CN111929245A - Coating defect detection device based on deep learning - Google Patents
Coating defect detection device based on deep learning Download PDFInfo
- Publication number
- CN111929245A CN111929245A CN202010932832.2A CN202010932832A CN111929245A CN 111929245 A CN111929245 A CN 111929245A CN 202010932832 A CN202010932832 A CN 202010932832A CN 111929245 A CN111929245 A CN 111929245A
- Authority
- CN
- China
- Prior art keywords
- fixed
- deep learning
- light source
- support
- industrial camera
- 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
- G01N21/01—Arrangements or apparatus for facilitating the optical investigation
-
- 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
- G01N21/898—Irregularities in textured or patterned surfaces, e.g. textiles, wood
Abstract
The invention relates to a coating defect detection device based on deep learning, which comprises a support, an industrial camera arranged on the support, an image acquisition device electrically connected with the industrial camera, a deep learning workstation electrically connected with the image acquisition device, a display device electrically connected with the deep learning workstation and a light source assembly used for emitting light to a to-be-detected gluing surface, wherein a lifting mechanism is arranged on the support, the industrial camera is fixed at the bottom end of the lifting mechanism, and the light source assembly comprises a light source adjusting seat fixed on the support, a light emitting body rotatably arranged on the light source adjusting seat and a reflecting concave mirror arranged on the outer side of the light emitting body. The invention can quickly distinguish and classify the defects of the gluing surface and has the characteristics of high recognition rate, quick detection and high efficiency.
Description
Technical Field
The invention relates to the technical field of machine vision and image processing, in particular to a coating defect detection device based on deep learning.
Background
The coating is widely applied to surface processing of various films, paper, non-woven fabrics and other materials, and is one of key links in the technical process of producing nanofiltration membranes and reverse osmosis membranes. In the coating industry, coated meltblown webs suffer from defects such as scratches, vertical lines, bright spots and wrinkles due to various factors. The presence of these defects results in reduced product yield.
At present, the domestic coating industry mainly has two methods for detecting defects in the melt-blowing process: purchasing foreign high special detection equipment and manual visual detection. The former has the defects of high price and long maintenance period, and the latter has the defects of small defects, difficulty in observation and strong subjectivity, and is essentially still dependent on personal experience of staff.
Therefore, it is desirable to provide a coating defect detecting apparatus based on deep learning, which can accurately and rapidly identify the defects of the coated surface.
Disclosure of Invention
In order to solve the above problems, an object of the present invention is to provide a coating defect detecting apparatus based on deep learning, which can accurately and rapidly identify the defects of the coated surface.
In order to achieve the purpose, the invention provides the following technical scheme: the utility model provides a coating defect detection device based on degree of depth study, it includes the support, sets up industrial camera on the support, with image acquisition equipment that industrial camera electricity is connected, with the degree of depth study workstation that image acquisition equipment electricity is connected, with display device that degree of depth study workstation electricity is connected and be used for sending the light source subassembly of light to waiting to detect the spreading surface, be equipped with elevating system on the support, industrial camera fixes the elevating system bottom, the light source subassembly is including fixing light source regulation seat, rotation on the support are installed luminous body and setting on the light source regulation seat are in the reflection concave mirror in the luminous body outside.
The lens of the industrial camera is vertically aligned with the to-be-detected gluing surface, and the distance between the lens of the industrial camera and the to-be-detected gluing surface is 500-600 mm.
Elevating system includes riser, regulating plate and the lift cylinder fixed with the support, an equal vertical fixation end plate in both ends about the board, a spout is enclosed into with two end plates to the riser, the lift cylinder is fixed on the end plate of riser upper end, a regulating plate upper end vertical fixation slider, the piston rod of lift cylinder passes the end plate of riser upper end with the slider is fixed, the regulating plate lower extreme has a fixed frame, the industry camera is fixed on fixed frame, the lift cylinder drives the slider is in reciprocate in the spout.
The deep learning workstation is also connected with a device operation parameter control mechanism used for adjusting the coating process.
The light source adjusting seat comprises a connecting plate fixed on the support, a rotary cylinder fixed on the side face of the connecting plate, a rotary disc rotatably connected to the driving end of the rotary cylinder and an inclined rod fixed on the side face of the rotary disc, the inclined rod is inclined outwards relative to the rotary disc, and the light emitting body is fixed at the tail end of the inclined rod.
The reflecting concave mirror is fixed on the outer side of the luminous body through a mirror bracket, and the mirror bracket is fixed with the inclined rod.
Compared with the prior art, the invention has the beneficial effects that: the lifting mechanism is arranged to realize the ascending or descending of the industrial camera, and is used for adjusting the distance between the industrial camera and the gluing surface and adjusting the relative position of the industrial camera and the light source assembly; by arranging the learning workstation, the defects of the glue coating surface of the melt-blown cloth are distinguished and classified by using a machine vision-based deep neural network detection method after image signals transmitted by the image acquisition equipment are received, and the recognition rate is high; through setting up luminous body, reflection concave mirror, the circular motion is done to the relative carousel center of luminous body, and the angle between the angle of incidence of the light source of control luminous body transmission and the spreading surface of waiting to detect is the acute angle, and the light source of luminous body transmission is treated and is detected the spreading surface and polish, can reach good effect of polishing.
Drawings
FIG. 1 is a general view of a coating defect detecting apparatus based on deep learning according to the present invention;
FIG. 2 is a schematic view of the support, the lifting mechanism, and the industrial camera of FIG. 1;
fig. 3 is a schematic view of the structure of the light source module in fig. 1.
Detailed Description
The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
Referring to fig. 1 to 3, the present invention is a coating defect detecting apparatus based on deep learning, which includes a bracket 100, an industrial camera 200 disposed on the bracket 100, an image capturing device 300 electrically connected to the industrial camera 200, a deep learning workstation 400 electrically connected to the image capturing device 300, a display device 500 electrically connected to the deep learning workstation 400, and a light source assembly 600 for emitting light to a to-be-detected coated surface.
Referring to fig. 1 and 2, the bracket 100 is provided with a lifting mechanism 110, and the industrial camera 200 is fixed to a bottom end of the lifting mechanism 110. The lifting mechanism 110 includes a vertical plate 111 fixed to the support 100, an adjustment plate 112, and a lifting cylinder 113.
Referring to fig. 2, an end plate 114 is vertically fixed at both upper and lower ends of a vertical plate 111, the vertical plate 111 and the two end plates 114 enclose a sliding slot 115, a lifting cylinder 113 is fixed on the end plate 114 at the upper end of the vertical plate 111, a sliding block 116 is vertically fixed at the upper end of an adjusting plate 112, and a piston rod of the lifting cylinder 113 passes through the end plate 114 at the upper end of the vertical plate 111 and is fixed with the sliding block 116.
Referring to fig. 2, the lower end of the adjusting plate 112 has a fixing frame 117, the industrial camera 200 is fixed on the fixing frame 117, and the lifting cylinder 113 drives the sliding block 116 to move up and down in the sliding groove 115, so as to realize the ascending or descending of the industrial camera 200, and adjust the distance between the industrial camera 200 and the gluing surface.
The industrial camera 200 is a CCD camera for collecting defective images of the coated surface of the meltblown and performing digital conversion of the images, wherein the CCD camera is used to generate sharp images to obtain details of the object to be measured. The industrial camera 200 is fixed at the bottom end of the lifting mechanism 110 and is vertically aligned with the glue coating surface of the melt-blown fabric, and the distance between the lens of the industrial camera 200 and the glue coating surface of the melt-blown fabric is 500-600 mm.
The industrial camera 200 is arranged at the outlet of the oven, the distance from the oven is not less than 1.5m, the oven is used for drying the glue coating surface of the melt-blown fabric, the industrial camera 200 shoots the image of the dried glue coating surface, and the shot image signal is input to the image acquisition equipment 300.
The image capturing device 300 is an image capturing card, and is configured to complete conversion from an analog signal to a digital signal, i.e., a/D conversion, and then transmit the image data to the deep learning workstation 400 through the communication module for analysis.
After receiving the image signal transmitted from the image acquisition device 300, the deep learning workstation 400 uses a deep neural network detection method based on machine vision to perform glue coating surface defect discrimination and defect classification of the melt-blown fabric, so as to transmit the glue coating surface defect to the external display device 500, thereby facilitating people to rapidly identify and find the defect of the glue coating surface of the melt-blown fabric. Referring to fig. 1, the deep learning station 400 is further connected with a control mechanism 700, and the control mechanism 700 adjusts the equipment operation parameters in the coating process according to the instruction of the deep learning station 400, so as to avoid the deterioration of the defects.
Referring to fig. 3, the light source assembly 600 includes a light source adjusting base 610 fixed on the bracket 100, a light body 620 rotatably installed on the light source adjusting base 610, and a reflective concave mirror 630 disposed outside the light body 620.
Referring to fig. 3, the light source adjusting base 610 includes a connection plate 611 fixed on the bracket 100, a rotary cylinder 612 fixed on a side surface of the connection plate 611, a rotary plate 613 rotatably connected to a driving end of the rotary cylinder 612, and a diagonal rod 614 fixed on a side surface of the rotary plate 613. The inclined rod 614 inclines outwards relative to the rotating disc 613, the light-emitting body 620 is fixed at the tail end of the inclined rod 614, the rotating cylinder 612 drives the rotating disc 613 to rotate, and the light-emitting body 620 makes circular motion relative to the center of the rotating disc 613. The concave reflecting mirror 630 is fixed to the outside of the luminous body 620 by a mirror holder 640. The frame 640 is secured to the tilt rod 614.
By arranging the light-emitting body 620 and the reflecting concave mirror 630, the light-emitting body 620 makes circular motion relative to the center of the turntable 613, the angle between the incident angle of the light source emitted by the light-emitting body 620 and the glue coating surface to be detected is controlled to be an acute angle, the light source emitted by the light-emitting body 620 polishes the surface of the glue coating surface to be detected, diffuse reflection is formed at the defect part of the glue coating surface, and the diffuse reflection is separated from the radiated light area of the smooth glue coating surface without the defect part, so that the characteristic shooting of the micro.
The basic principle of the invention is as follows: the industrial camera 200 shoots the defective image of the glue coating surface of the melt-blown fabric and completes the digital conversion of the image; the image acquisition device 300 completes the conversion from the analog signal to the digital signal, and then transmits the image data to the deep learning workstation 400 for analysis through the communication module; the deep learning workstation 400 uses a deep neural network detection method based on machine vision to perform glue coating surface defect discrimination and defect classification of the melt-blown fabric, so that the glue coating surface defect is transmitted to the external display device 500, and people can conveniently and rapidly identify and find the defect of the glue coating surface of the melt-blown fabric.
Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that changes, modifications, substitutions and alterations can be made in these embodiments without departing from the principles and spirit of the invention, the scope of which is defined in the appended claims and their equivalents.
Claims (6)
1. The utility model provides a coating defect detection device based on deep learning which characterized in that: the light source device comprises a support, an industrial camera arranged on the support, an image acquisition device electrically connected with the industrial camera, a deep learning workstation electrically connected with the image acquisition device, a display device electrically connected with the deep learning workstation and a light source assembly used for emitting light to a gluing surface to be detected, wherein a lifting mechanism is arranged on the support, the industrial camera is fixed at the bottom end of the lifting mechanism, and the light source assembly comprises a light source adjusting seat fixed on the support, a light body rotatably arranged on the light source adjusting seat and a reflecting concave mirror arranged outside the light body.
2. The coating defect detection device based on deep learning of claim 1, characterized in that: the lens of the industrial camera is vertically aligned with the to-be-detected gluing surface, and the distance between the lens of the industrial camera and the to-be-detected gluing surface is 500-600 mm.
3. The coating defect detection device based on deep learning of claim 1, characterized in that: elevating system includes riser, regulating plate and the lift cylinder fixed with the support, an equal vertical fixation end plate in both ends about the board, a spout is enclosed into with two end plates to the riser, the lift cylinder is fixed on the end plate of riser upper end, a regulating plate upper end vertical fixation slider, the piston rod of lift cylinder passes the end plate of riser upper end with the slider is fixed, the regulating plate lower extreme has a fixed frame, the industry camera is fixed on fixed frame, the lift cylinder drives the slider is in reciprocate in the spout.
4. The coating defect detection device based on deep learning of claim 1, characterized in that: the deep learning workstation is also connected with a device operation parameter control mechanism used for adjusting the coating process.
5. The coating defect detection device based on deep learning of claim 1, characterized in that: the light source adjusting seat comprises a connecting plate fixed on the support, a rotary cylinder fixed on the side of the connecting plate, a rotary disc rotatably connected to the driving end of the rotary cylinder, and an inclined rod fixed on the side face of the rotary disc, the inclined rod inclines outwards relative to the rotary disc, and the light-emitting body is fixed at the tail end of the inclined rod.
6. The coating defect detection device based on deep learning of claim 5, wherein: the reflecting concave mirror is fixed on the outer side of the luminous body through a mirror bracket, and the mirror bracket is fixed with the inclined rod.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010932832.2A CN111929245A (en) | 2020-09-08 | 2020-09-08 | Coating defect detection device based on deep learning |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010932832.2A CN111929245A (en) | 2020-09-08 | 2020-09-08 | Coating defect detection device based on deep learning |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN111929245A true CN111929245A (en) | 2020-11-13 |
Family
ID=73309215
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010932832.2A Pending CN111929245A (en) | 2020-09-08 | 2020-09-08 | Coating defect detection device based on deep learning |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111929245A (en) |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN202330294U (en) * | 2011-11-08 | 2012-07-11 | 中国科学院深圳先进技术研究院 | Surface defect detection device based on machine vision |
| CN205643191U (en) * | 2016-04-08 | 2016-10-12 | 胡文港 | Fabric quality on -line measuring equipment based on machine vision |
| CN207866727U (en) * | 2018-01-19 | 2018-09-14 | 西安工程大学 | Color based on GPU deep learning work stations knits shirt cut-parts defect detection device |
| CN110658202A (en) * | 2019-09-30 | 2020-01-07 | 贵州航天云网科技有限公司 | Industrial component appearance defect detection method based on deep learning |
| CN110726735A (en) * | 2019-09-03 | 2020-01-24 | 北京精思博智科技有限公司 | Full-automatic circuit board defect detection system and method based on deep learning |
| CN111103292A (en) * | 2019-12-31 | 2020-05-05 | 深圳市智信精密仪器有限公司 | Mobile phone defect inspection visual device based on deep learning |
-
2020
- 2020-09-08 CN CN202010932832.2A patent/CN111929245A/en active Pending
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN202330294U (en) * | 2011-11-08 | 2012-07-11 | 中国科学院深圳先进技术研究院 | Surface defect detection device based on machine vision |
| CN205643191U (en) * | 2016-04-08 | 2016-10-12 | 胡文港 | Fabric quality on -line measuring equipment based on machine vision |
| CN207866727U (en) * | 2018-01-19 | 2018-09-14 | 西安工程大学 | Color based on GPU deep learning work stations knits shirt cut-parts defect detection device |
| CN110726735A (en) * | 2019-09-03 | 2020-01-24 | 北京精思博智科技有限公司 | Full-automatic circuit board defect detection system and method based on deep learning |
| CN110658202A (en) * | 2019-09-30 | 2020-01-07 | 贵州航天云网科技有限公司 | Industrial component appearance defect detection method based on deep learning |
| CN111103292A (en) * | 2019-12-31 | 2020-05-05 | 深圳市智信精密仪器有限公司 | Mobile phone defect inspection visual device based on deep learning |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN103323457B (en) | Fruit appearance defect real-time online detection system and detection method | |
| CN104101608B (en) | Intelligent detecting device for detecting defects of multi-type irregularly shaped product | |
| CN203124215U (en) | Frame sealant coating machine | |
| CN211839085U (en) | Machine vision detection device | |
| CN113252561A (en) | Cookware surface defect detection system and method | |
| CN112730463A (en) | Device and method for detecting defects of glass cover plate based on coaxial grating light source | |
| CN109772721A (en) | Engine commutator open defect automatic online vision inspection apparatus and detection method | |
| CN106353336A (en) | Lens coating automatic detection system | |
| CN115219512A (en) | PE film surface crystal point detection equipment | |
| CN109596633B (en) | Light guide plate defect detection equipment and detection method | |
| CN111929245A (en) | Coating defect detection device based on deep learning | |
| CN212759762U (en) | Online automatic detection system for appearance defects of washing machine | |
| CN209542487U (en) | Lens defects detection device | |
| CN209565257U (en) | Engine commutator open defect automatic online vision inspection apparatus | |
| CN111729860A (en) | Online automatic detection system and method for appearance defects of washing machine | |
| CN208860768U (en) | A kind of ball Sphere Detector | |
| CN207689390U (en) | A kind of bottleneck inner wall defect on-line measuring device | |
| CN217766111U (en) | Automobile door trim appearance detection device | |
| CN109932367A (en) | A kind of curved surface carved image acquisition device | |
| CN215218599U (en) | Multi-surface appearance detection mechanism | |
| CN205361990U (en) | Motor for cell phone brush automatic checkout device | |
| CN208984547U (en) | Equipment for on-line checking transparent substrate defect | |
| CN208636217U (en) | A kind of fresh flower detection device | |
| CN219830881U (en) | Bottled liquid impurity detection equipment based on rim plate | |
| CN106092894B (en) | A kind of membrane filtration separation detecting device |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| WD01 | Invention patent application deemed withdrawn after publication | ||
| WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20201113 |