CN112452801A - High-efficiency online detection device for defects of glass bottles and working method thereof - Google Patents
High-efficiency online detection device for defects of glass bottles and working method thereof Download PDFInfo
- Publication number
- CN112452801A CN112452801A CN202011054840.8A CN202011054840A CN112452801A CN 112452801 A CN112452801 A CN 112452801A CN 202011054840 A CN202011054840 A CN 202011054840A CN 112452801 A CN112452801 A CN 112452801A
- Authority
- CN
- China
- Prior art keywords
- image
- conveying device
- detection
- ccd camera
- glass bottles
- 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
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B07—SEPARATING SOLIDS FROM SOLIDS; SORTING
- B07C—POSTAL SORTING; SORTING INDIVIDUAL ARTICLES, OR BULK MATERIAL FIT TO BE SORTED PIECE-MEAL, e.g. BY PICKING
- B07C5/00—Sorting according to a characteristic or feature of the articles or material being sorted, e.g. by control effected by devices which detect or measure such characteristic or feature; Sorting by manually actuated devices, e.g. switches
- B07C5/34—Sorting according to other particular properties
- B07C5/3404—Sorting according to other particular properties according to properties of containers or receptacles, e.g. rigidity, leaks, fill-level
- B07C5/3408—Sorting according to other particular properties according to properties of containers or receptacles, e.g. rigidity, leaks, fill-level for bottles, jars or other glassware
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B07—SEPARATING SOLIDS FROM SOLIDS; SORTING
- B07C—POSTAL SORTING; SORTING INDIVIDUAL ARTICLES, OR BULK MATERIAL FIT TO BE SORTED PIECE-MEAL, e.g. BY PICKING
- B07C5/00—Sorting according to a characteristic or feature of the articles or material being sorted, e.g. by control effected by devices which detect or measure such characteristic or feature; Sorting by manually actuated devices, e.g. switches
- B07C5/36—Sorting apparatus characterised by the means used for distribution
- B07C5/361—Processing or control devices therefor, e.g. escort memory
- B07C5/362—Separating or distributor mechanisms
-
- 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
- G01N21/8806—Specially adapted optical and illumination features
-
- 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
- G01N21/90—Investigating the presence of flaws or contamination in a container or its contents
-
- 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
- G01N21/90—Investigating the presence of flaws or contamination in a container or its contents
- G01N21/9009—Non-optical constructional details affecting optical inspection, e.g. cleaning mechanisms for optical parts, vibration reduction
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B07—SEPARATING SOLIDS FROM SOLIDS; SORTING
- B07C—POSTAL SORTING; SORTING INDIVIDUAL ARTICLES, OR BULK MATERIAL FIT TO BE SORTED PIECE-MEAL, e.g. BY PICKING
- B07C2501/00—Sorting according to a characteristic or feature of the articles or material to be sorted
- B07C2501/0063—Using robots
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B07—SEPARATING SOLIDS FROM SOLIDS; SORTING
- B07C—POSTAL SORTING; SORTING INDIVIDUAL ARTICLES, OR BULK MATERIAL FIT TO BE SORTED PIECE-MEAL, e.g. BY PICKING
- B07C2501/00—Sorting according to a characteristic or feature of the articles or material to be sorted
- B07C2501/0072—Sorting of 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
- G01N21/84—Systems specially adapted for particular applications
- G01N2021/845—Objects on a conveyor
-
- 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
- G01N21/8806—Specially adapted optical and illumination features
- G01N2021/8812—Diffuse illumination, e.g. "sky"
- G01N2021/8816—Diffuse illumination, e.g. "sky" by using multiple sources, e.g. LEDs
-
- 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
Landscapes
- Physics & Mathematics (AREA)
- Health & Medical Sciences (AREA)
- Life Sciences & Earth Sciences (AREA)
- Chemical & Material Sciences (AREA)
- Analytical Chemistry (AREA)
- Biochemistry (AREA)
- General Health & Medical Sciences (AREA)
- General Physics & Mathematics (AREA)
- Immunology (AREA)
- Pathology (AREA)
- Engineering & Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Signal Processing (AREA)
- Investigating Materials By The Use Of Optical Means Adapted For Particular Applications (AREA)
Abstract
A high-efficiency online detection device for defects of glass bottles and a working method thereof comprise a conveying mechanism, a detection mechanism, a grabbing mechanism and a waste pool; the conveying mechanism comprises a PLC (programmable logic controller), a servo driver, a servo motor and a conveying device, the PLC, the servo driver and the servo motor are sequentially connected, the servo motor provides power for the operation of the conveying device, and the glass bottles are placed on the conveying device; the detection mechanism comprises a light source, a CCD camera, a lens, a photoelectric sensor and an image acquisition card, wherein the light source comprises a bottleneck annular LED positive light source and a bottleneck LED backlight source, and the bottleneck annular LED positive light source 211 is arranged below the CCD camera 22. The high-efficiency online detection device for the defects of the glass bottle and the working method thereof adopt non-contact visual detection to carry out all-around defect detection on the bottle mouth, the bottle body and the bottle bottom of the glass bottle, and have the advantages of simple working method, high precision, high efficiency and high intelligent degree.
Description
Technical Field
The invention belongs to the technical field of defect detection, and particularly relates to a high-efficiency online glass bottle defect detection device and a working method thereof.
Background
The defects of the glass bottle as a packaging container relate to the quality of the packaging. When the glass bottle is manufactured, the common defects of bottle mouth fracture, notch, glass can body surface scratch, unevenness and the like are inevitably generated in the complex production process of raw material screening, processing melting and annealing forming. After the glass bottles with defects are filled with products, pressure difference is generated after liquid or gas is filled, so that pressure exists inside and outside the bottles, the bottles are exploded after being violently impacted, great potential safety hazards exist, and even the smooth operation of the whole production process is influenced. Therefore, before filling production, the detection of the canned glass bottles and the removal of the glass bottles with defects are very important.
In order to guarantee the quality of glass bottle products, the quality of glass bottles is generally detected by quality inspectors in China at present. The manual spot inspection has the defects of missing inspection, false inspection, low efficiency and the like. Along with the increasingly standard and strict control of the state on the glass bottle standard, the cost of enterprises is continuously increased due to the product quality problem, the labor cost is increased year by year, and the enterprises urgently need to realize transformation and upgrading from manual spot inspection to automatic and intelligent detection. Therefore, it is necessary to develop an efficient online detection device for defects of glass bottles and a working method thereof to solve the above technical problems.
Chinese patent application No. CN201110077446.0 discloses a glass bottle defect detecting platform, the detecting mechanism includes a bottle mouth detecting device, a bottle bottom detecting device, a bottle shoulder detecting device and a bottle body detecting device, the detecting mechanism is connected with a control host, and no improvement is made and improved in efficiency, accuracy and response speed of the detecting mechanism.
Disclosure of Invention
The purpose of the invention is as follows: in order to overcome the defects, the invention aims to provide an efficient online glass bottle defect detection device and a working method thereof, which have the advantages of reasonable structural design, high automation, high precision, high efficiency and high intelligent degree, adopt non-contact visual detection to carry out comprehensive defect detection on a bottle opening, a bottle body and a bottle bottom of a glass bottle, avoid the defects of missed detection, false detection, low efficiency and the like, and have simple working method and wide application prospect.
The purpose of the invention is realized by the following technical scheme:
a high-efficiency online detection device for defects of glass bottles is characterized by comprising a conveying mechanism, a detection mechanism, a grabbing mechanism and a waste pool; the conveying mechanism comprises a PLC (programmable logic controller), a servo driver, a servo motor and a conveying device, the PLC, the servo driver and the servo motor are sequentially connected, the servo motor provides power for the operation of the conveying device, and the glass bottles are placed on the conveying device; the detection mechanism comprises a light source, a CCD camera, a lens, a photoelectric sensor and an image acquisition card, wherein the light source comprises a bottleneck annular LED positive light source and a bottleneck annular LED backlight source, the bottleneck annular LED positive light source is arranged below the CCD camera, the bottleneck annular LED backlight source is arranged above the CCD camera, the lens is arranged at the bottom of the CCD camera, the photoelectric sensor is arranged at the two sides of the front end and the tail end of the conveying device, and the CCD camera is connected with the image acquisition card; the grabbing mechanism comprises a mechanical arm, a mechanical arm and a mechanical arm control cabinet, the mechanical arm is positioned on one side of the tail end of the conveying device, the mechanical arm is installed on the mechanical arm and faces the conveying device, and the mechanical arm control cabinet is connected with the mechanical arm; the waste tank is positioned on the right side of the grabbing mechanism; the PLC controller adopts RJ45 direct connection line industrial computer, the photoelectric sensor serial ports connects the industrial computer, image acquisition card adopts IEEE1394 to connect the industrial computer, the manipulator switch board passes through the communication and the industrial computer of ethernet and is connected.
The high-efficiency online glass bottle defect detection device is reasonable in structural design, adopts non-contact visual detection to perform all-around defect detection on the bottle opening, the bottle body and the bottle bottom of a glass bottle, is high in automation, precision and intelligence, and avoids the defects of missed detection, false detection, low efficiency and the like.
The glass bottle passes through conveyer and carries out motion from left to right, and when the position of glass bottle through detection mechanism, do not detect out bottleneck, body, bottle end and have the defect, then can enter into next process voluntarily, if detect and have any kind of problem, the industrial computer will control manipulator control mechanism, and the implementation of drive manipulator is snatched the rejection process, puts into the waste product pond with this glass bottle.
Wherein, transport mechanism has played the effect of series connection detection mechanism, snatching mechanism, waste product pond, has realized the physical connection between detection mechanism, snatch mechanism, the waste product pond. The PLC controller of the conveying mechanism controls the start and stop of the servo motor and the coordinated movement of the servo motor through the servo driver, the servo motor provides power, and the conveying device is used for conveying glass bottles.
Wherein, during the detection, the light that the natural light source shined is reflected on the glass bottle surface, and the angle also constantly changes moreover, and this can cause great influence to the detection precision, consequently, the natural light source can't reach the detection requirement. In order to improve the detection precision, a bottle mouth annular LED positive light source and a bottle body LED backlight source are arranged above the front end of the conveying device, the annular LED positive light source completely eliminates shadows through multiple reflections on the hemispherical inner wall, the illumination of a diffusion LED light source in the whole space area can be achieved, the good irradiation effect on uneven surface of an object to be detected is achieved, and the detection effect on the bottle mouth of the glass bottle is good; when the back surface is irradiated, the light-transmitting part and the light-proof part are separated, the white part is the light-transmitting part, the black part is the light-proof part, the detection effect on scratches and mixed foreign matters of a transparent object to be detected is good, and the detection effect on the surface of the glass bottle body is good. Furthermore, although the CCD camera displays the defect position in the detection image, photoelectric sensors are also arranged on the two sides of the front end and the tail end of the conveying device and used for detecting whether the bottom of the glass bottle is defective or not, and the precision of detecting the bottom defect of the glass bottle is further improved. The CCD camera is matched with the lens to be practical, and the shot and displayed picture has excellent resolution, so that the detection control mechanism can be favorable for processing the positions of good and bad glass bottles and the positions of defect positions and the like.
Further, the above high-efficiency online glass bottle defect detection device comprises a driving roller, a transmission belt, a driven roller, a bend roller, a carrier roller and a conveying device; the driving roller is arranged at the front end of the conveying device, and the servo motor is arranged below the driving roller and drives the driving roller through a gear; the driven roller is arranged at the tail end of the conveying device and is connected with the driving roller through a transmission belt; the turnabout drum is arranged below one side of the driving belt close to the driving drum, and the supporting roller is arranged below the other side of the driving belt; and a group of carrier rollers are arranged below the transmission belt.
The conveying device is reasonable in structure and simple in operation, can finish the transportation and the positioning of glass bottles, the PLC controls the start and stop of the servo motor and the coordinated movement of the servo motor through the servo driver in the process of conveying the glass bottles, the servo motor provides power, and the control of the running speed, the advancing or retreating direction, the starting or stopping is met through the matching of the driving roller, the transmission chain, the driven roller, the direction-changing roller, the supporting roller and the supporting roller. The carrier roller is used for supporting the transmission belt, reducing the running resistance of the transmission belt and ensuring that the suspension degree of the transmission belt does not exceed a certain limit so as to ensure that the glass bottle runs stably on the transmission belt.
Further, in the efficient online glass bottle defect detection device, the outer side of each carrier roller is wrapped with a rubber protective layer.
The protective layer that every bearing roller outside all wrapped up rubber not only can not harm the glass bottle like this, also can not harm the conveyer belt, has improved the life of conveyer belt.
Further, the working method of the high-efficiency online glass bottle defect detection device sequentially comprises the following steps:
(1) the conveying mechanism is reset, after the resetting is completed, the conveying mechanism is started, the light source is started, the industrial personal computer receives a trigger signal from the production line, the industrial personal computer sends a PLC (programmable logic controller) starting signal, the glass bottle reaches a conveying position, the PLC starts to control the servo motor to drive the conveying device to transmit through the servo driver, and the conveying device moves by a distance of one pitch every time;
(2) when the glass bottle reaches a set detection station of the detection mechanism, the PLC controls the transmission device to stop running and sends a CCD camera photographing signal to the industrial personal computer;
(3) the CCD camera takes a picture and transmits the acquired image to the industrial personal computer through the image acquisition card for image processing;
(4) the industrial personal computer sends the detection result to the PLC and the manipulator control cabinet, the conveying device runs, the qualified glass bottles are conveyed to the next procedure bottle stacking device through the conveying device, the unqualified manipulator is controlled through the manipulator control cabinet, the grabbing and rejecting procedures are implemented, and the glass bottles are placed into a waste product pool.
Further, in the working method of the above high-efficiency online glass bottle defect detection apparatus, the image processing in the step (3) includes the following steps:
(1) image preprocessing: the CCD camera takes a picture and transmits the acquired image to an industrial personal computer through an image acquisition card, the industrial personal computer firstly carries out filtering processing on the image, and the filtering processing comprises noise removal, histogram equalization, template selection and image transformation operation to obtain an image to be registered;
(2) feature extraction: based on the self characteristics of the reference image and the image to be registered, carrying out feature point extraction operation on the image to be registered according to the registration requirement and the purpose to be achieved;
(3) and (3) feature matching: after the feature extraction is finished, matching the corresponding features in the reference image and the image to be registered one by one, deleting the features which cannot be matched, and calculating the spatial corresponding relation of the relation between the reference image and the image to be registered;
(4) estimating model parameters: selecting image conversion which can best reflect the deformation property between the two images according to the spatial corresponding relation between the reference image and the image to be registered, solving model parameters between the reference image and the image to be registered, and solving a conversion model;
(5) image gray difference value: after the transformation model is solved, in the model parameter calculation of the image, a gray difference method is selected to remove floating points, so that the image processing is realized.
On an industrial production line, due to the shaking of a conveyor belt, unstable operation and interference of other external factors, the glass bottles inevitably shake in the conveying device, so that the adverse effect caused by shaking of the glass bottles in the transmission process is overcome through image processing, the detection result is more stable and accurate, and the anti-interference performance is enhanced.
Further, the working method of the high-efficiency online glass bottle defect detection device, wherein the defect processing and judging in the step (4) comprises the following steps:
(1) establishing a defect detection model: based on a YOLOv3 framework, a glass bottle appearance defect detection model based on a convolutional neural network is designed and constructed; the network structure of the glass bottle appearance defect detection model mainly comprises a convolution layer, a batch normalization processing layer and a jump layer connection module, wherein an activation function adopts a Leaky ReLu function;
(2) and (4) defect judgment: the glass bottle appearance defect detection model divides the processed image into SxS cells with the same size, each cell is allocated with 3 anchor point frames which are responsible for predicting 3 boundary frames at the center of the cell, and defect target task detection is carried out on 3 feature maps with different scales and sizes; the glass bottle appearance defect detection model adopts a Logistic regression method to predict each boundary frame, a Logistic function is used as a classifier when category prediction is carried out, then a square sum loss function is used for calculating a prediction frame positioning error, an IOU error and a classification error to be used as final loss, various appearance quality defects of the glass bottle are finally judged, different defects are accurately classified, and the defects are displayed and stored on an industrial personal computer.
Compared with the prior art, the invention has the following beneficial effects:
(1) the high-efficiency online detection device for the defects of the glass bottles, disclosed by the invention, has a reasonable structural design, adopts non-contact visual detection to carry out comprehensive defect detection on the bottle openings, the bottle bodies and the bottle bottoms of the glass bottles, has the advantages of high automation, high precision, high efficiency and high intelligent degree, avoids the defects of missed detection, false detection, low efficiency and the like, and has a wide application prospect;
(2) the working method of the high-efficiency online detection device for the defects of the glass bottles is simple, and the adverse effect caused by shaking of the glass bottles in the transmission process is overcome through image processing, so that the detection result is more stable and accurate, and the anti-interference performance is enhanced; through the improvement of defect processing judgment, the judgment precision is obviously improved, the quality and the qualification rate of glass bottle products are improved, and the labor cost is reduced.
Drawings
FIG. 1 is a layout diagram of a high-efficiency online glass bottle defect detection device according to the present invention;
in the figure: the automatic bottle opening and closing device comprises a conveying mechanism 1, a PLC 11, a servo driver 12, a servo motor 13, a gear 131, a conveying device 14, a driving roller 141, a conveying belt 142, a driven roller 143, a direction-changing roller 144, a carrier roller 145, a carrier roller 146, a detection mechanism 2, a light source 21, a bottle opening annular LED positive light source 211, a bottle body LED backlight source 212, a CCD camera 22, a lens 23, a photoelectric sensor 24, an image acquisition card 25, a grabbing mechanism 3, a mechanical arm 31, a mechanical arm 32, a mechanical arm control cabinet 33, a waste pool 4 and an industrial personal computer 5.
Detailed Description
The technical solutions in the embodiments of the present invention will be described clearly and completely with reference to specific experimental data and fig. 1, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, but 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.
As shown in fig. 1, the following embodiments provide a high-efficiency online detection device for defects of glass bottles, which includes a conveying mechanism 1, a detection mechanism 2, a grabbing mechanism 3, and a waste tank 4; the conveying mechanism 1 comprises a PLC (programmable logic controller) 11, a servo driver 12, a servo motor 13 and a conveying device 14, wherein the PLC 11, the servo driver 12 and the servo motor 13 are sequentially connected, the servo motor 13 provides power for the operation of the conveying device 14, and glass bottles are placed on the conveying device 14; the detection mechanism 2 comprises a light source 21, a CCD camera 22, a lens 23, a photoelectric sensor 24 and an image acquisition card 25, wherein the light source 21 comprises a bottleneck annular LED positive light source 211 and a body LED backlight source 212, the bottleneck annular LED positive light source 211 is arranged below the CCD camera 22, the body LED backlight source 212 is arranged above the CCD camera 22, the lens 23 is arranged at the bottom of the CCD camera 22, the photoelectric sensor 24 is arranged at the front end and the tail end of the conveying device 14, and the CCD camera 22 is connected with the image acquisition card; the grabbing mechanism 3 comprises a mechanical arm 31, a mechanical hand 32 and a mechanical hand control cabinet 33, wherein the mechanical arm 31 is positioned on one side of the tail end of the conveying device 14, the mechanical hand 32 is installed on the mechanical arm 31, the mechanical hand 32 faces the conveying device 14, and the mechanical hand control cabinet 33 is connected with the mechanical hand 32; the waste tank 4 is positioned at the right side of the grabbing mechanism 3; the PLC controller 11 adopts RJ45 direct connection industrial computer 5, the photoelectric sensor 24 serial port is connected with the industrial computer 5, the image acquisition card 25 adopts IEEE1394 to connect the industrial computer 5, and the manipulator control cabinet 33 is connected with the industrial computer 5 through the communication of Ethernet.
Further, the conveying device 14 comprises a driving roller 141, a driving belt 142, a driven roller 143, a direction-changing roller 144, a carrier roller 145 and a carrier roller 146; the driving roller 141 is arranged at the front end of the conveying device 14, the servo motor 13 is arranged below the driving roller 141, and the driving roller 141 is driven by the gear 131; the driven roller 143 is arranged at the end of the conveying device 14, and the driven roller 143 is connected with the driving roller 141 through a transmission belt 142; the direction-changing roller 144 is arranged below one side of the driving roller 141 of the transmission belt 142, and the idler 145 is arranged below the other side of the transmission belt 142; a set of rollers 146 is disposed below the belt 142.
Further, the outer sides of the carrier rollers 146 are all wrapped with rubber protection layers.
Further, the CCD camera 22 is a high-resolution industrial digital CCD camera, the lens 23 is a double telecentric machine vision lens, and the manipulator 32 is a six-axis manipulator.
Examples
The high-efficiency online detection device for the defects of the glass bottles sequentially comprises the following steps:
(1) the conveying mechanism 1 is reset, after the resetting is completed, the conveying mechanism 1 is started, the light source 21 is started, the industrial personal computer 5 receives a trigger signal from a production line, the industrial personal computer 5 sends a PLC (programmable logic controller) 11 starting signal, the glass bottle reaches a conveying position, the PLC 11 starts to control the servo motor 13 to drive the conveying device 14 to transmit through the servo driver 12, and the conveying device 14 moves one pitch distance every time;
(2) when the glass bottle reaches a set detection station of the detection mechanism 2, the PLC 11 controls the transmission device 14 to stop running and sends a photographing signal of the CCD camera 22 to the industrial personal computer 5;
(3) the CCD camera 22 takes a picture and transmits the acquired image to the industrial personal computer 5 through the image acquisition card 25 for image processing;
(4) through the defect processing and judgment of the industrial personal computer 5, the industrial personal computer 5 sends the detection result to the PLC controller 11 and the manipulator control cabinet 33, the conveying device 14 runs, the qualified glass bottles are conveyed to the next procedure bottle stacking device through the conveying device 14, the unqualified manipulator 32 is controlled through the manipulator control cabinet 33, the grabbing and rejecting procedures are implemented, and the glass bottles are placed into the waste product tank 4.
Wherein, the image processing of step (3) comprises the following steps:
(1) image preprocessing: the CCD camera 22 shoots and transmits the collected image to the industrial personal computer 5 through the image collecting card 25, the industrial personal computer 5 firstly carries out filtering processing on the image, and the filtering processing comprises noise removal, histogram equalization, template selection and image transformation operation to obtain an image to be registered;
(2) feature extraction: based on the self characteristics of the reference image and the image to be registered, carrying out feature point extraction operation on the image to be registered according to the registration requirement and the purpose to be achieved;
(3) and (3) feature matching: after the feature extraction is finished, matching the corresponding features in the reference image and the image to be registered one by one, deleting the features which cannot be matched, and calculating the spatial corresponding relation of the relation between the reference image and the image to be registered;
(4) estimating model parameters: selecting image conversion which can best reflect the deformation property between the two images according to the spatial corresponding relation between the reference image and the image to be registered, solving model parameters between the reference image and the image to be registered, and solving a conversion model;
(5) image gray difference value: after the transformation model is solved, in the model parameter calculation of the image, a gray difference method is selected to remove floating points, so that the image processing is realized.
Wherein, the defect processing and judging of the step (4) comprises the following steps:
(1) establishing a defect detection model: based on a YOLOv3 framework, a glass bottle appearance defect detection model based on a convolutional neural network is designed and constructed; the network structure of the glass bottle appearance defect detection model mainly comprises a convolution layer, a batch normalization processing layer and a jump layer connection module, wherein an activation function adopts a Leaky ReLu function;
(2) and (4) defect judgment: the glass bottle appearance defect detection model divides the processed image into SxS cells with the same size, each cell is allocated with 3 anchor point frames which are responsible for predicting 3 boundary frames at the center of the cell, and defect target task detection is carried out on 3 feature maps with different scales and sizes; the glass bottle appearance defect detection model adopts a Logistic regression method to predict each boundary frame, a Logistic function is used as a classifier when category prediction is carried out, then a square sum loss function is used for calculating a prediction frame positioning error, an IOU error and a classification error to be used as final loss, various appearance quality defects of the glass bottle are finally judged, different defects are accurately classified, and the defects are displayed and stored on the industrial personal computer 5.
Effect verification:
P/% | R/% | F1/% | Ap/% | velocity/(frame. s)-1) | |
Examples | 97.6 | 95.2 | 96.4 | 92.3 | 78 |
The high-efficiency online detection device for the defects of the glass bottles obtained by the embodiment adopts the working method of the high-efficiency online detection device for the defects of the glass bottles, and the accuracy rate P and the recall rate R, F are measured in 5000 tests1The values, average accuracy Ap, test time were validated and the results are shown in table 1.
Therefore, the high-efficiency online detection device for the defects of the glass bottles is far higher than manual detection in accuracy rate, recall rate, average accuracy and test time, achieves the purposes of high accuracy and high efficiency, has high intelligent degree, and avoids the defects of missed detection, false detection, low efficiency and the like.
The specific working method of the invention is many, and the above description is only the preferred embodiment of the invention. It should be noted that the above examples are only for illustrating the present invention, and are not intended to limit the scope of the present invention. It will be apparent to those skilled in the art that various modifications can be made without departing from the principles of the invention and these modifications are to be considered within the scope of the invention.
Claims (7)
1. The high-efficiency online detection device for the defects of the glass bottles is characterized by comprising a conveying mechanism (1), a detection mechanism (2), a grabbing mechanism (3) and a waste product pool (4); the conveying mechanism (1) comprises a PLC (programmable logic controller) (11), a servo driver (12), a servo motor (13) and a conveying device (14), the PLC (11), the servo driver (12) and the servo motor (13) are sequentially connected, the servo motor (13) provides power for the operation of the conveying device (14), and glass bottles are placed on the conveying device (14); the detection mechanism (2) comprises a light source (21), a CCD camera (22), a lens (23), a photoelectric sensor (24) and an image acquisition card (25), wherein the light source (21) comprises a bottleneck annular LED positive light source (211) and a body LED backlight source (212), the bottleneck annular LED positive light source (211) is installed below the CCD camera (22), the body LED backlight source (212) is installed above the CCD camera (22), the lens (23) is installed at the bottom of the CCD camera (22), the photoelectric sensor (24) is installed at the front end and the two sides of the tail end of the conveying device (14), and the CCD camera (22) is connected with the image acquisition card; the grabbing mechanism (3) comprises a mechanical arm (31), a mechanical hand (32) and a mechanical hand control cabinet (33), the mechanical arm (31) is located on one side of the tail end of the conveying device (14), the mechanical hand (32) is installed on the mechanical arm (31), the mechanical hand (32) faces the conveying device (14), and the mechanical hand control cabinet (33) is connected with the mechanical hand (32); the waste tank (4) is positioned on the right side of the grabbing mechanism (3); PLC controller (11) adopt RJ45 directly to link line industrial computer (5), photoelectric sensor (24) serial ports connection industrial computer (5), image acquisition card (25) adopt IEEE1394 to connect industrial computer (5), manipulator switch board (33) are connected with industrial computer (5) through the communication of ethernet.
2. The high-efficiency online defect detection device for glass bottles of claim 1, wherein the conveying device (14) comprises a driving roller (141), a transmission belt (142), a driven roller (143), a redirection roller (144), a carrier roller (145) and a carrier roller (146); the driving roller (141) is arranged at the front end of the conveying device (14), the servo motor (13) is arranged below the driving roller (141), and the driving roller (141) is driven by the gear (131); the driven roller (143) is arranged at the tail end of the conveying device (14), and the driven roller (143) is connected with the driving roller (141) through a transmission belt (142); the direction-changing roller (144) is arranged below one side of the transmission belt (142) close to the driving roller (141), and the idler (145) is arranged below the other side of the transmission belt (142); and a group of carrier rollers (146) are arranged below the transmission belt (142).
3. The high-efficiency online glass bottle defect detection device as claimed in claim 2, wherein the outer sides of the carrier rollers (146) are wrapped with rubber protection layers.
4. The high-efficiency online glass bottle defect detection device as claimed in claim 1, wherein the CCD camera (22) is a high-resolution industrial digital CCD camera, the lens (23) is a double telecentric machine vision lens, and the manipulator (32) is a six-axis manipulator.
5. The high-efficiency on-line detection device for the defects of the glass bottles as claimed in any one of claims 1 to 4, which is characterized by comprising the following steps in sequence:
(1) the conveying mechanism (1) is reset, after the resetting is completed, the conveying mechanism (1) is started, the light source (21) is started, the industrial personal computer (5) receives a trigger signal from a production line, the industrial personal computer (5) sends a PLC (programmable logic controller) starting signal, the glass bottles reach a conveying position, the PLC (11) starts to control the servo motor (13) to drive the conveying device (14) to transmit through the servo driver (12), and the conveying device (14) moves a distance of one pitch once in transmission;
(2) when the glass bottle reaches a set detection station of the detection mechanism (2), the PLC (11) controls the transmission device (14) to stop running and sends a photographing signal of the CCD camera (22) to the industrial personal computer (5);
(3) the CCD camera (22) takes a picture and transmits the collected image to the industrial personal computer (5) through the image collecting card (25) for image processing;
(4) through the defect processing and judgment of the industrial personal computer (5), the industrial personal computer (5) sends a detection result to the PLC (programmable logic controller) controller (11) and the manipulator control cabinet (33), the conveying device (14) operates, qualified glass bottles are conveyed to the next procedure bottle stacking device through the conveying device (14), unqualified glass bottles control the manipulator (32) through the manipulator control cabinet (33), the grabbing and rejecting procedure is implemented, and the glass bottles are placed into the waste pool (4).
6. The working method of the high-efficiency online defect detecting device for glass bottles of claim 5, wherein the image processing of the step (3) comprises the following steps:
(1) image preprocessing: the CCD camera (22) shoots and transmits the collected image to the industrial personal computer (5) through the image collecting card (25), the industrial personal computer (5) firstly carries out filtering processing on the image, and the filtering processing comprises noise removal, histogram equalization, template selection and image transformation operation to obtain an image to be registered;
(2) feature extraction: based on the self characteristics of the reference image and the image to be registered, carrying out feature point extraction operation on the image to be registered according to the registration requirement and the purpose to be achieved;
(3) and (3) feature matching: after the feature extraction is finished, matching the corresponding features in the reference image and the image to be registered one by one, deleting the features which cannot be matched, and calculating the spatial corresponding relation of the relation between the reference image and the image to be registered;
(4) estimating model parameters: selecting image conversion which can best reflect the deformation property between the two images according to the spatial corresponding relation between the reference image and the image to be registered, solving model parameters between the reference image and the image to be registered, and solving a conversion model;
(5) image gray difference value: after the transformation model is solved, in the model parameter calculation of the image, a gray difference method is selected to remove floating points, so that the image processing is realized.
7. The method for operating the high-efficiency online defect detecting device for glass bottles as claimed in claim 5, wherein the defect processing and judging in the step (4) comprises the following steps:
(1) establishing a defect detection model: based on a YOLOv3 framework, a glass bottle appearance defect detection model based on a convolutional neural network is designed and constructed; the network structure of the glass bottle appearance defect detection model mainly comprises a convolution layer, a batch normalization processing layer and a jump layer connection module, wherein an activation function adopts a Leaky ReLu function;
(2) and (4) defect judgment: the glass bottle appearance defect detection model divides the processed image into SxS cells with the same size, each cell is allocated with 3 anchor point frames which are responsible for predicting 3 boundary frames at the center of the cell, and defect target task detection is carried out on 3 feature maps with different scales and sizes; the glass bottle appearance defect detection model adopts a Logistic regression method to predict each boundary frame, a Logistic function is used as a classifier when category prediction is carried out, then a square sum loss function is used for calculating a prediction frame positioning error, an IOU error and a classification error to be used as final loss, various appearance quality defects of the glass bottle are finally judged, different defects are accurately classified, and the defects are displayed and stored on an industrial personal computer (5).
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202011054840.8A CN112452801A (en) | 2020-09-28 | 2020-09-28 | High-efficiency online detection device for defects of glass bottles and working method thereof |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202011054840.8A CN112452801A (en) | 2020-09-28 | 2020-09-28 | High-efficiency online detection device for defects of glass bottles and working method thereof |
Publications (1)
Publication Number | Publication Date |
---|---|
CN112452801A true CN112452801A (en) | 2021-03-09 |
Family
ID=74832983
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202011054840.8A Pending CN112452801A (en) | 2020-09-28 | 2020-09-28 | High-efficiency online detection device for defects of glass bottles and working method thereof |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN112452801A (en) |
Cited By (13)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN112964717A (en) * | 2021-03-24 | 2021-06-15 | 安徽中青检验认证服务有限公司 | Quality authentication device and method for checking and identifying |
CN113070240A (en) * | 2021-03-25 | 2021-07-06 | 南京工业大学 | Copper plate surface defect detection and automatic classification method based on machine vision and deep learning |
CN113428640A (en) * | 2021-07-05 | 2021-09-24 | 赵诣泓 | Intelligent bar counter system and operation method |
CN113504263A (en) * | 2021-06-24 | 2021-10-15 | 海南电网有限责任公司电力科学研究院 | Insulator abnormal heating defect detection device and application method thereof |
CN113538417A (en) * | 2021-08-24 | 2021-10-22 | 安徽顺鼎阿泰克科技有限公司 | Transparent container defect detection method and device based on multi-angle and target detection |
CN113533376A (en) * | 2021-06-29 | 2021-10-22 | 成都泓睿科技有限责任公司 | Glass bottle breakage defect detection system and detection method |
CN114018947A (en) * | 2021-10-21 | 2022-02-08 | 武汉理工大学 | High reflection of light plastic bottle lid surface flaw detection device |
CN115035119A (en) * | 2022-08-12 | 2022-09-09 | 山东省计算中心(国家超级计算济南中心) | Glass bottle bottom flaw image detection and removal device, system and method |
CN115078397A (en) * | 2022-07-20 | 2022-09-20 | 菲特(天津)检测技术有限公司 | Medicine bottle detection system and method and electronic equipment |
CN115445964A (en) * | 2022-11-10 | 2022-12-09 | 江苏尚纯自动化技术有限公司 | Intelligent monitoring system for filling production and conveying and detection method |
CN116493363A (en) * | 2023-05-04 | 2023-07-28 | 广东冠鸿智能装备有限公司 | Packaging bottle inspection cleaning system and equipment |
CN117383806A (en) * | 2023-10-08 | 2024-01-12 | 凯盛君恒药玻(重庆)有限公司 | Intelligent management method, control system and bottle making machine device for bottle making production |
CN118010761A (en) * | 2024-04-02 | 2024-05-10 | 七海测量技术(深圳)有限公司 | Curved glass defect detection device and method based on machine vision |
Citations (8)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
DE19605133A1 (en) * | 1996-02-13 | 1997-08-14 | Kronseder Maschf Krones | Inspection machine for vessels |
CN102539443A (en) * | 2011-12-29 | 2012-07-04 | 天津普达软件技术有限公司 | Bottle body defect automatic detection device based on machine vision and method thereof |
CN203091296U (en) * | 2013-02-01 | 2013-07-31 | 泸州市慧江机械制造有限公司 | System for detecting ceramic bottle or glass bottle |
CN204544803U (en) * | 2014-12-23 | 2015-08-12 | 山东明佳科技有限公司 | A kind of vial empty bottle sorting system |
CN107179325A (en) * | 2017-06-27 | 2017-09-19 | 建湖国创机械制造有限公司 | A kind of Clear glass bottles and jars all-around intelligent detection means |
CN207502417U (en) * | 2017-12-14 | 2018-06-15 | 山东明佳科技有限公司 | For the image collecting device of on-line checking |
CN109225912A (en) * | 2018-09-29 | 2019-01-18 | 芜湖中义玻璃有限公司 | A kind of vial quality testing device for eliminating |
CN110431405A (en) * | 2017-02-06 | 2019-11-08 | 东洋玻璃株式会社 | The check device of vial |
-
2020
- 2020-09-28 CN CN202011054840.8A patent/CN112452801A/en active Pending
Patent Citations (8)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
DE19605133A1 (en) * | 1996-02-13 | 1997-08-14 | Kronseder Maschf Krones | Inspection machine for vessels |
CN102539443A (en) * | 2011-12-29 | 2012-07-04 | 天津普达软件技术有限公司 | Bottle body defect automatic detection device based on machine vision and method thereof |
CN203091296U (en) * | 2013-02-01 | 2013-07-31 | 泸州市慧江机械制造有限公司 | System for detecting ceramic bottle or glass bottle |
CN204544803U (en) * | 2014-12-23 | 2015-08-12 | 山东明佳科技有限公司 | A kind of vial empty bottle sorting system |
CN110431405A (en) * | 2017-02-06 | 2019-11-08 | 东洋玻璃株式会社 | The check device of vial |
CN107179325A (en) * | 2017-06-27 | 2017-09-19 | 建湖国创机械制造有限公司 | A kind of Clear glass bottles and jars all-around intelligent detection means |
CN207502417U (en) * | 2017-12-14 | 2018-06-15 | 山东明佳科技有限公司 | For the image collecting device of on-line checking |
CN109225912A (en) * | 2018-09-29 | 2019-01-18 | 芜湖中义玻璃有限公司 | A kind of vial quality testing device for eliminating |
Non-Patent Citations (5)
Title |
---|
刘国华: "《HALCON编程及工程应用》", 30 September 2019, 西安电子科技大学出版社 * |
希尔德(SCHILDT,H.): "《高级TURBO PROLOG》", 31 July 1997, 石油大学出版社 * |
李茂林: "《低压电器及电控配电设备选用手册》", 31 January 1988, 辽宁科学技术出版社 * |
王星: "基于图像配准的药用玻璃瓶印刷字的缺陷检测", 《中国优秀硕士学位论文全文数据库 信息科技辑》 * |
陈宏彩等: "基于YOLOv3的医药玻璃瓶缺陷检测方法", 《包装工程》 * |
Cited By (19)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN112964717A (en) * | 2021-03-24 | 2021-06-15 | 安徽中青检验认证服务有限公司 | Quality authentication device and method for checking and identifying |
CN113070240A (en) * | 2021-03-25 | 2021-07-06 | 南京工业大学 | Copper plate surface defect detection and automatic classification method based on machine vision and deep learning |
CN113504263B (en) * | 2021-06-24 | 2022-12-27 | 海南电网有限责任公司电力科学研究院 | Insulator abnormal heating defect detection device and application method thereof |
CN113504263A (en) * | 2021-06-24 | 2021-10-15 | 海南电网有限责任公司电力科学研究院 | Insulator abnormal heating defect detection device and application method thereof |
CN113533376A (en) * | 2021-06-29 | 2021-10-22 | 成都泓睿科技有限责任公司 | Glass bottle breakage defect detection system and detection method |
CN113533376B (en) * | 2021-06-29 | 2024-02-27 | 成都泓睿科技有限责任公司 | Glass bottle breakage defect detection system and detection method |
CN113428640A (en) * | 2021-07-05 | 2021-09-24 | 赵诣泓 | Intelligent bar counter system and operation method |
CN113538417A (en) * | 2021-08-24 | 2021-10-22 | 安徽顺鼎阿泰克科技有限公司 | Transparent container defect detection method and device based on multi-angle and target detection |
CN114018947A (en) * | 2021-10-21 | 2022-02-08 | 武汉理工大学 | High reflection of light plastic bottle lid surface flaw detection device |
CN114018947B (en) * | 2021-10-21 | 2023-11-07 | 武汉理工大学 | High reflection of light plastic bottle lid surface flaw detection device |
CN115078397A (en) * | 2022-07-20 | 2022-09-20 | 菲特(天津)检测技术有限公司 | Medicine bottle detection system and method and electronic equipment |
CN115078397B (en) * | 2022-07-20 | 2022-11-15 | 菲特(天津)检测技术有限公司 | Medicine bottle detection system and method and electronic equipment |
CN115035119A (en) * | 2022-08-12 | 2022-09-09 | 山东省计算中心(国家超级计算济南中心) | Glass bottle bottom flaw image detection and removal device, system and method |
CN115445964A (en) * | 2022-11-10 | 2022-12-09 | 江苏尚纯自动化技术有限公司 | Intelligent monitoring system for filling production and conveying and detection method |
CN116493363A (en) * | 2023-05-04 | 2023-07-28 | 广东冠鸿智能装备有限公司 | Packaging bottle inspection cleaning system and equipment |
CN116493363B (en) * | 2023-05-04 | 2023-10-20 | 广东冠鸿智能装备有限公司 | Packaging bottle inspection and cleaning method and equipment |
CN117383806A (en) * | 2023-10-08 | 2024-01-12 | 凯盛君恒药玻(重庆)有限公司 | Intelligent management method, control system and bottle making machine device for bottle making production |
CN118010761A (en) * | 2024-04-02 | 2024-05-10 | 七海测量技术(深圳)有限公司 | Curved glass defect detection device and method based on machine vision |
CN118010761B (en) * | 2024-04-02 | 2024-06-21 | 七海测量技术(深圳)有限公司 | Curved glass defect detection device and method based on machine vision |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN112452801A (en) | High-efficiency online detection device for defects of glass bottles and working method thereof | |
US6504606B2 (en) | Integrated soft bag inspection system | |
CN103143508B (en) | Machine-vision-based online grading and packaging production line for bananas | |
KR20220070313A (en) | AA Visual Inspection System and Method | |
CN102218406B (en) | Intelligent detection device of defects of mobile phone outer shell based on machine vision | |
CN101105460A (en) | Multifunctional bottle checker | |
CN104634790A (en) | Capsule detection method and high-speed fully-automatic detection device | |
CN111545483B (en) | Chopstick sorting detection device and detection algorithm thereof | |
CN102621156A (en) | Image-processing-based automatic micro part sorting system | |
CN101865861A (en) | Automatic detection system for hardware appearance | |
CN103308531A (en) | Device and method for automatically detecting defects of glass medicine bottle | |
CN106353336A (en) | Lens coating automatic detection system | |
CN104614375A (en) | Nondestructive testing system for multi-layer food wrappage | |
CN111136027A (en) | Salted duck egg quality sorting device and method based on convolutional neural network | |
CN105728339A (en) | Visual automatic detecting system for ceramic cartridge | |
CN204346935U (en) | The location of food multi-packaging thing nondestructive detection system and dust guard | |
CN212301356U (en) | Wheel hub welding seam visual detection device | |
CN113894055A (en) | Hardware surface defect detection and classification system and method based on machine vision | |
CN203140320U (en) | Real-time banana grading and packaging production line based on nondestructive testing technology | |
CN111558542A (en) | Ceramic tile surface quality online detection sorting system and method | |
CN114359155A (en) | Film laminating method and system | |
CN202522527U (en) | Automatic glass medicine bottle flaw detection device | |
CN107742355B (en) | Automatic device of conveying letter sorting coin | |
CN204448621U (en) | The location of food multi-packaging thing nondestructive detection system and device for eliminating | |
CN216728343U (en) | Multistation automobile chassis foundry goods defect automatic check out system |
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 | ||
RJ01 | Rejection of invention patent application after publication | ||
RJ01 | Rejection of invention patent application after publication |
Application publication date: 20210309 |