CN114674830A - Bottle cap flaw detection module on high-speed production line - Google Patents
Bottle cap flaw detection module on high-speed production line Download PDFInfo
- Publication number
- CN114674830A CN114674830A CN202210398891.5A CN202210398891A CN114674830A CN 114674830 A CN114674830 A CN 114674830A CN 202210398891 A CN202210398891 A CN 202210398891A CN 114674830 A CN114674830 A CN 114674830A
- Authority
- CN
- China
- Prior art keywords
- product
- detection
- bottle cap
- station
- unit
- 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
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N21/00—Investigating or analysing materials by the use of optical means, i.e. using sub-millimetre waves, infrared, visible or ultraviolet light
- G01N21/84—Systems specially adapted for particular applications
- G01N21/88—Investigating the presence of flaws or contamination
- G01N21/8851—Scan or image signal processing specially adapted therefor, e.g. for scan signal adjustment, for detecting different kinds of defects, for compensating for structures, markings, edges
-
- 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
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N21/00—Investigating or analysing materials by the use of optical means, i.e. using sub-millimetre waves, infrared, visible or ultraviolet light
- G01N21/84—Systems specially adapted for particular applications
- G01N21/88—Investigating the presence of flaws or contamination
- G01N21/8851—Scan or image signal processing specially adapted therefor, e.g. for scan signal adjustment, for detecting different kinds of defects, for compensating for structures, markings, edges
- G01N2021/8883—Scan or image signal processing specially adapted therefor, e.g. for scan signal adjustment, for detecting different kinds of defects, for compensating for structures, markings, edges involving the calculation of gauges, generating models
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N21/00—Investigating or analysing materials by the use of optical means, i.e. using sub-millimetre waves, infrared, visible or ultraviolet light
- G01N21/84—Systems specially adapted for particular applications
- G01N21/88—Investigating the presence of flaws or contamination
- G01N21/89—Investigating the presence of flaws or contamination in moving material, e.g. running paper or textiles
- G01N21/892—Investigating the presence of flaws or contamination in moving material, e.g. running paper or textiles characterised by the flaw, defect or object feature examined
- G01N2021/8924—Dents; Relief flaws
Abstract
The invention is suitable for the technical field of bottle cap production, and provides a bottle cap flaw detection module on a high-speed production line, which comprises: the acquisition unit is used for acquiring product images of a plurality of stations; the acquisition unit is used for intercepting a picture of the detection area through the acquired product image; the detection unit is used for inputting the intercepted detection area picture into a pre-trained deep learning model for detection; the judging unit is used for judging that the corresponding product is a defective product when detecting that the defect exists in the detection area picture; the control unit is used for controlling the removing device to remove the defective products, and the invention has the beneficial effects that: the detection efficiency is improved, and the missing detection is reduced.
Description
Technical Field
The invention belongs to the technical field of bottle cap production, and particularly relates to a bottle cap flaw detection module on a high-speed production line.
Background
The bottle cap is mainly used for packaging food such as drinks, edible oil and the like, and along with the increasing importance of consumers on food health, the requirements on the good and flawless food packaging are increased day by day. The reverse-pushing food manufacturers have higher and higher quality requirements on food packaging bottle caps, so that the bottle cap manufacturers add a visual detection module on a high-speed production line to replace manual visual detection with low efficiency and easy omission.
In the prior art, manual visual inspection is mostly adopted or a traditional AOI (automatic optical inspection) mode is adopted, the manual visual inspection is assisted by a high-brightness light source, the defects are visually detected by manually adjusting the angle, the defects are observed, namely, the defects are manually removed, and the high-efficiency full-automatic production cannot be realized; traditional AOI detects and need set up the visual detection region alone to different products, and the step is loaded down with trivial details, and simultaneously for avoiding background interference, the detection region can artificially reduce, causes to miss to examine, is difficult to more to match the production efficiency of producing the line at a high speed.
Disclosure of Invention
The embodiment of the invention aims to provide a bottle cap flaw detection module on a high-speed production line, and aims to solve the problems in the background technology.
The embodiment of the invention is realized in such a way that the bottle cap flaw detection module on the high-speed production line comprises:
the acquisition unit is used for acquiring product images of a plurality of stations;
the acquisition unit is used for intercepting a picture of the detection area through the acquired product image;
the detection unit is used for inputting the intercepted detection area picture into a pre-trained deep learning model for detection;
the judging unit is used for judging that the corresponding product is a defective product when detecting that the defect exists in the detection area picture;
and the control unit is used for controlling the rejecting device to reject the defective products.
As a further aspect of the present invention, the image capturing device for the products at the plurality of stations at least includes 6 cameras, and when the number of the cameras is 6, the cameras are respectively arranged at the top of the first station, 4 sides of the second station, and the bottom of the third station, wherein the camera at the first station is used for matching with the low-angle ring light to detect black spots and deformation defects at the top of the bottle cap, the camera at the second station is used for matching with the prism, the coaxial light to detect black spots at the sides, poor combination, and poor printing defects, and the camera at the third station is used for matching with the endoscope lens, the high-angle ring light to detect black spots, foreign matters, deformation, and hole defects inside the bottle cap.
As still further aspect of the present invention, the acquisition unit includes:
the inductor unit is used for inducing whether the product is in place or not through the photoelectric switch;
and the triggering subunit is used for triggering the camera to pick pictures after sensing that the product is in place.
As a further aspect of the present invention, the acquiring unit specifically includes:
the acquisition subunit is used for calculating the position of the product through a machine vision positioning algorithm after the image is acquired;
and the intercepting subunit is used for intercepting the detection area picture according to the relative coordinates of the detection area and the product position.
As a further aspect of the present invention, the step of training the deep learning model includes:
when the model is trained in the previous period, manually marking pictures in the detection area, and dividing the pictures into good products and a plurality of various flaws;
and inputting the manual label into the basic model for deep learning to obtain a deep learning model.
As a further aspect of the present invention, the module further includes:
the comparison subunit is used for comparing the manual re-inspection result uploaded after the defective products are removed with the corresponding defective product removal information;
and the display subunit is used for displaying the comparison result on the human-computer interaction interface.
As a further aspect of the present invention, the acquiring unit further includes:
and the brightness compensation subunit is used for detecting the ambient brightness of the station when acquiring the product image, and when the ambient brightness is lower than a preset brightness threshold value, different light sources and lenses are used for supplementing to ensure that each detection surface can clearly image.
According to the bottle cap flaw detection module on the high-speed production line, provided by the embodiment of the invention, the flaws are collected by the cameras in an all-dimensional dead-angle-free shooting mode by adopting 3 stations and at least 6 cameras, and the clear imaging of each detection surface can be ensured by using different light sources, lenses and lenses.
Drawings
Fig. 1 is a schematic diagram of a main structure of a bottle cap defect detecting module on a high-speed production line.
Fig. 2 is a flow chart of the operation of the bottle cap defect detecting module in the high-speed production line.
Fig. 3 is a schematic structural diagram of a collecting unit in a bottle cap defect detecting module on a high-speed production line.
Fig. 4 is a schematic structural diagram of an acquisition unit of a bottle cap defect detection module on a high-speed production line.
Fig. 5 is a schematic structural diagram of a review display unit in a bottle cap defect detection module on a high-speed production line.
Detailed Description
In order to make the objects, technical solutions and advantages of the present invention more apparent, the present invention is described in further detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention.
Specific implementations of the present invention are described in detail below with reference to specific embodiments.
The invention provides a bottle cap flaw detection module on a high-speed production line, which solves the technical problem in the background technology.
As shown in fig. 1-2, a main structural schematic diagram of a bottle cap defect detecting module in a high-speed production line according to an embodiment of the present invention is shown, where the bottle cap defect detecting module in the high-speed production line includes:
the acquisition unit 10 is used for acquiring product images of a plurality of stations;
the acquisition unit 20 is used for intercepting a detection area picture through the acquired product image;
the detection unit 30 is used for inputting the intercepted detection area picture into a pre-trained deep learning model for detection;
the judging unit 40 is configured to judge that the corresponding product is a defective product when it is detected that a defect exists in the detection area picture;
and the control unit 50 is used for controlling the rejecting device to reject the defective products.
The highest production speed of the high-speed bottle cap production line can reach 800 per minute, and oil stains, black spots, poor combination, poor printing and other defects exist in the production process. The existing bottle cap pressing equipment adopts the mode of collecting firstly and then off-line manual visual inspection, the efficiency is extremely low, in addition, the storage process is easy to cause extrusion and deformation, the product quality and the production yield are influenced, different patterns exist at the top of the bottle cap and around the bottle cap, the background is relatively complex, more defects are easily ignored by human eyes, the omission is caused, meanwhile, the inner wall of the bottle cap is deep, the bottle cap can be seen completely by rotating a plurality of angles, the product position is calculated by a traditional machine vision positioning algorithm after the images are obtained, and the detection area is intercepted according to relative coordinates. And during model training in the early stage, manually marking pictures in the detection area, dividing the pictures into a plurality of types of good products and various flaws, and performing deep learning model training to obtain a model. In the actual production process, the trained model is read, deep learning model prediction is carried out on each product, whether the product is a defective product or not is judged, and if the product is the defective product, a rejecting signal is sent out. The rejecting device rejects the defective products and displays the prediction result on the UI.
As a preferred embodiment of the present invention, the image capturing device for the products at several stations comprises at least 6 cameras, when the number of the cameras is 6, the cameras are respectively arranged at the top of a first station, 4 sides of a second station and the bottom of a third station, wherein the camera at the first station is used for matching with a low-angle ring to detect black spots and deformation defects at the top of the bottle cap, the camera at the second station is used for matching with a prism, coaxial to detect black spots on the sides, poor combination and poor printing defects, and the camera at the third station is used for matching with an endoscope lens, a high-angle ring to detect black spots, foreign matters, deformation and hole defects inside the bottle cap.
The detection system that this module corresponds is divided into 3 stations, disposes 6 industrial cameras altogether. The method comprises the following steps that 1 top camera is arranged at a first station, and defects such as black spots and deformation at the top of a bottle cap are detected in a matched mode through a low-angle ring; the second station is provided with 4 side cameras, and defects such as side black spots, poor combination, poor printing and the like are detected by matching with the prism and the coaxial light; the third station is provided with 1 bottom camera, detects defects such as black spots, foreign matters, deformation, holes and the like in the bottle cap by matching with an endoscope lens and a high-angle ring light, and the endoscope lens realizes that a single camera covers 360-degree visual field of the inner wall of the bottle cap; the photoelectric switch is adopted to sense that the product is in place, the camera is triggered to acquire images, the position of the images is guaranteed to be stable, the visual field waste of the camera is reduced, and the detection precision and speed are improved.
As shown in fig. 3, as another preferred embodiment of the present invention, the collecting unit 10 includes:
the sensor sub-unit 101 is used for sensing whether the product is in place or not through the photoelectric switch;
and the triggering subunit 102 is used for triggering the camera to pick up the picture after sensing that the product is in place.
As shown in fig. 4, as another preferred embodiment of the present invention, the obtaining unit 20 specifically includes:
the acquisition subunit 201 is configured to calculate a product position through a machine vision positioning algorithm after acquiring the image;
and the intercepting subunit 202 is configured to intercept the detection area picture according to the relative coordinates of the detection area and the product position.
As another preferred embodiment of the present invention, the step of training the deep learning model includes:
when the model is trained in the previous period, manually marking pictures in the detection area, and dividing the pictures into good products and a plurality of various flaws;
the method comprises the steps of inputting an artificial label into a basic model for deep learning to obtain a deep learning model, wherein the basic model can be a model obtained based on a Convolutional Neural Network (CNN), the Convolutional Neural Network (CNN) is a feed-forward Neural network (fed Neural network) which contains convolution calculation and has a deep structure, the Convolutional Neural network is one of representative algorithms of deep learning (deep learning), and the Convolutional Neural network has a representation learning (representation learning) capability and can carry out translation invariant classification on input information according to a hierarchical structure of the Convolutional Neural network.
As shown in fig. 5, as another preferred embodiment of the present invention, the module further includes a review display unit 60:
a comparison subunit 601, configured to, after defective products are removed, upload results of manual review and compare the results with corresponding product defective product removal information;
and the display subunit 602 is configured to display the comparison result on a human-computer interaction interface.
As another preferred embodiment of the present invention, the obtaining unit 20 further includes:
and the brightness compensation subunit 203 is used for detecting the ambient brightness of the station when acquiring the product image, and when the ambient brightness is lower than a preset brightness threshold, different light sources and lenses are used for supplementing to ensure that each detection surface can clearly image.
According to the bottle cap defect detection module on the high-speed production line, provided by the embodiment of the invention, the defects are ensured to be collected by the cameras by adopting a mode of taking pictures in all directions without dead angles by adopting 3 stations and at least 6 cameras, and the clear imaging of each detection surface is ensured by using different light sources, lenses and lenses. The method combines the traditional machine vision algorithm and the deep learning algorithm, greatly improves the algorithm detection capability, has strong robustness, does not need to reduce the detection range, and reduces the missing detection.
In order to load the above method and system to operate smoothly, the system may include more or less components than those described above, or combine some components, or different components, besides the various modules described above, for example, input/output devices, network access devices, buses, processors, memories, and the like.
The processor may be a Central Processing Unit (CPU), other general purpose processor, a Digital Signal Processor (DSP), an Application Specific Integrated Circuit (ASIC), an off-the-shelf programmable gate array (FPGA) or other programmable logic device, discrete gate or transistor logic, discrete hardware components, etc. A general purpose processor may be a microprocessor or the processor may be any conventional processor or the like that is the control center for the system and that connects the various components using various interfaces and lines.
The memory may be used to store computer and system programs and/or modules, and the processor may perform the various functions described above by operating or executing the computer programs and/or modules stored in the memory and invoking data stored in the memory. The memory may mainly include a program storage area and a data storage area, where the program storage area may store an operating system, an application program required by at least one function (such as an information collection template presentation function, a product information distribution function, and the like), and the like. The storage data area may store data created according to the use of the berth-state display system (e.g., product information acquisition templates corresponding to different product types, product information that needs to be issued by different product providers, etc.), and the like. In addition, the memory may include high speed random access memory, and may also include non-volatile memory, such as a hard disk, a memory, a plug-in hard disk, a Smart Media Card (SMC), a Secure Digital (SD) card, a flash memory card (FlashCard), at least one disk storage device, a flash memory device, or other volatile solid state storage device.
It should be understood that, although the steps in the flowcharts of the embodiments of the present invention are shown in sequence as indicated by the arrows, the steps are not necessarily executed in sequence as indicated by the arrows. The steps are not performed in the exact order shown and described, and may be performed in other orders, unless explicitly stated otherwise. Moreover, at least a portion of the steps in various embodiments may include multiple sub-steps or multiple stages that are not necessarily performed at the same time, but may be performed at different times, and the order of performance of the sub-steps or stages is not necessarily sequential, but may be performed in turn or alternately with other steps or at least a portion of the sub-steps or stages of other steps.
The technical features of the embodiments described above may be arbitrarily combined, and for the sake of brevity, all possible combinations of the technical features in the embodiments described above are not described, but should be considered as being within the scope of the present specification as long as there is no contradiction between the combinations of the technical features.
The above-mentioned embodiments only express several embodiments of the present invention, and the description thereof is more specific and detailed, but not construed as limiting the scope of the present invention. It should be noted that, for a person skilled in the art, several variations and modifications can be made without departing from the inventive concept, which falls within the scope of the present invention. Therefore, the protection scope of the present patent shall be subject to the appended claims.
The above description is only for the purpose of illustrating the preferred embodiments of the present invention and is not to be construed as limiting the invention, and any modifications, equivalents and improvements made within the spirit and principle of the present invention are intended to be included within the scope of the present invention.
Claims (7)
1. The utility model provides a bottle lid flaw detection module on high-speed production line which characterized in that includes:
the acquisition unit is used for acquiring product images of a plurality of stations;
the acquisition unit is used for intercepting a picture of a detection area through an acquired product image;
the detection unit is used for inputting the intercepted detection area picture into a pre-trained deep learning model for detection;
the judging unit is used for judging that the corresponding product is a defective product when detecting that the defect exists in the detection area picture;
and the control unit is used for controlling the rejecting device to reject the defective products.
2. The bottle cap defect detecting module set in the high-speed production line of claim 1, wherein the image capturing device for the product at the plurality of stations comprises at least 6 cameras, and when the number of the cameras is 6, the cameras are respectively arranged at the top of the first station, 4 sides of the second station and the bottom of the third station, wherein the camera at the first station is used for matching with the low-angle ring to detect black spots and deformed defects at the top of the bottle cap, the camera at the second station is used for matching with the prism, the coaxial light to detect black spots at the sides, poor combination and poor printing defects, and the camera at the third station is used for matching with the endoscope lens and the high-angle ring to detect black spots, foreign matters, deformation and hole defects inside the bottle cap.
3. The module for detecting defects of bottle caps on a high-speed production line according to claim 1 or 2, wherein the collecting unit comprises:
the inductor unit is used for inducing whether the product is in place or not through the photoelectric switch;
and the trigger subunit is used for triggering the camera to pick the picture after sensing that the product is in place.
4. The bottle cap defect detecting module of claim 1, wherein the obtaining unit specifically comprises:
the acquisition subunit is used for calculating the position of the product through a machine vision positioning algorithm after the image is acquired;
and the intercepting subunit is used for intercepting the detection area picture according to the relative coordinates of the detection area and the product position.
5. The module of claim 2, wherein the step of training the deep learning model comprises:
when the model is trained in the previous period, manually marking pictures in the detection area, and dividing the pictures into good products and a plurality of various flaws;
and inputting the manual label into the basic model for deep learning to obtain a deep learning model.
6. The module of claim 5, further comprising:
the comparison subunit is used for comparing the manual re-inspection result uploaded after the defective products are removed with the corresponding defective product removal information;
and the display subunit is used for displaying the comparison result on a human-computer interaction interface.
7. The module for detecting defects of bottle caps on a high-speed production line according to claim 1, wherein the obtaining unit further comprises:
and the brightness compensation subunit is used for detecting the ambient brightness of the station when acquiring the product image, and when the ambient brightness is lower than a preset brightness threshold value, different light sources and lenses are used for supplementing to ensure that each detection surface can clearly image.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210398891.5A CN114674830A (en) | 2022-04-16 | 2022-04-16 | Bottle cap flaw detection module on high-speed production line |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210398891.5A CN114674830A (en) | 2022-04-16 | 2022-04-16 | Bottle cap flaw detection module on high-speed production line |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN114674830A true CN114674830A (en) | 2022-06-28 |
Family
ID=82078129
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210398891.5A Pending CN114674830A (en) | 2022-04-16 | 2022-04-16 | Bottle cap flaw detection module on high-speed production line |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114674830A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115588123A (en) * | 2022-12-12 | 2023-01-10 | 河北省科学院应用数学研究所 | Milk powder cover safety disc abnormity classification method, device, terminal and storage medium |
| CN117533736A (en) * | 2023-12-12 | 2024-02-09 | 苏州奥特兰恩自动化设备有限公司 | Automatic feeding control system, method and medium |
-
2022
- 2022-04-16 CN CN202210398891.5A patent/CN114674830A/en active Pending
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115588123A (en) * | 2022-12-12 | 2023-01-10 | 河北省科学院应用数学研究所 | Milk powder cover safety disc abnormity classification method, device, terminal and storage medium |
| CN117533736A (en) * | 2023-12-12 | 2024-02-09 | 苏州奥特兰恩自动化设备有限公司 | Automatic feeding control system, method and medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US20220366558A1 (en) | System and method for ai visual inspection | |
| CN107966454A (en) | A kind of end plug defect detecting device and detection method based on FPGA | |
| CN114674830A (en) | Bottle cap flaw detection module on high-speed production line | |
| Liu et al. | Automatic detection technology of surface defects on plastic products based on machine vision | |
| CN201707305U (en) | Inspection system for inspecting quality of container caps | |
| CN105403147A (en) | Embedded bottle pre-form detection system and detection method | |
| US11657599B2 (en) | Method for detecting appearance of six sides of chip multi-layer ceramic capacitor based on artificial intelligence | |
| CN210071686U (en) | Fruit grading plant based on orthogonal binocular machine vision | |
| CN115184359A (en) | Surface defect detection system and method capable of automatically adjusting parameters | |
| CN106546600A (en) | A kind of PET bottle multi-orientation detection method and apparatus based on machine vision | |
| KR20220164124A (en) | System for inspecting product defects by type based on a deep learning model | |
| CN107576600B (en) | Quick detection method for matcha granularity grade | |
| CN114445330A (en) | Method and system for detecting appearance defects of components | |
| CN116441190A (en) | Longan detection system, method, equipment and storage medium | |
| CN114972246A (en) | Die-cutting product surface defect detection method based on deep learning | |
| JP4910800B2 (en) | Screw parts inspection device and inspection method | |
| Duong et al. | Vision inspection system for pharmaceuticals | |
| CN211741094U (en) | Surface defect detection device for transparent mirror printing and gold stamping process | |
| CN101995325A (en) | Appearance detection method and system of image sensor | |
| CN111105413B (en) | Intelligent spark plug appearance defect detection system | |
| KR20220134019A (en) | Systems and methods for imaging reflective objects | |
| CN112748110A (en) | Surface defect detection device for transparent mirror printing and gold stamping process | |
| CN110866917A (en) | Tablet type and arrangement mode identification method based on machine vision | |
| JP2005241304A (en) | Appearance inspection method | |
| CN112763506A (en) | Flaw detection method and device with AOI and AI functions |
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 |