CN117517348B - Surface defect detection system based on microcrystalline glass panel finished product - Google Patents
Surface defect detection system based on microcrystalline glass panel finished product Download PDFInfo
- Publication number
- CN117517348B CN117517348B CN202311509599.7A CN202311509599A CN117517348B CN 117517348 B CN117517348 B CN 117517348B CN 202311509599 A CN202311509599 A CN 202311509599A CN 117517348 B CN117517348 B CN 117517348B
- Authority
- CN
- China
- Prior art keywords
- frame
- image
- ceramic panel
- glass ceramic
- detection
- 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
- 238000001514 detection method Methods 0.000 title claims abstract description 113
- 230000007547 defect Effects 0.000 title claims abstract description 74
- 239000011521 glass Substances 0.000 title claims description 27
- 239000002241 glass-ceramic Substances 0.000 claims abstract description 72
- 230000007246 mechanism Effects 0.000 claims abstract description 15
- 238000004140 cleaning Methods 0.000 claims description 47
- 230000005540 biological transmission Effects 0.000 claims description 25
- 238000012545 processing Methods 0.000 claims description 25
- 238000013528 artificial neural network Methods 0.000 claims description 11
- 238000000605 extraction Methods 0.000 claims description 8
- 238000003709 image segmentation Methods 0.000 claims description 6
- 238000007781 pre-processing Methods 0.000 claims description 6
- 230000011218 segmentation Effects 0.000 claims description 6
- 238000013527 convolutional neural network Methods 0.000 claims description 5
- 238000004364 calculation method Methods 0.000 claims description 3
- 238000001914 filtration Methods 0.000 claims description 3
- 238000000926 separation method Methods 0.000 claims description 3
- 238000000034 method Methods 0.000 abstract description 10
- 239000000463 material Substances 0.000 abstract description 4
- 230000000295 complement effect Effects 0.000 abstract description 3
- 239000000428 dust Substances 0.000 description 10
- 239000012535 impurity Substances 0.000 description 10
- 230000008569 process Effects 0.000 description 4
- 230000004044 response Effects 0.000 description 4
- 230000003044 adaptive effect Effects 0.000 description 3
- 230000000694 effects Effects 0.000 description 3
- 230000033001 locomotion Effects 0.000 description 3
- 230000003287 optical effect Effects 0.000 description 3
- 238000012549 training Methods 0.000 description 3
- 239000003795 chemical substances by application Substances 0.000 description 2
- 239000013078 crystal Substances 0.000 description 2
- 238000013461 design Methods 0.000 description 2
- 238000010586 diagram Methods 0.000 description 2
- 239000013081 microcrystal Substances 0.000 description 2
- 239000000203 mixture Substances 0.000 description 2
- 230000001502 supplementing effect Effects 0.000 description 2
- 238000004458 analytical method Methods 0.000 description 1
- 239000006121 base glass Substances 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000007664 blowing Methods 0.000 description 1
- 239000002131 composite material Substances 0.000 description 1
- 238000002425 crystallisation Methods 0.000 description 1
- 230000008025 crystallization Effects 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000018109 developmental process Effects 0.000 description 1
- 238000009826 distribution Methods 0.000 description 1
- 239000000284 extract Substances 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
- 238000003062 neural network model Methods 0.000 description 1
- 239000002245 particle Substances 0.000 description 1
- 238000003904 radioactive pollution Methods 0.000 description 1
- 238000010187 selection method Methods 0.000 description 1
- 238000005245 sintering Methods 0.000 description 1
- 239000000126 substance Substances 0.000 description 1
- 230000000153 supplemental effect Effects 0.000 description 1
- 238000010408 sweeping Methods 0.000 description 1
- 238000012360 testing method Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/136—Segmentation; Edge detection involving thresholding
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B08—CLEANING
- B08B—CLEANING IN GENERAL; PREVENTION OF FOULING IN GENERAL
- B08B13/00—Accessories or details of general applicability for machines or apparatus for cleaning
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B08—CLEANING
- B08B—CLEANING IN GENERAL; PREVENTION OF FOULING IN GENERAL
- B08B5/00—Cleaning by methods involving the use of air flow or gas flow
- B08B5/02—Cleaning by the force of jets, e.g. blowing-out cavities
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B65—CONVEYING; PACKING; STORING; HANDLING THIN OR FILAMENTARY MATERIAL
- B65G—TRANSPORT OR STORAGE DEVICES, e.g. CONVEYORS FOR LOADING OR TIPPING, SHOP CONVEYOR SYSTEMS OR PNEUMATIC TUBE CONVEYORS
- B65G49/00—Conveying systems characterised by their application for specified purposes not otherwise provided for
- B65G49/05—Conveying systems characterised by their application for specified purposes not otherwise provided for for fragile or damageable materials or articles
- B65G49/06—Conveying systems characterised by their application for specified purposes not otherwise provided for for fragile or damageable materials or articles for fragile sheets, e.g. glass
- B65G49/062—Easels, stands or shelves, e.g. castor-shelves, supporting means on vehicles
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B65—CONVEYING; PACKING; STORING; HANDLING THIN OR FILAMENTARY MATERIAL
- B65G—TRANSPORT OR STORAGE DEVICES, e.g. CONVEYORS FOR LOADING OR TIPPING, SHOP CONVEYOR SYSTEMS OR PNEUMATIC TUBE CONVEYORS
- B65G49/00—Conveying systems characterised by their application for specified purposes not otherwise provided for
- B65G49/05—Conveying systems characterised by their application for specified purposes not otherwise provided for for fragile or damageable materials or articles
- B65G49/06—Conveying systems characterised by their application for specified purposes not otherwise provided for for fragile or damageable materials or articles for fragile sheets, e.g. glass
- B65G49/063—Transporting devices for sheet glass
-
- 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
- G01N21/8851—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
- 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
- G01N21/95—Investigating the presence of flaws or contamination characterised by the material or shape of the object to be examined
- G01N21/958—Inspecting transparent materials or objects, e.g. windscreens
-
- 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
- G01N2021/8887—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 based on image processing techniques
-
- 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
- G06T2207/20021—Dividing image into blocks, subimages or windows
Landscapes
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Health & Medical Sciences (AREA)
- Engineering & Computer Science (AREA)
- Chemical & Material Sciences (AREA)
- Analytical Chemistry (AREA)
- Biochemistry (AREA)
- General Health & Medical Sciences (AREA)
- Immunology (AREA)
- Pathology (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Theoretical Computer Science (AREA)
- Signal Processing (AREA)
- Investigating Materials By The Use Of Optical Means Adapted For Particular Applications (AREA)
Abstract
The invention relates to advanced nonmetallic material technology, and particularly discloses a surface defect detection system based on a glass ceramic panel finished product, which comprises an annular outer frame and a detection unit, wherein the detection unit comprises an annular detection frame coaxially arranged in the outer frame, the upper part of the annular detection frame is in sliding connection with the outer frame through a sliding mechanism, the annular detection frame is divided into a first semi-ring and a second semi-ring from top to bottom by taking the center position of the annular detection frame as a reference, a plurality of image acquisition devices are arranged on the inner ring surface of the first semi-ring at equal intervals, a plurality of complementary light sources are arranged on the inner ring surface of the second semi-ring at equal intervals, the image acquisition devices and the complementary light sources are in one-to-one correspondence, and the surface images of the glass ceramic panel are acquired through the image acquisition devices, so that defect detection is carried out on the surface of the glass ceramic panel while the glass ceramic panel is conveyed, the detection procedure is greatly reduced, the detection man-hour is shortened, and the efficiency is higher.
Description
Technical Field
The invention relates to the technical field of advanced nonmetallic materials, in particular to a surface defect detection system based on a glass ceramic panel finished product.
Background
The microcrystalline glass is base glass with specific composition and added with crystal nucleus agent (or not added with crystal nucleus agent), and is crystallized and heat treated at certain temperature to form great amount of micro crystals homogeneously in the glass to form compact multiphase composite of microcrystalline phase and glass phase. Microcrystalline glass is a mixture of glass and microcrystals made from appropriate glass particles by sintering and crystallization. The material has the advantages of hard and compact texture, no pollution in the production process, no radioactive pollution of the product, and good thermal, chemical, biological, optical and electrical properties, and is a novel advanced environment-friendly nonmetallic material.
For the glass ceramic plate, defect detection is required to be carried out on the finished glass panel product after the glass ceramic plate is prepared so as to ensure the quality of the finished glass ceramic plate product, but at present, when the glass ceramic plate product is detected, the defect detection is carried out on the glass panel by manually holding an optical detection instrument, the detection method is time-consuming and labor-consuming, the detection consistency is poor, the detection precision is affected, and the non-contact detection is realized on the glass panel by the optical detection system, so that a plurality of inconveniences still exist in actual use; when detecting, it can only carry out single face by face to the glass ceramic panel and detect, and is difficult to carry out the omnidirectional to the upper and lower surface of glass panel, causes its detection efficiency lower, and when detecting, is difficult to clean the glass ceramic panel in advance to influence the detection effect of glass ceramic panel.
Disclosure of Invention
The invention aims to provide a surface defect detection system based on a glass ceramic panel finished product, so as to solve the problems in the prior art.
The invention is realized by the following technical scheme:
The surface defect detection system based on the microcrystalline glass panel finished product comprises an annular outer frame and a detection unit, wherein the detection unit comprises an annular detection frame coaxially arranged in the outer frame, the upper part of the annular detection frame is in sliding connection with the outer frame through a sliding mechanism, the annular detection frame is divided into a first half ring and a second half ring from top to bottom by taking the center position of the annular detection frame as a reference, a plurality of image acquisition devices are equidistantly arranged on the inner ring surface of the first half ring, a plurality of supplementary light sources are equidistantly arranged on the inner ring surface of the second half ring, the image acquisition devices are in one-to-one correspondence with the plurality of supplementary light sources, the image acquisition devices acquire surface images of the microcrystalline glass panel, and the plurality of supplementary light sources are used for carrying out light source irradiation on the microcrystalline glass panel when the image acquisition devices acquire the surface images of the microcrystalline glass panel so as to realize light supplementing when the image acquisition devices acquire the images of the microcrystalline glass panel, and the acquired images are clearer;
The detection unit further comprises an image recognition processing module which is arranged on the outer frame and is in signal connection with the image acquisition devices, wherein the image recognition processing module is used for receiving surface image information of the glass-ceramic panel transmitted by the image acquisition devices and detecting and recognizing defects of the glass-ceramic panel based on the surface image information;
The glass ceramic panel conveying device is characterized in that a base is further arranged at the bottom of the outer frame, a conveying frame is further arranged on the upper portion of the base, the conveying frame is of a rectangular frame structure, a plurality of groups of conveying rollers used for conveying glass ceramic panels are arranged in the conveying frame along the length direction of the conveying frame, the conveying rollers are rotatably connected with the conveying frame through rotating shafts, gaps exist between the conveying rollers, the outer frame is located in one of the gaps, the center position of the outer frame is located on a uniform horizontal plane with the conveying rollers, the bottom of the outer frame is connected with the base, a driving mechanism is arranged on one side of the inner portion of the conveying frame, and the driving mechanism comprises a power unit used for driving the conveying rollers to rotate, and a transmission unit connected with the power unit and used for driving the annular detection frame to rotate in the outer frame.
It should be noted that, this scheme is based on the defect of current to glass ceramic panel surface defect technique, specially propose a surface defect detecting system based on glass ceramic panel finished product, it is through outer frame, annular detection frame, detecting element and transportation frame's setting, after glass ceramic panel prepares, when carrying glass ceramic panel through the transportation frame, annular detection frame can carry out rotary motion at outer frame through slide mechanism in drive unit's drive, and because annular detection frame divide into first semi-ring and second semi-ring, and first semi-ring and second semi-ring are equipped with image acquisition device and the supplementary light source that corresponds each other respectively, therefore when the transportation frame carries glass ceramic panel, annular detection frame can rotate thereupon, and then carry out comprehensive collection to glass ceramic panel's upper and lower surface image through the image acquisition device that its inner ring set up, with this greatly promoted detecting system's detection scope, and improve detection efficiency, in this scheme, simultaneously, based on above-mentioned structure cooperation setting, can realize carrying glass ceramic panel carries out the simultaneous defect detection to its surface, consequently, greatly reduced the detection process, and shortened and detected work hour more high in efficiency.
Further, the image recognition processing module comprises a preprocessing module, an image segmentation module and a defect recognition module which are sequentially connected through signals;
The preprocessing module is used for carrying out image enhancement processing after receiving surface original image information;
the image segmentation module is used for carrying out thresholding segmentation on the surface image information subjected to image enhancement processing, obtaining a gray threshold value of the surface image information through image gray level statistics calculation, and carrying out binarization segmentation on the image based on the gray threshold value so as to obtain binarized image information;
the defect identification module is used for extracting defect characteristic information in the binarized image information through the neural network so as to carry out defect judgment and identification, and outputting the defect judgment and identification to an external main control computer.
Based on the module design, the glass ceramic panel surface image processing device is convenient for processing original image information after the glass ceramic panel surface image is acquired by the image acquisition device, so that the processed image information is convenient for defect detection and identification, and the detection accuracy is improved.
Further, the image enhancement processing specifically includes filtering noise in the image information by a median filter, performing texture expression and separation on the image information after denoising based on a Gabor filter to obtain texture feature histogram information, performing adaptive histogram equalization on an original image, and fusing the image subjected to the adaptive histogram equalization with a histogram of the texture feature obtained before to obtain image information after contrast enhancement. Based on the processing, the contrast of the image can be enhanced while the texture characteristics of the original image are maintained, so that the image is clearer and easy to analyze and identify.
Preferably, the neural network is a fast R-CNN neural network structure comprising a feature extraction network, a region suggestion network, a region feature alignment, a defect classification and regression network, and an output layer. It should be noted that, the neural network model is an end-to-end target detection network, and compared with the traditional two-stage method of defect feature extraction and classifier model, the end-to-end target detection model can learn the feature and identification information of the defect from the original image more directly, integrate the defect feature extraction and defect identification into a neural network structure, and directly perform defect identification without resorting to the classifier model. The method can learn the characteristics and the identification information of the defects from the original image more directly, and has better performance and robustness.
Specifically, the power unit includes: the conveying device comprises a driving motor, a driving wheel and driven wheels, wherein one side of the driving motor is connected with a conveying frame, an output shaft is arranged at the output end of the driving wheel, the driving wheels are sleeved on one end part of the output shaft, the driven wheels are provided with a plurality of rotating shafts which are sleeved on one ends of a plurality of groups of conveying rollers one by one, and the driven wheels are in transmission connection with the driving wheels through transmission belts. Based on the structure, the power unit can provide driving force for a plurality of groups of conveying rollers on the conveying frame, so that the conveying rollers are driven to rotate, and the microcrystalline glass panel is conveyed.
Preferably, the transmission unit includes a driven gear and a rack; the driven gear is keyed on an output shaft of the output end of the driving motor, the rack is annular and is positioned on the outer surface of the annular detection frame, and the rack is in meshed connection with the transmission gear. Through the structure, when the power unit outputs power, the transmission unit can transmit the power to the annular detection frame, so that the annular detection frame rotates in the outer frame, and the microcrystalline glass panel on the conveying roller is subjected to 360-degree rotation detection and identification through the rotation of the annular detection frame.
Still preferably, the transportation frame is further provided with a cleaning table, the cleaning table is provided with a cleaning fan, the cleaning fan is rotatably connected with the cleaning table through a rotating piece at the bottom of the cleaning fan, and the cleaning fan rotates on the cleaning table through the rotating piece and cleans the surface of the glass ceramic panel on the conveying roller in a fan shape.
It should be noted that, after the glass ceramic panel is prepared, dust impurities are inevitably attached to the glass panel in the transportation process, so that when the defect detection and identification are performed, the attached dust impurities can greatly influence the definition of image acquisition, and further cause the defect misidentification to occur when the defect detection and identification are performed by the detection system, and further influence the accuracy of the detection structure. In view of this, in this scheme, especially through setting up the cleaning fan on cleaning the platform, so when the conveyer is carrying glass ceramic panel, cleaning the fan and can begin the work and blow the dust impurity that adheres to the glass ceramic panel upper surface, thereby avoid dust impurity to adhere to on the glass ceramic panel, and influence final defect detection recognition result, simultaneously, the cleaning fan rotates with cleaning the platform through the rotating member, so when cleaning the fan and beginning work, its accessible rotating member rotates, so as to realize carrying out the sweeping that is fan-shaped region to the glass ceramic panel upper surface, thereby further promote the cleaning region who cleans the fan, thereby greatly improve defect detection system's result of use.
Compared with the prior art, the invention has the following advantages and beneficial effects:
1. According to the invention, after the microcrystalline glass panel is prepared, the microcrystalline glass panel is conveyed through the conveying frame, the annular detecting frame can be driven by the transmission unit to rotate in the outer frame through the sliding mechanism, and as the annular detecting frame is divided into the first semi-ring and the second semi-ring, and the image acquisition devices and the complementary light sources which correspond to each other are respectively arranged in the first semi-ring and the second semi-ring, the annular detecting frame can rotate along with the microcrystalline glass panel when the conveying frame conveys the microcrystalline glass panel, and then the image acquisition devices arranged on the inner ring of the annular detecting frame comprehensively acquire the upper surface image and the lower surface image of the microcrystalline glass panel, so that the detection range of the detecting system is greatly improved, and the detection efficiency is improved.
2. According to the invention, the cleaning fan is arranged on the cleaning table, so that when the conveyor conveys the glass ceramic panel, the cleaning fan can start to work and blow dust impurities attached to the upper surface of the glass ceramic panel, so that the dust impurities are prevented from attaching to the glass ceramic panel to influence the final defect detection and identification result, and meanwhile, the cleaning fan is rotationally connected with the cleaning table through the rotating piece, so that when the cleaning fan starts to work, the cleaning fan can rotate through the rotating piece to realize blowing the upper surface of the glass ceramic panel in a sector area, and the cleaning area of the cleaning fan is further improved, so that the use effect of the defect detection system is greatly improved.
Drawings
The accompanying drawings, which are included to provide a further understanding of embodiments of the application and are incorporated in and constitute a part of this specification, illustrate embodiments of the application and together with the description serve to explain the principles of the application. In the drawings:
FIG. 1 is a schematic view of the structure of the present invention, with dashed arrows indicating the direction of rotation;
FIG. 2 is a schematic top view of the present invention, which is intended to show a specific structure;
FIG. 3 is a block diagram illustrating an image recognition processing module according to the present invention;
FIG. 4 is a schematic view of the structure of the outer frame and the annular detecting frame of the present invention;
FIG. 5 is a schematic view of a partial structure of a power unit of the present invention, intended to show its specific structure;
FIG. 6 is a schematic side view of the annular test rack of the present invention;
FIG. 7 is a schematic view of the structure of the transmission unit of FIG. 6, partially enlarged, according to the present invention;
fig. 8 is a schematic structural view of the cleaning fan of the present invention, and the dashed arrow is intended to show the operation state thereof.
The reference numerals are represented as follows: 1. an outer frame; 10. an annular detection frame; 100. an image acquisition device; 101. supplementing a light source; 2. a base; 20. a transport rack; 21. a conveying roller; 220. a power unit; 2200. a driving motor; 2201. an output shaft; 2202. a driving wheel; 2203. driven wheel; 2204. a drive belt; 221. a transmission unit; 2210. a transmission gear; 2211. a rack; 30. a rotating member; 300. a rotating motor; 301. a rotating wheel; 302. a vertical rod; 303. a cross bar; 304. rectangular grooves; 305. a rotating bearing; 31. cleaning a fan; 4. a sliding mechanism; 40. a chute; 41. a pulley; 5. glass ceramic panels.
Detailed Description
For the purpose of making apparent the objects, technical solutions and advantages of the present invention, the present invention will be further described in detail with reference to the following examples and the accompanying drawings, wherein the exemplary embodiments of the present invention and the descriptions thereof are for illustrating the present invention only and are not to be construed as limiting the present invention. It should be noted that the present invention is already in a practical development and use stage.
Example 1
Referring to fig. 1 to 3, the embodiment discloses a surface defect detection system based on a glass ceramic panel finished product, which comprises an annular outer frame 1 and a detection unit, wherein the detection unit comprises an annular detection frame 10 coaxially arranged in the outer frame 1, the upper part of the annular detection frame 10 is in sliding connection with the outer frame 1 through a sliding mechanism 4, the annular detection frame 10 is divided into a first semi-ring and a second semi-ring from top to bottom by taking the center position of the annular detection frame as a reference, a plurality of image acquisition devices 100 are equidistantly arranged on the inner ring surface of the first semi-ring, a plurality of supplementary light sources 101 are equidistantly arranged on the inner ring surface of the second semi-ring, the plurality of image acquisition devices 100 are in one-to-one correspondence with the plurality of supplementary light sources 101, the plurality of supplementary light sources 101 are used for carrying out light source irradiation on the glass ceramic panel 5 when the image acquisition devices 100 acquire the surface image of the glass ceramic panel 5, so that the image acquisition devices 100 can acquire the image of the glass ceramic panel 5 clearly, and the image acquisition is more convenient;
the detection unit further comprises an image recognition processing module which is arranged on the outer frame 1 and is in signal connection with the plurality of image acquisition devices 100, wherein the image recognition processing module is used for receiving surface image information of the glass ceramic panel 5 transmitted by the image acquisition devices 100 and detecting and recognizing defects of the glass ceramic panel 5 based on the surface image information;
the bottom of outer frame 1 still is equipped with base 2 the upper portion of base 2 still is equipped with transport frame 20, transport frame 20 is rectangular frame structure transport frame 20's inside is equipped with multiunit along its length direction and is used for carrying glass ceramic panel 5's conveying roller 21, multiunit conveying roller 21's both ends all rotate through the pivot and transport frame 20 and be connected, and all have the clearance between the multiunit conveying roller 21, outer frame 1 is located one of them clearance, and outer frame 1's centre of a circle position department and conveying roller 21 are in unified horizontal plane, outer frame 1's bottom is connected with base 2, just transport frame 20's inside one side installs actuating mechanism, actuating mechanism includes the power unit 220 that is used for driving multiunit conveying roller 21 pivoted to and is connected, is used for driving annular detection frame 10 and carries out pivoted drive unit 221 in outer frame 1 with the transmission.
Based on the above embodiment, this scheme is through outer frame 1, annular detection frame 10, detecting element and transportation frame 20's setting, after glass ceramic panel 5 prepares, when carrying glass ceramic panel 5 through transportation frame 20, annular detection frame 10 can carry out rotary motion at outer frame 1 through slide mechanism 4in the drive of drive unit 221, and because annular detection frame 10 divide into first semi-ring and second semi-ring, and be equipped with image acquisition device 100 and the supplementary light source 101 that corresponds each other respectively in first semi-ring and the second semi-ring, therefore when transportation frame 20 carries glass ceramic panel 5, annular detection frame 10 can rotate along with it, and then the image acquisition device 100 that sets up through its inner ring carries out comprehensive collection to glass ceramic panel 5's upper and lower surface image, thereby promote detecting system's detection scope greatly, and improve detection efficiency, simultaneously, in this scheme, can realize carrying glass ceramic panel 5 simultaneously, carry out defect detection to its surface, consequently, greatly reduced the detection process, and shortened the detection man-hour, make its efficiency higher.
In the above scheme, the image recognition processing module comprises a preprocessing module, an image segmentation module and a defect recognition module which are sequentially connected through signals;
The preprocessing module is used for carrying out image enhancement processing after receiving surface original image information, specifically, the image enhancement processing comprises filtering noise in the image information through a median filter, carrying out texture expression and separation on the image information after denoising processing based on a Gabor filter to obtain texture characteristic histogram information, carrying out self-adaptive histogram equalization on the original image, fusing the image subjected to the self-adaptive histogram equalization with a histogram of texture characteristics obtained before to obtain image information after contrast enhancement, so that the contrast of the image is enhanced on the premise of keeping the texture characteristics of the original image, the image can be clearer and easier to analyze and recognize, and further, after the Gabor filter is applied, a Gabor filter response image can be obtained, and can be used for describing and separating textures through counting the texture characteristics in the response image, specifically, carrying out region division on the Gabor filter response image, dividing the image into a plurality of sub-regions, then calculating the histogram of the Gabor response image according to the characteristic description of pixel value distribution condition of each sub-region; and combining the histograms of the sub-areas to obtain the histogram of the texture characteristics of the whole image. Thus, through the histogram of the texture features, further analysis and processing of the image, such as texture classification, texture recognition, etc., can be performed;
the image segmentation module is used for carrying out thresholding segmentation on the surface image information subjected to image enhancement processing, obtaining a gray threshold value of the surface image information through image gray level statistics calculation, and carrying out binarization segmentation on the image based on the gray threshold value so as to obtain binarized image information; it should be noted that, when calculating the gray threshold, it is calculated based on the Otsu algorithm, further, the Otsu algorithm is an adaptive threshold selection method based on maximization of the inter-class variance, and an optimal gray threshold is determined by maximizing the inter-class variance of the image, specifically, traversing all possible thresholds, classifying the image into two categories, wherein the pixels below or equal to the threshold are one category, the pixels above the threshold are the other category, calculating the inter-class variance corresponding to each threshold, and selecting the threshold with the maximum inter-class variance as the optimal gray threshold;
The defect identification module is used for extracting defect characteristic information in the binarized image information through a neural network to carry out defect judgment and identification, and outputting the defect judgment and identification to an external main control computer, wherein the neural network is of a Faster R-CNN neural network structure and comprises a characteristic extraction network, a region suggestion network, a region characteristic alignment, a defect classification and regression network and an output layer, and the input layer is used for receiving the processed image information as input; the feature extraction layer extracts defect features in the input image based on the convolutional neural network; the target detection layer predicts the position and the category of the defect through an additional layer by utilizing an output characteristic diagram of the convolutional neural network; finally, the output layer is used for outputting the position and category information of the defect. In the training stage, training can be performed on the end-to-end target detection model by using training data with labels, and network weight is adjusted by optimizing a loss function so that defects can be accurately detected and identified; compared with the traditional two-stage method of defect feature extraction and classifier model, the end-to-end target detection model can learn the feature and identification information of the defects from the original image more directly, integrate the defect feature extraction and defect identification into a neural network structure, and directly perform defect identification without resorting to the classifier model. The method can learn the characteristics and the identification information of the defects from the original image more directly, and has better performance and robustness.
Therefore, in the scheme, through the module design, the image acquisition device 100 is convenient to process the original image information after acquiring the surface image of the glass ceramic panel 5, so that the processed image information is convenient to detect and identify defects, and the detection accuracy is greatly improved.
Example 2
As shown in fig. 1 to 7, in this embodiment, based on embodiment 1, in order to further ensure that the defect detection system can synchronously detect defects on the surface of the glass-ceramic panel 5 when the glass-ceramic panel 5 is conveyed by the conveying frame, the power unit 220 is specifically described herein, and the power unit 220 includes: the automatic conveying device comprises a driving motor 2200, a driving wheel 2202 and driven wheels 2203, wherein one side of the driving motor 2200 is connected with a conveying frame 20, an output shaft 2201 is arranged at the output end of the driving motor, the driving wheel 2202 is sleeved on one end of the output shaft 2201, the driven wheels 2203 are provided with a plurality of driven wheels and are sleeved on rotating shafts at one ends of a plurality of groups of conveying rollers 21 one by one, and the driven wheels 2203 are in transmission connection with the driving wheel 2202 through a transmission belt 2204. Based on the above structure, the power unit 220 can provide driving force for the multiple groups of conveying rollers 21 on the conveying frame 20, so that the conveying rollers 21 are driven to rotate, and the glass ceramic panel 5 is conveyed. Preferably, the transmission unit 221 includes a driven gear and rack 2211; the driven gear is keyed on the output shaft 2201 at the output end of the driving motor 2200, the rack 2211 is ring-shaped, and is positioned on the outer surface of the ring-shaped detection frame 10, and the rack 2211 is in meshed connection with the transmission gear 2210. With the above structure, when the power unit 220 outputs power, the transmission unit 221 can transmit the power to the annular detection frame 10, so that the annular detection frame 10 rotates in the outer frame 1, and the rotation of the annular detection frame 10 is used for 360-degree rotation detection and identification of the glass ceramic panel 5 on the conveying roller 21.
Specifically, when the driving motor 2200 drives the driving wheel 2202 to rotate through the output shaft 2201, the driving wheel 2202 can drive the driven wheel 2203 to rotate through the transmission belt 2204, so that the driven wheel 2203 drives the multiple groups of conveying rollers 21 to rotate, and the conveying rollers 21 rotate and then convey the glass ceramic panel 5 thereon; meanwhile, when the output shaft 2201 of the driving motor 2200 rotates, the transmission gear 2210 also rotates, so that after the transmission gear 2210 rotates, the annular detection frame 10 is driven by the rack 2211 to rotate in the outer frame 1, and the annular detection frame 10 rotates and acquires the surface image of the glass ceramic panel 5 through the image acquisition device 100 of the inner ring surface of the annular detection frame while the conveying roller 21 conveys the glass ceramic panel 5; it should be noted that, in order to facilitate the sliding of the annular detecting frame 10 in the outer frame 1, the sliding mechanism 4 is specially provided between the two, and the specific structure includes a sliding groove 40 formed on the inner ring surface of the outer frame 1, and a plurality of pulleys 41 installed on the outer ring surface of the annular detecting frame 10 and extending into the sliding groove 40 to be in sliding fit with the sliding groove.
Example 3
As shown in fig. 1 to 8, according to the above embodiment, after the glass ceramic panel 5 is manufactured, dust impurities are inevitably attached to the glass panel during transportation, so that when defect detection and identification are performed, the attached dust impurities greatly affect the definition of image acquisition, and further cause error defect identification when the defect detection and identification are performed by the detection system, thereby affecting the accuracy of the detection structure; therefore, it is further preferable that the transportation frame 20 is further provided with a cleaning table, the cleaning table is provided with a cleaning fan 31, the cleaning fan 31 is rotatably connected with the cleaning table through a rotating member 30 at the bottom of the cleaning fan 31, and the cleaning fan 31 rotates on the cleaning table through the rotating member 30 and cleans the surface of the glass ceramic panel 5 on the transportation roller 21 in a fan shape.
The structure of the rotating member 30 is described herein, which includes a rotating bearing 305 disposed on the cleaning table and connected to the bottom of the cleaning fan 31, a rotating motor 300 disposed at a position on one side of the rotating bearing 305, an output end of the rotating motor 300 is connected with a rotating wheel 301, a vertical rod 302 is disposed at an outer edge position of the rotating wheel 301, a cross rod 303 is disposed at one end of the rotating motor 300, a rectangular groove 304 is opened at a position of the shaft of the cross rod 303 corresponding to the vertical rod 302, and a swivel ring adapted to the rectangular groove 304 is embedded in the rectangular groove, and the swivel ring is sleeved on the vertical rod 302. Therefore, the arrangement of the rotating piece 30 is helpful to drive the cleaning fan 31 to rotate in a sector area; specifically, after the rotating motor 300 starts to work, the rotating wheel 301 can be driven to rotate, so that the rotating wheel 301 rotates to drive the vertical rod 302 at the outer edge of the rotating wheel to circumferentially rotate, and after the vertical rod 302 rotates, the rotating ring can be driven to move in the rectangular groove 304, so that the cross rod 303 can be driven to reciprocally rotate in a fan shape through the movement of the rotating ring, and further the cleaning fan 31 is driven to rotate on the cleaning table through the rotating bearing 305, and a specific moving track is shown in fig. 8.
Therefore, based on the above embodiment, the dust and impurities attached to the upper surface of the glass ceramic panel 5 can be cleaned, so that the dust and impurities are prevented from being attached to the surface of the glass ceramic panel 5, and the image definition acquired by the image acquisition device 100 is affected, so that the defect misidentification of the defect detection system is effectively avoided, the detection precision of the defect detection system is greatly ensured, and the use effect of the defect detection system is greatly improved.
Meanwhile, it should be noted that the image capturing device 100 in this embodiment is preferably a CCD camera, and the supplemental light source 101 is a laser light, which is obtained through market purchase.
The foregoing detailed description of the invention has been presented for purposes of illustration and description, and it should be understood that the invention is not limited to the particular embodiments disclosed, but is intended to cover all modifications, equivalents, alternatives, and improvements within the spirit and principles of the invention.
Claims (4)
1. The surface defect detection system based on the glass ceramic panel finished product comprises an annular outer frame (1) and a detection unit, and is characterized in that the detection unit comprises an annular detection frame (10) coaxially arranged in the outer frame (1), the upper part of the annular detection frame (10) is in sliding connection with the outer frame (1) through a sliding mechanism (4), the annular detection frame (10) is divided into a first semi-ring and a second semi-ring from top to bottom by taking the center position of the annular detection frame as a reference, a plurality of image acquisition devices (100) are arranged on the inner ring surface of the first semi-ring at equal intervals, a plurality of supplementary light sources (101) are arranged on the inner ring surface of the second semi-ring at equal intervals, the image acquisition devices (100) and the supplementary light sources (101) are in one-to-one correspondence, the image acquisition devices (100) acquire surface images of the glass ceramic panel (5), and the supplementary light sources (101) are used for carrying out light source irradiation on the glass ceramic panel (5) when the image acquisition devices (100) acquire surface images of the glass ceramic panel (5), so that the image acquisition of the glass ceramic panel (5) is more clear, and the image acquisition is facilitated;
The detection unit further comprises an image recognition processing module which is arranged on the outer frame (1) and is in signal connection with the plurality of image acquisition devices (100), wherein the image recognition processing module is used for receiving surface image information of the glass ceramic panel (5) transmitted by the image acquisition devices (100) and detecting and recognizing defects of the glass ceramic panel (5) based on the surface image information;
the image recognition processing module comprises a preprocessing module, an image segmentation module and a defect recognition module which are sequentially connected through signals;
The preprocessing module is used for carrying out image enhancement processing after receiving surface original image information;
The image segmentation module is used for carrying out thresholding segmentation on the surface image information subjected to image enhancement processing, obtaining a gray threshold value of the surface image information through image gray level statistics calculation, and carrying out binarization segmentation on the image based on the gray threshold value to obtain binarized image information, wherein the gray threshold value is calculated based on an Otsu algorithm when the gray threshold value is calculated;
The defect identification module is used for extracting defect characteristic information in the binarized image information through a neural network so as to carry out defect judgment and identification, and outputting the defect characteristic information to an external main control computer;
The image enhancement processing specifically comprises filtering noise in image information through a median filter, carrying out texture expression and separation on the image information subjected to denoising processing based on a Gabor filter to obtain texture feature histogram information, carrying out self-adaptive histogram equalization on an original image, and fusing the image subjected to the self-adaptive histogram equalization with a histogram of the texture feature obtained before to obtain image information subjected to contrast enhancement;
The neural network is a Faster R-CNN neural network structure and comprises a feature extraction network, a region suggestion network, a region feature alignment, a defect classification and regression network and an output layer;
The bottom of outer frame (1) still is equipped with base (2) the upper portion of base (2) still is equipped with transportation frame (20), transportation frame (20) are rectangular frame structure transportation frame (20) inside is equipped with multiunit transportation roller (21) that are used for carrying microcrystalline glass panel (5) along its length direction, multiunit transportation roller (21) both ends all are connected through pivot and transportation frame (20) rotation, and all have the clearance between multiunit transportation roller (21), outer frame (1) are located one of them clearance, and the centre of a circle position department of outer frame (1) is in unified horizontal plane with transportation roller (21), the bottom of outer frame (1) is connected with base (2), just driving mechanism is installed to inside one side of transportation frame (20), driving mechanism is including being used for driving multiunit transportation roller (21) pivoted power unit (220) to be connected with power unit (220) transmission, and be used for driving annular detection frame (10) and carry out pivoted transmission unit (221) in outer frame (1).
2. The glass-ceramic panel finished product-based surface defect detection system of claim 1, wherein the power unit (220) comprises: the automatic conveying device comprises a driving motor (2200), a driving wheel (2202) and a driven wheel (2203), wherein one side of the driving motor (2200) is connected with a conveying frame (20), an output shaft (2201) is arranged at the output end of the driving wheel (2202), the driving wheel (2202) is sleeved on one end of the output shaft (2201), the driven wheel (2203) is provided with a plurality of rotating shafts which are sleeved on one ends of a plurality of groups of conveying rollers (21) one by one, and the driven wheel (2203) and the driving wheel (2202) are in transmission connection through a transmission belt (2204).
3. The glass-ceramic panel finished product-based surface defect detection system of claim 2, wherein the transmission unit (221) comprises a driven gear and rack (2211); the driven gear is keyed on an output shaft (2201) at the output end of the driving motor (2200), the rack (2211) is annular and is positioned on the outer surface of the annular detection frame (10), and the rack (2211) is in meshed connection with the transmission gear (2210).
4. A surface defect detection system based on glass ceramic panel finished products according to claim 3, characterized in that a cleaning table is further arranged on the transportation frame (20), a cleaning fan (31) is arranged on the cleaning table, the cleaning fan (31) is rotationally connected with the cleaning table through a rotating piece (30) at the bottom of the cleaning fan, and the cleaning fan (31) rotates on the cleaning table through the rotating piece (30) and cleans the surface of the glass ceramic panel (5) on the conveying roller (21) in a fan shape.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311509599.7A CN117517348B (en) | 2023-11-14 | 2023-11-14 | Surface defect detection system based on microcrystalline glass panel finished product |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311509599.7A CN117517348B (en) | 2023-11-14 | 2023-11-14 | Surface defect detection system based on microcrystalline glass panel finished product |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN117517348A CN117517348A (en) | 2024-02-06 |
| CN117517348B true CN117517348B (en) | 2024-05-14 |
Family
ID=89760223
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202311509599.7A Active CN117517348B (en) | 2023-11-14 | 2023-11-14 | Surface defect detection system based on microcrystalline glass panel finished product |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117517348B (en) |
Citations (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| SU1096567A1 (en) * | 1982-10-10 | 1984-06-07 | Научно-производственное объединение "Атомкотломаш" | Device for ultrasonic inspection of pipe welded seams |
| CN1789990A (en) * | 2005-12-01 | 2006-06-21 | 渤海船舶重工有限责任公司 | Automatic online detection method for defects on upper and lower surfaces during steel plate pretreatment process |
| CN106248686A (en) * | 2016-07-01 | 2016-12-21 | 广东技术师范学院 | Glass surface defects based on machine vision detection device and method |
| CN108698774A (en) * | 2016-02-25 | 2018-10-23 | 康宁股份有限公司 | The method and apparatus that edge surface inspection is carried out to movable glass coiled material |
| CN109668909A (en) * | 2017-10-13 | 2019-04-23 | 南京敏光视觉智能科技有限公司 | A kind of glass defect detection method |
| CN110940681A (en) * | 2019-12-11 | 2020-03-31 | 湘潭市汇丰设备制造有限公司 | Automatic nondestructive testing device for steam pressure pipeline with heat insulation layer |
| CN111306415A (en) * | 2020-03-31 | 2020-06-19 | 宁夏巨能机器人股份有限公司 | Camera support for detecting defects of buffer shell |
| CN113504238A (en) * | 2021-06-04 | 2021-10-15 | 广东华中科技大学工业技术研究院 | Glass surface defect collecting device and detection method |
| CN114092389A (en) * | 2021-09-13 | 2022-02-25 | 浙江大学 | Glass panel surface defect detection method based on small sample learning |
| CN115060748A (en) * | 2022-05-26 | 2022-09-16 | 大连理工大学宁波研究院 | Surround centre gripping formula pipeline CT scanning imaging device |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| DE102004004301B4 (en) * | 2004-01-28 | 2013-01-31 | Siemens Aktiengesellschaft | Imaging tomography device with compensating rings to compensate for an imbalance |
-
2023
- 2023-11-14 CN CN202311509599.7A patent/CN117517348B/en active Active
Patent Citations (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| SU1096567A1 (en) * | 1982-10-10 | 1984-06-07 | Научно-производственное объединение "Атомкотломаш" | Device for ultrasonic inspection of pipe welded seams |
| CN1789990A (en) * | 2005-12-01 | 2006-06-21 | 渤海船舶重工有限责任公司 | Automatic online detection method for defects on upper and lower surfaces during steel plate pretreatment process |
| CN108698774A (en) * | 2016-02-25 | 2018-10-23 | 康宁股份有限公司 | The method and apparatus that edge surface inspection is carried out to movable glass coiled material |
| CN106248686A (en) * | 2016-07-01 | 2016-12-21 | 广东技术师范学院 | Glass surface defects based on machine vision detection device and method |
| CN109668909A (en) * | 2017-10-13 | 2019-04-23 | 南京敏光视觉智能科技有限公司 | A kind of glass defect detection method |
| CN110940681A (en) * | 2019-12-11 | 2020-03-31 | 湘潭市汇丰设备制造有限公司 | Automatic nondestructive testing device for steam pressure pipeline with heat insulation layer |
| CN111306415A (en) * | 2020-03-31 | 2020-06-19 | 宁夏巨能机器人股份有限公司 | Camera support for detecting defects of buffer shell |
| CN113504238A (en) * | 2021-06-04 | 2021-10-15 | 广东华中科技大学工业技术研究院 | Glass surface defect collecting device and detection method |
| CN114092389A (en) * | 2021-09-13 | 2022-02-25 | 浙江大学 | Glass panel surface defect detection method based on small sample learning |
| CN115060748A (en) * | 2022-05-26 | 2022-09-16 | 大连理工大学宁波研究院 | Surround centre gripping formula pipeline CT scanning imaging device |
Also Published As
| Publication number | Publication date |
|---|---|
| CN117517348A (en) | 2024-02-06 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN103090804B (en) | Automatic detection system and detection method of finished product magnet ring image | |
| CN105548212B (en) | A kind of touched panel glass surface blemish on-line automaticization detection system and method | |
| CN104502356A (en) | Automatic detection method for defects of inner surface of sliding bearing on basis of computer vision | |
| CN109598715B (en) | Material granularity online detection method based on machine vision | |
| CN107486415A (en) | Thin bamboo strip defect on-line detecting system and detection method based on machine vision | |
| CN105388162A (en) | Raw material silicon wafer surface scratch detection method based on machine vision | |
| CN113578764B (en) | Non-woven fabric production detection device based on Internet of things | |
| CN110394307B (en) | Tobacco leaf automated inspection letter sorting system | |
| CN111563519B (en) | Tea impurity identification method and sorting equipment based on Stacking weighting integrated learning | |
| CN112950602B (en) | Candy defect detection device and classification method based on depth camera and machine learning | |
| CN102941189A (en) | Controllable rotary angle insection tray device of fruit sorter | |
| CN113894055A (en) | Hardware surface defect detection and classification system and method based on machine vision | |
| CN103234974A (en) | Lithium battery non-destructive detection method | |
| CN113504238A (en) | Glass surface defect collecting device and detection method | |
| CN111672773A (en) | Product surface defect detection system and method based on machine vision | |
| CN117517348B (en) | Surface defect detection system based on microcrystalline glass panel finished product | |
| CN207222383U (en) | Plank sorting system | |
| CN114359155A (en) | Film laminating method and system | |
| CN203069148U (en) | Finished magnetic ring image automatic detection system | |
| CN105973903B (en) | A kind of Oral liquid bottle lid detection method | |
| CN210358133U (en) | Automatic inspection equipment for realizing product appearance inspection | |
| CN112461854A (en) | Method and device for intelligently optimizing AOI detection of final inspection of cover plate | |
| CN114677597B (en) | Gear defect visual inspection method and system based on improved YOLOv5 network | |
| CN112474415B (en) | Fruit sorting machine | |
| CN111161215B (en) | Detection apparatus capable of identifying color of printer |
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 |