging metal cap | |
| Yuanyuan | Research on automatic visual inspection method for character on cartridge fuse based on template matching | |
| CN113588655A (en) | Detection device for surface defects of MDF fiber lines and working method thereof |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant | ||
| CP01 | Change in the name or title of a patent holder |
Address after: 300457 unit msd-g1-1001, TEDA, No.57, 2nd Street, Tianjin Economic and Technological Development Zone, Binhai New Area, Tianjin Patentee after: Tianjin Zhongke intelligent identification Co.,Ltd. Address before: 300457 unit msd-g1-1001, TEDA, No.57, 2nd Street, Tianjin Economic and Technological Development Zone, Binhai New Area, Tianjin Patentee before: TIANJIN ZHONGKE INTELLIGENT IDENTIFICATION INDUSTRY TECHNOLOGY RESEARCH INSTITUTE Co.,Ltd. |
|
| CP01 | Change in the name or title of a patent holder |