WO2023021209A1 - System and method for inspecting an object - Google Patents
System and method for inspecting an object Download PDFInfo
- Publication number
- WO2023021209A1 WO2023021209A1 PCT/EP2022/073256 EP2022073256W WO2023021209A1 WO 2023021209 A1 WO2023021209 A1 WO 2023021209A1 EP 2022073256 W EP2022073256 W EP 2022073256W WO 2023021209 A1 WO2023021209 A1 WO 2023021209A1
- Authority
- WO
- WIPO (PCT)
- Prior art keywords
- light pattern
- imaging device
- pattern
- input light
- defect
- Prior art date
Links
- 238000000034 method Methods 0.000 title claims description 27
- 238000003384 imaging method Methods 0.000 claims abstract description 50
- 230000007547 defect Effects 0.000 claims abstract description 47
- 238000012545 processing Methods 0.000 claims abstract description 22
- 239000011521 glass Substances 0.000 claims description 61
- 239000002245 particle Substances 0.000 claims description 35
- 238000001514 detection method Methods 0.000 claims description 28
- 239000003086 colorant Substances 0.000 claims description 14
- 239000012530 fluid Substances 0.000 claims description 14
- 239000000835 fiber Substances 0.000 claims description 12
- 230000002123 temporal effect Effects 0.000 claims description 12
- 238000010801 machine learning Methods 0.000 claims description 8
- 238000013145 classification model Methods 0.000 claims description 6
- 238000005299 abrasion Methods 0.000 claims description 3
- 238000013473 artificial intelligence Methods 0.000 claims description 3
- 239000000428 dust Substances 0.000 claims description 3
- 238000004458 analytical method Methods 0.000 claims description 2
- 230000008859 change Effects 0.000 claims description 2
- 230000007704 transition Effects 0.000 claims description 2
- 238000000926 separation method Methods 0.000 claims 1
- 238000012706 support-vector machine Methods 0.000 description 7
- 238000012360 testing method Methods 0.000 description 6
- 230000003287 optical effect Effects 0.000 description 5
- 238000004590 computer program Methods 0.000 description 4
- 238000002790 cross-validation Methods 0.000 description 4
- 239000002184 metal Substances 0.000 description 4
- 239000007788 liquid Substances 0.000 description 3
- 230000008569 process Effects 0.000 description 3
- 238000010200 validation analysis Methods 0.000 description 3
- 238000011109 contamination Methods 0.000 description 2
- 230000004069 differentiation Effects 0.000 description 2
- -1 dirt Substances 0.000 description 2
- 229940079593 drug Drugs 0.000 description 2
- 239000003814 drug Substances 0.000 description 2
- 238000007689 inspection Methods 0.000 description 2
- 230000001788 irregular Effects 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 230000004044 response Effects 0.000 description 2
- 238000012549 training Methods 0.000 description 2
- 230000002159 abnormal effect Effects 0.000 description 1
- 230000006399 behavior Effects 0.000 description 1
- 238000010276 construction Methods 0.000 description 1
- 238000013135 deep learning Methods 0.000 description 1
- 230000001678 irradiating effect Effects 0.000 description 1
- 230000010287 polarization Effects 0.000 description 1
- 230000009467 reduction Effects 0.000 description 1
- 230000003595 spectral effect Effects 0.000 description 1
- 239000012906 subvisible particle Substances 0.000 description 1
- 229960005486 vaccine Drugs 0.000 description 1
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N21/00—Investigating or analysing materials by the use of optical means, i.e. using sub-millimetre waves, infrared, visible or ultraviolet light
- G01N21/84—Systems specially adapted for particular applications
- G01N21/88—Investigating the presence of flaws or contamination
- G01N21/90—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
- G01N21/9018—Dirt detection in containers
- G01N21/9027—Dirt detection in containers in containers after filling
-
- 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
- G01N21/9036—Investigating the presence of flaws or contamination in a container or its contents using arrays of emitters or receivers
-
- 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
- G01N21/9054—Inspection of sealing surface and container finish
-
- 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
- G01N21/8806—Specially adapted optical and illumination features
- G01N2021/8845—Multiple wavelengths of illumination or detection
Definitions
- the present disclosure relates to a system and method for inspecting an object using a light pattern, and more particularly to identifying a defect such as an immobile glass particle in glass vials and syringes using a light pattern.
- Glass vials and syringes are used to store and administer medicines, drugs, vaccines and the like to patients. Glass vials and syringes may break due to the presence of surface flaws and application of stress during tube forming, tube-to-container converting or pharmaceutical filling operations creating immobile glass particles and/or chips.
- the immobile glass particles are small subvisible particles to large visible chips.
- the detection of immobile glass particles in liquid filled syringes or vials is a difficult problem to solve due to false negatives caused by dirt, air bubbles and fibers on or in the glass.
- EP3312592A1 discloses a device that detects air bubbles and glass particles in a container filled with liquid by irradiating the container with light.
- the device employs at least two light sources which differ from one another in their spectral light color and/or their polarization.
- the use of multiple light sources that differ from one another in their optical properties is quite inefficient, cumbersome, expensive, and leads to generation of a large number of false negatives in detection of defects.
- the present invention relates to a system and method for inspecting an object, as set out in the appended claims.
- a system for inspecting an object includes a light source for generating an input light pattern, the object for receiving the input light pattern, an imaging device for capturing an image of the object that includes an output light pattern, and a processing module for analysing one or more output light patterns in one or more images of the object to identify a defect in the object.
- the light source is disposed at a left side of the object for backlighting the object with the input light pattern
- the imaging device is disposed at a right side of the object, such that the light source and the imaging device face each other, with the object in between.
- the object when the object is one of: a glass vial and a syringe filled with a fluid, the object acts as a cylindrical convex lens that refract the input light pattern to generate the output light pattern.
- the input light pattern includes a plurality of colours disposed in parallel with respect to each other, and each colour separated from corresponding adjacent colour by a black line.
- the output light pattern is a clear inverted input light pattern with respective black lines, when the imaging device captures the image at a first focal point that is outside the object and is at a first distance from the imaging device.
- the output light pattern is a blurred inverted input light pattern including a smooth pattern among adjacent colours without black lines, when the imaging device captures the image at a second focal point that is inside the object and is at a second distance from the imaging device.
- the first and second distances are determined based on refractive indices of the object, and the fluid inside the object.
- the processing module is configured to identify an immobile glass particle in the object as the defect, upon detecting a randomly dispersed irregularly shaped light in the smooth pattern, and wherein the randomly dispersed irregularly shaped light has a position which remains constant and a colour which changes across the one or more images obtained by rotating the object.
- the processing module is configured to identify an air bubble in the object as the defect, upon detecting a miniature light pattern similar to the input light pattern in the output light pattern, and wherein the air bubble acts as a concave lens that receives the input light pattern and generates the miniature input light pattern at a left side of the air bubble.
- the processing module includes an Artificial Intelligence (Al) based processing system that employs a machine learning model for identifying the defect in the object, wherein the machine learning model includes at least one of: a classification model, a temporal classification model, an object detection model, a temporal object detection model, and an anomaly detection model.
- Al Artificial Intelligence
- the light source projects the input light pattern on the object
- the imaging device captures one or more images of the projected light pattern on the object.
- the system further includes a robotic arm attached to the object for rotating the object and enabling the imaging device to capture the one or more images of the object in one or more different orientations.
- the object is configured to perform at least one of: reflect and refract the input light pattern.
- the object includes one of: a glass syringe, a glass vial, a contact lens, a plastic tube, and a metal surface.
- the defect includes at least one of: an immobile glass particle in the object, an air bubble, a dust particle, a fiber particle, a glass abrasion, and a deformation defect in the object.
- a method for inspecting an object includes generating an input light pattern by a light source, capturing an image of the object by an imaging device that includes an output light pattern, and analysing one or more output light patterns in one or more images of the object to identify a defect in the object.
- the method further includes generating the output light pattern as a clear inverted input light pattern with respective black lines, upon capturing the image at a first focal point that is outside the object and is at a first distance from the imaging device.
- the method further includes generating the output light pattern as a blurred inverted input light pattern including a smooth pattern among adjacent colours without black lines, upon capturing the image at a second focal point that is inside the object and is at a second distance from the imaging device.
- the method further includes identifying an immobile glass particle in the object as the defect, upon detecting a randomly dispersed irregularly shaped light in the smooth pattern, and wherein the randomly dispersed irregularly shaped light has a position which remains constant and a colour which changes across the one or more images obtained by rotating the object.
- the method further includes identifying an air bubble in the object as the defect, upon detecting a miniature light pattern similar to the input light pattern in the smooth pattern, and wherein the air bubble acts as a concave lens that receives the input light pattern and generates the miniature input light pattern at a left side of the air bubble.
- a non-transitory computer readable medium having stored thereon computer-executable instructions which, when executed by a processor, cause the processor to generate an input light pattern by a light source, capture an image of an object by an imaging device that includes an input light pattern, and analyse one or more output light patterns in one or more images of the object to identify a defect in the object.
- a computer program comprising program instructions for causing a computer program to carry out the above method which may be embodied on a record medium, carrier signal or read-only memory.
- FIG.1 illustrates a system for inspecting an object, in accordance with an embodiment of the present invention
- FIG.2 illustrates an input light pattern used for inspecting the object, and images including first and second output light patterns, in accordance with an embodiment of the present invention
- FIG.3 illustrates various type of disruptions in the output light pattern caused by various defects in the object, in accordance with an embodiment of the present invention
- FIGs.4A and 4B illustrate an image of light pattern taken by the imaging device at a focal point inside the object when the air bubble is present in the object, in accordance with an embodiment of the present invention
- FIG. 5A illustrate a processing module for analysing the images captured by the imaging device to identify a defect in the object, in accordance with an embodiment of the present invention
- FIG. 5B illustrates a histogram of the cross-validation tests performed for the processing module, in accordance with an embodiment of the present invention
- FIG.5C illustrates first and second example model outputs generated based on glass syringe detection and non-glass syringe detection by the processing module, respectively, in accordance with an embodiment of the present invention
- FIG. 6 illustrates another type of input light pattern that may be used for inspecting an object, in accordance with an embodiment of the present invention
- FIG.7 is a flowchart illustrating a method for inspecting an object, in accordance with an embodiment of the present invention.
- FIG.8 illustrates validation test results of inspection of various number of glass syringes/vials, in accordance with an embodiment of the present invention.
- FIG.1 illustrates a system 100 for inspecting an object 102, in accordance with an embodiment of the present invention.
- FIG.2 illustrates an input light pattern 202 used for inspecting the object 102, and images including first and second output light patterns 204 and 206.
- the object 102 is typically any object that is capable of reflecting and/or refracting the input light pattern 202.
- Examples of the object 102 include but are not limited to: a glass vial filled with fluid, a glass syringe filled with fluid, a contact lens, a plastic tube, a metal surface.
- the inspecting the object 102 may include identifying a defect such as an immobile glass particle, an air bubble, a dust particle, a fiber particle, a glass abrasion, and a deformation defect in the object
- the defect may include imperfections on the metal surface that reflect light.
- the defect may include an immobile glass particle, an air bubbles and dirt in the fluid and also deformation (non-smooth) defect in the glass.
- the input light pattern 202 includes a plurality of colours disposed in parallel with respect to each other, and each colour separated from corresponding adjacent colour by a black line.
- the input light pattern 202 is not limited to the one shown herein, it may include other custom light patterns based on the type of object and defect to be identified.
- the light source 104 when the object 102 includes a glass vial/syringe filled with fluid, the light source 104 generates a custom input light pattern based on optical properties of the glass and fluid to emphasize difference between glass particles, dirt, fibers and air bubbles.
- the system 100 includes a light source 104 for generating the input light pattern 202, the object 102 for receiving the input light pattern 202, an imaging device 106 for capturing images of the object 102 that includes the first and second output light patterns 204 and 206, and a processing module 108 for analysing the output light patterns 204 and 206 to identify a defect in the object 106.
- the system 100 further includes a robotic arm 110 attached to the object 102 for rotating the objectl 02 and enabling the imaging device 106 to capture the one or more images of the object 102 in one or more different orientations, i.e. from all required angles.
- the imaging device 106 includes, for example, a charge coupled (CCD) camera.
- the light source 104 is disposed at a left side of the object 102 for backlighting the object 102 with the input light pattern 202
- the imaging device 106 is disposed at a right side of the object 102, such that the light source 104 and the imaging device 106 face each other, with the object 102 is in between.
- the light source 104 and the imaging device 106 may be positioned in such a manner that the light source 104 projects the input light pattern 202 on the object 102, and the imaging device 106 captures one or more images of the projected light pattern to enable identification of defects by identifying disruptions in the projected light pattern.
- FIG.2 illustrates the object 102 as one of: a glass vial and a syringe filled with a fluid and images including first and second output light patterns 204 and 206.
- the object 102 refracts the output light patterns 204 and 206 when there is no defect in the object 102.
- the object 102 acts as a cylindrical convex lens that refract the input light pattern 202 and forms the output light patterns 204 and 206 at a right side thereof.
- the images including output light patterns 204 and 206 are captured by the imaging device 106 at two different focal points.
- the imaging device 106 generates an image including the first output light pattern 204 when it focusses on an outside wall of the object at a distance ‘D’ from the imaging device 106.
- the imaging device 106 captures an image of the object 102 at a first focal point that is outside a cylindrical wall of the object
- the output light pattern 204 is a clear inverted image of the input light pattern 202 with respective black lines.
- the imaging device 106 generates an image including the second output light pattern 206 when it focusses on an inside wall of the object 102 at a distance ‘F’ from the imaging device 106.
- the imaging device 106 captures an image of the object 102 at a second focal point that is inside of a cylindrical wall of the object 102.
- the output light pattern 206 is an inverted blurred image of the input light pattern 202.
- the output light pattern 206 includes a smooth pattern among adjacent colours, in which the transitions between colours are relatively smooth without black lines.
- the glass syringes/vials i.e. object 102 with no defects are imaged as a smooth surface with transitioning light colours. If there is any glass particle, dirt, fiber, or air bubble inside or at a surface of the glass vial/syringe (i.e. object 102), then the smooth pattern may be disrupted.
- the first and second distances ‘D’ and ‘F’ are determined based on the refractive index of the glass of the glass syringe/vial, and the fluid contained inside the glass syringe/vial.
- the text beside the second output light pattern 206 in fig. 2 notes: images of light pattern taken by cameras in different focal points.
- FIG.3 illustrates various type of disruptions in the output light pattern 300 caused by various defects in the object 102, i.e. glass syringes/vials filled with fluids.
- the output light pattern 300 is obtained by the imaging device 106 by focusing on an internal wall of the object 102.
- Figure 3 depicts air bubbles 304 and 306, glass particle 302 and dirt 308.
- the output light pattern 300 includes a randomly dispersed irregular shaped light at corresponding position of immobile glass particle 302.
- the position of the randomly dispersed irregularly shaped light in the output light pattern 300 remains constant but its colour changes across the images obtained by rotating the object 102.
- the glass particle 302 is irregular in shape and refract the light from the input light pattern in random ways. It is highlighted as it disperses the light randomly and its colour changes as it rotate.
- the unique light pattern response/signature for immobile glass particles enables the identification and differentiation with other contamination/particles.
- the object 102 i.e. the glass syringe/vial includes a defect such as dirt/fiber 308, then there is not much change in the output light pattern 300.
- the dirt and fibers undergo minimal changes as they are imaged while being rotated.
- the unique light pattern response/signature for dirt and fibers enables the identification and differentiation with other contamination/particles.
- the output light pattern 300 includes a miniature light pattern similar to the input light pattern at corresponding position of the air bubble in the object 102.
- Figure 4A depicts light pattern 202 on the left.
- On the right is a plan view of the syringe I vial.
- the text to the right of light pattern 400 in fig. 4A states: images of light pattern taken by camera focal point inside container.
- the text to the right of light pattern 400 in fig. 4B states: images of light pattern at left of bubble.
- the air bubble 304 near the edge of the syringe/vial 102 acts as a concave lens that generates a miniature light pattern 402 similar to the input light pattern 202 at a left side of the air bubble 304.
- the miniature light pattern 402 due to the air bubble 304 is in sharper focus than the light pattern 400 where there is no air bubble 304 and as the air bubble 304 acts as a concave lens, the miniature light pattern 402 is not inverted.
- the black lines in the miniature light pattern 402 are clearly visible in the air bubble 304 and as the miniature light pattern 402 is not inverted, the revealed bubble 304 is highlighted and easily identifiable.
- the air bubbles have a distinct pattern acting as optical lenses inverting the structured light pattern.
- the processing module 108 is an Artificial Intelligence (Al) based module that employs a machine learning model for processing the information captured by the imaging device 106 for identifying the defect in the object 102.
- Al Artificial Intelligence
- the box labelled 506 is entitled “Anomaly Detection”.
- the text under the graph reads: In this syringe abnormal compared to all other syringes that do not have immobile glass particles?
- the box labelled 504 is entitled “Object Detection”.
- the text under the image reads: Where is the defect located in this syringe??
- the box labelled 502 is entitled “Temporal Classification”.
- the text under the image reads: Find correlations between temporal images of the syringe?
- the box labelled 508 is entitled “Classifier Support Vector Machine”.
- the text under the image reads: Trained on the outputs from all the other modules to generate final predictions.
- the machine learning model includes at least one of: a classification model, a temporal classification model 502, an object detection model 504, a temporal object detection model, and an anomaly detection model 506.
- the Al based module 108 employs a single or ensemble of computer vision/machine learning/deep learning algorithms to accurately detect glass particles and exclude air bubbles, dirt and fibers.
- the ensemble modeling is a process where multiple diverse machine learning models may be combined to predict an outcome differentiating between air bubbles, dirt, fibers and immobile glass particles on internal surfaces of glass syringes/vials.
- the output of the anomaly detection module 506, the object detection module 504, and the temporal classification module 502 is provided to a Support Vector Machine (SVM) classifier 508.
- SVM Support Vector Machine
- FIG.5C illustrates first and second example model outputs 510 and 512 generated based on glass syringe detection and non-glass syringe detection by the processing module 108 respectively.
- Both the first and second example model outputs 510 and 512 illustrates model probability for anomaly detection model, object detection model, and temporal object detection model of the processing module 108.
- the plot beginning at around 1 .0 on the y-axis is anomaly detection
- the plot beginning just under 0.6 on the y-axis is object detection.
- the plot beginning at around 0.1 on the y-axis is temporal.
- the plot beginning at around 0.1 on the y-axis is anomaly detection
- the plot that rises just before frame 20 is temporal.
- the third plot is object detection.
- FIG. 6 illustrates another type of input light pattern 600 that may be used for inspecting an object such as the object 102.
- FIG.7 is a flowchart illustrating a method 700 for inspecting an object, in accordance with an embodiment of the present invention.
- the method includes generating an input light pattern by a light source.
- the method includes capturing an image of the object by an imaging device that includes an output light pattern.
- the method includes analysing one or more output light patterns in one or more images of the object to identify the defect in the object.
- FIG.8 illustrates validation test results 800 of inspection of various number of glass syringes/vials by implementing the method and system of the present invention. It may be noted that the overall defect detection rate is found to be
- the embodiments in the invention described with reference to the drawings comprise a computer apparatus and/or processes performed in a computer apparatus.
- the invention also extends to computer programs, particularly computer programs stored on or in a carrier adapted to bring the invention into practice.
- the program may be in the form of source code, object code, or a code intermediate source and object code, such as in partially compiled form or in any other form suitable for use in the implementation of the method according to the invention.
- the carrier may 1 3 comprise a storage medium such as ROM, e.g. a memory stick or hard disk.
- the carrier may be an electrical or optical signal which may be transmitted via an electrical or an optical cable or by radio or other means.
Abstract
The present invention relates to a system for inspecting an object. The system includes a light source for generating an input light pattern, the object for receiving the input light pattern, an imaging device for capturing an image of the object that includes an output light pattern, and a processing module for analysing one or more output light patterns in one or more images of the object to identify a defect in the object.
Description
TITLE
SYSTEM AND METHOD FOR INSPECTING AN OBJECT
FIELD
The present disclosure relates to a system and method for inspecting an object using a light pattern, and more particularly to identifying a defect such as an immobile glass particle in glass vials and syringes using a light pattern.
BACKGROUND
In pharmaceutical industry glass vials and syringes are used to store and administer medicines, drugs, vaccines and the like to patients. Glass vials and syringes may break due to the presence of surface flaws and application of stress during tube forming, tube-to-container converting or pharmaceutical filling operations creating immobile glass particles and/or chips.
The immobile glass particles are small subvisible particles to large visible chips. The detection of immobile glass particles in liquid filled syringes or vials is a difficult problem to solve due to false negatives caused by dirt, air bubbles and fibers on or in the glass.
EP3312592A1 discloses a device that detects air bubbles and glass particles in a container filled with liquid by irradiating the container with light. However, the device employs at least two light sources which differ from one another in their spectral light color and/or their polarization. The use of multiple light sources that differ from one another in their optical properties is quite inefficient, cumbersome, expensive, and leads to generation of a large number of false negatives in detection of defects.
In view of the above, there is a need for a system that is compact, can be set up easily to detect defects such as immobile glass particles in liquid filled syringes or vials, and should provide a significant reduction in false negatives in detection of immobile glass particles.
SUMMARY
The present invention relates to a system and method for inspecting an object, as set out in the appended claims.
In one aspect, there is provided a system for inspecting an object. The system includes a light source for generating an input light pattern, the object for receiving the input light pattern, an imaging device for capturing an image of the object that includes an output light pattern, and a processing module for analysing one or more output light patterns in one or more images of the object to identify a defect in the object.
In an embodiment of the present invention, the light source is disposed at a left side of the object for backlighting the object with the input light pattern, and the imaging device is disposed at a right side of the object, such that the light source and the imaging device face each other, with the object in between.
In an embodiment of the present invention, when the object is one of: a glass vial and a syringe filled with a fluid, the object acts as a cylindrical convex lens that refract the input light pattern to generate the output light pattern.
In an embodiment of the present invention, the input light pattern includes a plurality of colours disposed in parallel with respect to each other, and each colour separated from corresponding adjacent colour by a black line.
In an embodiment of the present invention, the output light pattern is a clear inverted input light pattern with respective black lines, when the imaging device captures the image at a first focal point that is outside the object and is at a first distance from the imaging device.
In an embodiment of the present invention, the output light pattern is a blurred inverted input light pattern including a smooth pattern among adjacent colours without black lines, when the imaging device captures the image at a second focal point that is inside the object and is at a second distance from the imaging device.
In an embodiment of the present invention, the first and second distances are determined based on refractive indices of the object, and the fluid inside the object.
In an embodiment of the present invention, the processing module is configured to identify an immobile glass particle in the object as the defect, upon detecting a randomly dispersed irregularly shaped light in the smooth pattern, and wherein the randomly dispersed irregularly shaped light has a position which remains constant and a colour which changes across the one or more images obtained by rotating the object.
In an embodiment of the present invention, the processing module is configured to identify an air bubble in the object as the defect, upon detecting a miniature light pattern similar to the input light pattern in the output light pattern, and wherein the air bubble acts as a concave lens that receives the input light pattern and generates the miniature input light pattern at a left side of the air bubble.
In an embodiment of the present invention, the processing module includes an Artificial Intelligence (Al) based processing system that employs a machine learning model for identifying the defect in the object, wherein the machine learning model includes at least one of: a classification model, a temporal classification model, an object detection model, a temporal object detection model, and an anomaly detection model.
In an embodiment of the present invention, the light source projects the input light pattern on the object, and the imaging device captures one or more images of the projected light pattern on the object.
In an embodiment of the present invention, the system further includes a robotic arm attached to the object for rotating the object and enabling the imaging device to capture the one or more images of the object in one or more different orientations.
In an embodiment of the present invention, the object is configured to perform at least one of: reflect and refract the input light pattern.
In an embodiment of the present invention, the object includes one of: a glass syringe, a glass vial, a contact lens, a plastic tube, and a metal surface.
In an embodiment of the present invention, the defect includes at least one of: an immobile glass particle in the object, an air bubble, a dust particle, a fiber particle, a glass abrasion, and a deformation defect in the object.
In another aspect, there is provided a method for inspecting an object. The method includes generating an input light pattern by a light source, capturing an image of the object by an imaging device that includes an output light pattern, and analysing one or more output light patterns in one or more images of the object to identify a defect in the object.
In an embodiment of the present invention, the method further includes generating the output light pattern as a clear inverted input light pattern with respective black lines, upon capturing the image at a first focal point that is outside the object and is at a first distance from the imaging device.
In an embodiment of the present invention, the method further includes generating the output light pattern as a blurred inverted input light pattern including a smooth pattern among adjacent colours without black lines, upon capturing the image at a second focal point that is inside the object and is at a second distance from the imaging device.
In an embodiment of the present invention, the method further includes identifying an immobile glass particle in the object as the defect, upon detecting a randomly dispersed irregularly shaped light in the smooth pattern, and wherein the randomly dispersed irregularly shaped light has a position which remains constant and a colour which changes across the one or more images obtained by rotating the object.
In an embodiment of the present invention, the method further includes identifying an air bubble in the object as the defect, upon detecting a miniature light pattern similar to the input light pattern in the smooth pattern, and wherein the air bubble acts as a concave lens that receives the input light pattern and generates the miniature input light pattern at a left side of the air bubble.
In a third aspect, there is provided a non-transitory computer readable medium having stored thereon computer-executable instructions which, when executed by a processor, cause the processor to generate an input light pattern by a light source, capture an image of an object by an imaging device that includes an input light pattern, and analyse one or more output light patterns in one or more images of the object to identify a defect in the object.
There is also provided a computer program comprising program instructions for causing a computer program to carry out the above method which may be embodied on a record medium, carrier signal or read-only memory.
BRIEF DESCRIPTION OF DRAWINGS
The novel features of the present invention are set forth in the appended claims hereto. The subject matter itself, however, as well as a preferred mode of use, further objectives, and advantages thereof, will best be understood by reference to the following detailed description of illustrative embodiments when read in conjunction with the accompanying drawings and wherein:
FIG.1 illustrates a system for inspecting an object, in accordance with an embodiment of the present invention;
FIG.2 illustrates an input light pattern used for inspecting the object, and images including first and second output light patterns, in accordance with an embodiment of the present invention;
FIG.3 illustrates various type of disruptions in the output light pattern caused by various defects in the object, in accordance with an embodiment of the present invention;
FIGs.4A and 4B illustrate an image of light pattern taken by the imaging device at a focal point inside the object when the air bubble is present in
the object, in accordance with an embodiment of the present invention;
FIG. 5A illustrate a processing module for analysing the images captured by the imaging device to identify a defect in the object, in accordance with an embodiment of the present invention;
FIG. 5B illustrates a histogram of the cross-validation tests performed for the processing module, in accordance with an embodiment of the present invention;
FIG.5C illustrates first and second example model outputs generated based on glass syringe detection and non-glass syringe detection by the processing module, respectively, in accordance with an embodiment of the present invention;
FIG. 6 illustrates another type of input light pattern that may be used for inspecting an object, in accordance with an embodiment of the present invention;
FIG.7 is a flowchart illustrating a method for inspecting an object, in accordance with an embodiment of the present invention; and
FIG.8 illustrates validation test results of inspection of various number of glass syringes/vials, in accordance with an embodiment of the present invention.
DETAILED DESCRIPTION OF INVENTION
FIG.1 illustrates a system 100 for inspecting an object 102, in accordance with an embodiment of the present invention. FIG.2 illustrates an input light pattern 202 used for inspecting the object 102, and images including first and second output light patterns 204 and 206.
The object 102 is typically any object that is capable of reflecting and/or refracting the input light pattern 202. Examples of the object 102 include but are not limited to: a glass vial filled with fluid, a glass syringe filled with fluid, a contact lens, a plastic tube, a metal surface. The inspecting the object 102 may include identifying a defect such as an immobile glass particle, an air bubble, a dust particle, a fiber particle, a glass abrasion, and a deformation defect in the object
102. For example, when the object 102 is a metal surface, the defect may include imperfections on the metal surface that reflect light. When the object 102 is a
glass vial/syringe filled with fluid, the defect may include an immobile glass particle, an air bubbles and dirt in the fluid and also deformation (non-smooth) defect in the glass.
The input light pattern 202 includes a plurality of colours disposed in parallel with respect to each other, and each colour separated from corresponding adjacent colour by a black line. However, it would be apparent to one of ordinary skill in the art, that the input light pattern 202 is not limited to the one shown herein, it may include other custom light patterns based on the type of object and defect to be identified. In an example, when the object 102 includes a glass vial/syringe filled with fluid, the light source 104 generates a custom input light pattern based on optical properties of the glass and fluid to emphasize difference between glass particles, dirt, fibers and air bubbles.
The system 100 includes a light source 104 for generating the input light pattern 202, the object 102 for receiving the input light pattern 202, an imaging device 106 for capturing images of the object 102 that includes the first and second output light patterns 204 and 206, and a processing module 108 for analysing the output light patterns 204 and 206 to identify a defect in the object 106. The system 100 further includes a robotic arm 110 attached to the object 102 for rotating the objectl 02 and enabling the imaging device 106 to capture the one or more images of the object 102 in one or more different orientations, i.e. from all required angles. The imaging device 106 includes, for example, a charge coupled (CCD) camera.
As shown herein, the light source 104 is disposed at a left side of the object 102 for backlighting the object 102 with the input light pattern 202, and the imaging device 106 is disposed at a right side of the object 102, such that the light source 104 and the imaging device 106 face each other, with the object 102 is in between. However, it would be apparent to one of ordinary skill in the art, that the light source 104 and the imaging device 106 may be positioned in such a manner that the light source 104 projects the input light pattern 202 on the object 102, and the imaging device 106 captures one or more images of the projected light pattern to enable identification of defects by identifying disruptions in the projected light pattern.
As stated in the text beside processing module 108 in fig. 1 , Acumen Al has many
modules which process information in their own unique way.
FIG.2 illustrates the object 102 as one of: a glass vial and a syringe filled with a fluid and images including first and second output light patterns 204 and 206. The object 102 refracts the output light patterns 204 and 206 when there is no defect in the object 102.
In the plan view, the object 102 acts as a cylindrical convex lens that refract the input light pattern 202 and forms the output light patterns 204 and 206 at a right side thereof. The images including output light patterns 204 and 206 are captured by the imaging device 106 at two different focal points.
The imaging device 106 generates an image including the first output light pattern 204 when it focusses on an outside wall of the object at a distance ‘D’ from the imaging device 106. Thus, the imaging device 106 captures an image of the object 102 at a first focal point that is outside a cylindrical wall of the object
102. The output light pattern 204 is a clear inverted image of the input light pattern 202 with respective black lines.
The imaging device 106 generates an image including the second output light pattern 206 when it focusses on an inside wall of the object 102 at a distance ‘F’ from the imaging device 106. Thus, the imaging device 106 captures an image of the object 102 at a second focal point that is inside of a cylindrical wall of the object 102. The output light pattern 206 is an inverted blurred image of the input light pattern 202. The output light pattern 206 includes a smooth pattern among adjacent colours, in which the transitions between colours are relatively smooth without black lines. Thus, the glass syringes/vials (i.e. object 102) with no defects are imaged as a smooth surface with transitioning light colours. If there is any glass particle, dirt, fiber, or air bubble inside or at a surface of the glass vial/syringe (i.e. object 102), then the smooth pattern may be disrupted.
Further, it is to be noted that when the object 102 includes a glass syringe/glass vial filled with fluid, the first and second distances ‘D’ and ‘F’ are determined based on the refractive index of the glass of the glass syringe/vial, and the fluid contained inside the glass syringe/vial.
The text beside the second output light pattern 206 in fig. 2 notes: images of light pattern taken by cameras in different focal points.
FIG.3 illustrates various type of disruptions in the output light pattern 300 caused by various defects in the object 102, i.e. glass syringes/vials filled with fluids. The output light pattern 300 is obtained by the imaging device 106 by focusing on an internal wall of the object 102. Figure 3 depicts air bubbles 304 and 306, glass particle 302 and dirt 308.
When the object 102, i.e. the glass syringe/vial includes a defect such as an immobile glass particle 302, then the output light pattern 300 includes a randomly dispersed irregular shaped light at corresponding position of immobile glass particle 302. The position of the randomly dispersed irregularly shaped light in the output light pattern 300 remains constant but its colour changes across the images obtained by rotating the object 102. The glass particle 302 is irregular in shape and refract the light from the input light pattern in random ways. It is highlighted as it disperses the light randomly and its colour changes as it rotate. The unique light pattern response/signature for immobile glass particles enables the identification and differentiation with other contamination/particles.
When the object 102, i.e. the glass syringe/vial includes a defect such as dirt/fiber 308, then there is not much change in the output light pattern 300. The dirt and fibers undergo minimal changes as they are imaged while being rotated. The unique light pattern response/signature for dirt and fibers enables the identification and differentiation with other contamination/particles.
When the object 102, i.e. the glass syringe/vial includes a defect such as an air bubble 304, then the output light pattern 300 includes a miniature light pattern similar to the input light pattern at corresponding position of the air bubble in the object 102.
Figure 4A depicts light pattern 202 on the left. On the right is a plan view of the syringe I vial. The text to the right of light pattern 400 in fig. 4A states: images of light pattern taken by camera focal point inside container. The text to the right
of light pattern 400 in fig. 4B states: images of light pattern at left of bubble.
Further, referring to FIGs.4A and 4B, the air bubble 304 near the edge of the syringe/vial 102 acts as a concave lens that generates a miniature light pattern 402 similar to the input light pattern 202 at a left side of the air bubble 304. The miniature light pattern 402 due to the air bubble 304 is in sharper focus than the light pattern 400 where there is no air bubble 304 and as the air bubble 304 acts as a concave lens, the miniature light pattern 402 is not inverted. The black lines in the miniature light pattern 402 are clearly visible in the air bubble 304 and as the miniature light pattern 402 is not inverted, the revealed bubble 304 is highlighted and easily identifiable. The air bubbles have a distinct pattern acting as optical lenses inverting the structured light pattern.
Referring FIGs.1 and 5A, the processing module 108 is an Artificial Intelligence (Al) based module that employs a machine learning model for processing the information captured by the imaging device 106 for identifying the defect in the object 102.
The text in the boxes of figure 5A reads as follows:
The box labelled 506 is entitled “Anomaly Detection”. The text under the graph reads: In this syringe abnormal compared to all other syringes that do not have immobile glass particles?
The box to the right of that shows the value 65% and below the text reads: Anomaly score for the syringe.
The box labelled 504 is entitled “Object Detection”. The text under the image reads: Where is the defect located in this syringe??
In the box to the right of that the text reads: Locations of the defects.
The box labelled 502 is entitled “Temporal Classification”. The text under the image reads: Find correlations between temporal images of the syringe?
In the box to the right of that the text reads: Probabilities of different defects based upon their behaviour.
The box labelled 508 is entitled “Classifier Support Vector Machine”. The text under
the image reads: Trained on the outputs from all the other modules to generate final predictions.
The machine learning model includes at least one of: a classification model, a temporal classification model 502, an object detection model 504, a temporal object detection model, and an anomaly detection model 506. The Al based module 108 employs a single or ensemble of computer vision/machine learning/deep learning algorithms to accurately detect glass particles and exclude air bubbles, dirt and fibers. The ensemble modeling is a process where multiple diverse machine learning models may be combined to predict an outcome differentiating between air bubbles, dirt, fibers and immobile glass particles on internal surfaces of glass syringes/vials.
In the processing module 108, the output of the anomaly detection module 506, the object detection module 504, and the temporal classification module 502 is provided to a Support Vector Machine (SVM) classifier 508. The SVM classifier
508 is trained on the SVM Classifier Training data, wherein cross validation is used to get an average accuracy of the final ensemble model. A histogram (as shown in FIG.5B, entitled “Stacked Model Test Set Accuracy Distribution. The wording in the middle of the graph to the right of 30 on the y-axis reads “Cross Validation Accuracy Distribution”) of the cross-validation tests is shown opposite giving an average accuracy of 96% as marked by the vertical line. The SVM may be finally trained on a complete SVM Classifier Training dataset prior to being used for testing on the validation dataset.
FIG.5C illustrates first and second example model outputs 510 and 512 generated based on glass syringe detection and non-glass syringe detection by the processing module 108 respectively. Both the first and second example model outputs 510 and 512 illustrates model probability for anomaly detection model, object detection model, and temporal object detection model of the processing module 108. On the graph on the left: the plot beginning at around 1 .0 on the y-axis is anomaly detection, the plot beginning just under 0.6 on the y-axis is object detection. The plot beginning at around 0.1 on the y-axis is temporal. On the graph on the right: the plot beginning at around 0.1 on the y-axis is anomaly detection, the plot that rises just before frame 20 is temporal. The third plot is object detection.
FIG. 6 illustrates another type of input light pattern 600 that may be used for
inspecting an object such as the object 102.
FIG.7 is a flowchart illustrating a method 700 for inspecting an object, in accordance with an embodiment of the present invention. At step 702, the method includes generating an input light pattern by a light source. At step 704, the method includes capturing an image of the object by an imaging device that includes an output light pattern. At step 706, the method includes analysing one or more output light patterns in one or more images of the object to identify the defect in the object.
Example
FIG.8 illustrates validation test results 800 of inspection of various number of glass syringes/vials by implementing the method and system of the present invention. It may be noted that the overall defect detection rate is found to be
96.5% (55 out of 57 syringes) with false positive rate as 0% (33 out of 33 syringes).
Although the present invention has been described with reference to specific embodiments, this description is not meant to be construed in a limiting sense. Various modifications of the disclosed embodiments, as well as alternate embodiments of the subject matter, will become apparent to persons skilled in the art upon reference to the description of the subject matter. It is therefore contemplated that such modifications can be made without departing from the spirit or scope of the present invention as defined.
The embodiments in the invention described with reference to the drawings comprise a computer apparatus and/or processes performed in a computer apparatus. However, the invention also extends to computer programs, particularly computer programs stored on or in a carrier adapted to bring the invention into practice. The program may be in the form of source code, object code, or a code intermediate source and object code, such as in partially compiled form or in any other form suitable for use in the implementation of the method according to the invention. The carrier may
1 3 comprise a storage medium such as ROM, e.g. a memory stick or hard disk. The carrier may be an electrical or optical signal which may be transmitted via an electrical or an optical cable or by radio or other means.
In the specification the terms "comprise, comprises, comprised and comprising" or any variation thereof and the terms include, includes, included and including" or any variation thereof are considered to be totally interchangeable and they should all be afforded the widest possible interpretation and vice versa.
The invention is not limited to the embodiments hereinbefore described but may be varied in both construction and detail.
Claims
1 . A system for inspecting an object, comprising: a light source for generating an input light pattern; the object for receiving the input light pattern; an imaging device for capturing an image of the object that includes an output light pattern; and a processing module for analysing one or more output light patterns in one or more images of the object to identify a defect in the object.
2. The system as claimed in claim 1 , wherein the light source is disposed at a left side of the object for backlighting the object with the input light pattern, and the imaging device is disposed at a right side of the object, such that the light source and the imaging device face each other, with the object in between.
3. The system as claimed in any preceding claim, wherein when the object is one of: a glass vial and a syringe filled with a fluid, the object acts as a cylindrical convex lens that refract the input light pattern to generate the output light pattern.
4. The system as claimed in any preceding claim, wherein the input light pattern includes a plurality of colours disposed in parallel with respect to each other.
5. The system as claimed in any of claim 1 -3, wherein the input light pattern includes a plurality of colours disposed in parallel with respect to each other, and each colour separated from corresponding adjacent colour by a black line.
6. The system as claimed in claim 5, wherein the output light pattern is a clear inverted input light pattern with respective black lines, when the imaging device captures the image at a first focal point that is outside the object and is at a first distance from the imaging device.
The system as claimed in claim 4, wherein there is no separation from corresponding adjacent colour by a line and, preferably, there is a smooth transition between the colours. The system as claimed in a n y of claims 4 and 5, wherein the output light pattern is a blurred inverted input light pattern including a smooth pattern among adjacent colors, when the imaging device captures the image at a second focal point that is inside the object and is at a second distance from the imaging device. The system as claimed in claim 5, wherein the output light pattern is a blurred inverted input light pattern including a smooth pattern among adjacent colors without black lines, when the imaging device captures the image at a second focal point that is inside the object and is at a second distance from the imaging device. The system as claimed in any of claims 6, 8 and 9, wherein the first and second distances are determined based on refractive indices of the object, and the fluid inside the object. The system as claimed in any of claims 7 -9 , wherein the processing module is configured to identify an immobile glass particle in the object as the defect, upon detecting a randomly dispersed irregularly shaped light in the smooth pattern, and wherein the randomly dispersed irregularly shaped light has a position which remains constant and a colour which changes across the one or more images obtained by rotating the object. The system as claimed in any of claims 7 - 9 , wherein the processing module is configured to identify an air bubble in the object as the defect, upon detecting a miniature light pattern similar to the input light pattern in the smooth pattern, and wherein the air bubble acts as a concave lens that receives the input light pattern, and generates the miniature light pattern at a left side of the air bubble. The system as claimed in any of claims 7 - 9 , wherein the processing module
16 is configured to identify a dirt particle and a fiber particle in the object as the defect, upon detecting a minimal change in the smooth pattern. The system as claimed in any preceding claim, wherein the processing module includes an Artificial Intelligence (Al) based processing system that employs a machine learning model for identifying the defect in the object, wherein the machine learning model includes at least one of: a classification model, a temporal classification model, an object detection model, a temporal object detection model, and an anomaly detection model. The system as claimed as claimed in any preceding claim, wherein the light source projects the input light pattern on the object, and the imaging device captures one or more images of the projected light pattern on the object. The system as claimed in any preceding claim further comprising a robotic arm attached to the object for rotating the object and enabling the imaging device to capture the one or more images of the object in one or more different orientations. The system as claimed in any preceding claim, wherein the defect includes at least one of: an immobile glass particle in the object, an air bubble, a dust particle, a fiber particle, a glass abrasion, and a deformation defect in the object. A method for inspecting an object, comprising: generating an input light pattern by a light source; capturing an image of the object by an imaging device that includes an output light pattern; and analysing one or more output light patterns in one or more images of the object to identify a defect in the object. The method as claimed in claim 18, wherein the light source is disposed at a left side of the object for backlighting the object with the input light pattern, and the imaging device is disposed at a right side of the object, such that the light source and the imaging device face each other, with the object in between, and the input light pattern includes a plurality of colours disposed
17 in parallel with respect to each other, and each colour separated from corresponding adjacent colour by a black line. The method as claimed in claim 18 further comprising generating the output light pattern as a clear inverted input light pattern with respective black lines, upon capturing the image at a first focal point that is outside the object and is at a first distance from the imaging device. The method as claimed in claim 18 further comprising generating the output light pattern as a blurred inverted input light pattern including a smooth pattern among adjacent colours without black lines, upon capturing the image at a second focal point that is inside the object and is at a second distance from the imaging device. The method as claimed in claim 20 further comprising identifying an immobile glass particle in the object as the defect, upon detecting a randomly dispersed irregularly shaped light in the smooth pattern, and wherein the randomly dispersed irregularly shaped light has a position which remains constant and a colour which changes across the one or more images obtained by rotating the object. The method as claimed in claim 20 further comprising identifying an air bubble in the object as the defect, upon detecting a miniature light pattern similar to the input light pattern in the smooth pattern, and wherein the air bubble acts as a concave lens that receives the input light pattern, and generates the miniature light pattern at a left side of the air bubble. A non-transitory computer readable medium having stored thereon computer-executable instructions which, when executed by a processor, cause the processor to: generate an input light pattern by a light source; capture an image of an object by an imaging device that includes an output light pattern; and
18 analyse one or more output light patterns in one or more images of the object to identify a defect in the object.
Applications Claiming Priority (2)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
GB2111915.1 | 2021-08-19 | ||
GB2111915.1A GB2609969A (en) | 2021-08-19 | 2021-08-19 | System and method for inspecting an object |
Publications (1)
Publication Number | Publication Date |
---|---|
WO2023021209A1 true WO2023021209A1 (en) | 2023-02-23 |
Family
ID=77913875
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
PCT/EP2022/073256 WO2023021209A1 (en) | 2021-08-19 | 2022-08-19 | System and method for inspecting an object |
Country Status (2)
Country | Link |
---|---|
GB (1) | GB2609969A (en) |
WO (1) | WO2023021209A1 (en) |
Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20090154789A1 (en) * | 2007-12-17 | 2009-06-18 | Gradience Imaging, Inc. | System and method for detecting optical defects |
WO2011128580A1 (en) * | 2010-04-13 | 2011-10-20 | Iris - Inspection Machines | Method for detecting defects in glassware articles, and installation for implementing said method |
EP3312592A1 (en) | 2016-10-21 | 2018-04-25 | Seidenader Maschinenbau GmbH | Device for detecting air bubbles in a container filled with liquid |
US20180156740A1 (en) * | 2016-12-07 | 2018-06-07 | Applied Vision Corporation | Identifying defects in transparent containers |
US20200408702A1 (en) * | 2019-06-26 | 2020-12-31 | Seidenader Maschinenbau Gmbh | Device for optical inspection of empty and liquid-filled containers |
US20210213486A1 (en) * | 2018-07-31 | 2021-07-15 | Amgen Inc. | Robotic system for performing pattern recognition-based inspection of pharmaceutical containers |
-
2021
- 2021-08-19 GB GB2111915.1A patent/GB2609969A/en active Pending
-
2022
- 2022-08-19 WO PCT/EP2022/073256 patent/WO2023021209A1/en active Application Filing
Patent Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20090154789A1 (en) * | 2007-12-17 | 2009-06-18 | Gradience Imaging, Inc. | System and method for detecting optical defects |
WO2011128580A1 (en) * | 2010-04-13 | 2011-10-20 | Iris - Inspection Machines | Method for detecting defects in glassware articles, and installation for implementing said method |
EP3312592A1 (en) | 2016-10-21 | 2018-04-25 | Seidenader Maschinenbau GmbH | Device for detecting air bubbles in a container filled with liquid |
US20180156740A1 (en) * | 2016-12-07 | 2018-06-07 | Applied Vision Corporation | Identifying defects in transparent containers |
US20210213486A1 (en) * | 2018-07-31 | 2021-07-15 | Amgen Inc. | Robotic system for performing pattern recognition-based inspection of pharmaceutical containers |
US20200408702A1 (en) * | 2019-06-26 | 2020-12-31 | Seidenader Maschinenbau Gmbh | Device for optical inspection of empty and liquid-filled containers |
Also Published As
Publication number | Publication date |
---|---|
GB2609969A (en) | 2023-02-22 |
GB202111915D0 (en) | 2021-10-06 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
TW229266B (en) | ||
JP6590876B2 (en) | Method and apparatus for non-destructive detection of non-dissolved particles in a fluid | |
CN105021628B (en) | Method for detecting surface defects of optical fiber image inverter | |
US7430047B2 (en) | Small container fluid dynamics to produce optimized inspection conditions | |
JP4246438B2 (en) | Lens inspection | |
Li et al. | Automated discrimination between digs and dust particles on optical surfaces with dark-field scattering microscopy | |
TWI294031B (en) | Method and system for inspecting optical devices | |
CN112912964A (en) | Defect detection for lyophilized pharmaceutical products using convolutional neural networks | |
US20230139131A1 (en) | Sheet Lighting for Particle Detection in Drug Product Containers | |
NO309213B1 (en) | Method and apparatus for inspecting liquid-filled containers | |
CN115769275A (en) | Deep learning platform for automatic vision inspection | |
CN108287165A (en) | Defect inspection method and defect inspection system | |
Karangwa et al. | Automatic optical inspection platform for real-time surface defects detection on plane optical components based on semantic segmentation | |
CN110709749B (en) | Combined bright field and phase contrast microscope system and image processing apparatus equipped therewith | |
WO2021054376A1 (en) | Learning process device and inspection device | |
WO2023021209A1 (en) | System and method for inspecting an object | |
Wang | RETRACTED ARTICLE: Application of deep learning to detect defects on the surface of steel balls in an IoT environment | |
JP2021157735A (en) | Image identification system, image identification device, program, and trained model | |
JP2014025884A (en) | Visual inspection method and visual inspection device | |
Soltane et al. | Estimating and monitoring laser-induced damage size on glass windows with a deep-learning-based pipeline | |
Nguyen et al. | Detection of weak micro-scratches on aspherical lenses using a Gabor neural network and transfer learning | |
FR3138213A1 (en) | Method and device for inspecting glass containers in at least two ways with a view to classifying the containers according to glass defects | |
WO2021181482A1 (en) | Microscopic image capturing method and microscopic image capturing device | |
KR20240057531A (en) | Device and method for quality inspecting of automotive parts based on ai | |
JPH0789054B2 (en) | Product defect inspection method and equipment |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
121 | Ep: the epo has been informed by wipo that ep was designated in this application |
Ref document number: 22769104 Country of ref document: EP Kind code of ref document: A1 |
|
WWE | Wipo information: entry into national phase |
Ref document number: 2022769104 Country of ref document: EP |
|
NENP | Non-entry into the national phase |
Ref country code: DE |
|
ENP | Entry into the national phase |
Ref document number: 2022769104 Country of ref document: EP Effective date: 20240319 |