Method for detecting defects in horizontal mold seams for glass containers

Abstract

The present invention relates to a method for detecting defects in a horizontal mold seal (JH) for a finish (B) of a plurality of glass containers (R), the method comprising the following steps: placing the container (R) between a light source and a camera, ensuring rotation of the container (R) for one rotation, acquiring an image by the camera at each container rotation increment such that the number of images per rotation is greater than 36, and analyzing the captured images for each container, the images being captured such that a contour of the end of the finish is detected in each image, comparing the contour of the end of the finish in the image with a reference contour of the end of the finish to detect deviations between these contours, and a defect in the horizontal mold seal (JH) for a container is detected if at least one image of the container has a deviation.

Description

Technical Field

[0001] The present invention relates to the technical field of inspecting objects, hollow articles, or generally transparent or translucent containers such as glass bottles, jars or flasks.

[0002] More specifically, the object of the present invention relates to the field of inspection of glass containers for detecting the presence or absence of defects in the finishing part of such glass containers with a horizontal mold seam in the finishing part.

Background Art

[0003] Generally, a container has a bottom, and the vertical wall rising from there ends at a part called the finishing part. There are various types of finishing parts depending on the equipped closing system, and as shown in FIG. 1, it includes an annular finishing part surface S at the upper part and an unfinished part CB at the lower part. This FIG. 1 represents a threaded finishing part including a thread on a cylindrical vertical part. It is known that glass containers are manufactured by a molding machine called an IS machine that includes various independent molding sections to which drops of malleable glass are supplied. Each of these molding sections is equipped with at least one blanking cavity with a blank mold and the same number of finishing cavities, and each of the finishing cavities has a blow mold where the container takes its final shape at high temperature.

[0004] Conventionally, the finishing part is formed in the blank mold. When the blank is transferred to the blow mold, the finishing part is already formed and the blank is held by the finishing part. For this purpose, the finishing part includes two half molds forming the vertical wall of the finishing part and the surface of the finishing part or the seamThe finish is formed by a finishing die, which consists of a mold section called a ring that forms the surface. The blank die also includes two half-dies for the blank body to form the walls of the blank body. During the press-and-blow process, a punch presses the glass against the half-dies for the blank body. During the blow-and-blow process, the punch shortens and enters the finishing die, but it is compressed air that presses the glass against the half-dies for the blank body. Therefore, the outside of the vertical wall of the finish is formed by the two half-dies for the finish, the top surface of the finish is formed by the ring, and the inside of the finish is formed by the punch. As a result, the finish of each container R has vertical molds on both sides of its vertical wall, corresponding to the interface between the two half-dies, as shown in Figure 1. seam The JV and the horizontal mold corresponding to the interface between the finishing mold section called the ring and the two finishing half-molds. seam Including JH. These finishing molds seam As the mold deteriorates with use, adjustments are made to leave marks or make them more or less noticeable on the container.

[0005] Horizontal mold seam It should be considered that JH is located slightly below the surface S of the finished part of the container. Also, the horizontal of the container mold seam It appears necessary to detect defects that affect the container and eliminate containers containing defects that could affect the aesthetic properties of the container, or, more seriously, could pose a real danger to the user.

[0006] In cutting-edge technology, AGR INTERNATIONAL offers DSG for accurately measuring the dimensional properties of containers. 40 lab monitoring machines show horizontal molds known as knockouts and flanges. seam This proposes the possibility of detecting defects in a horizontal mold. The machine, in particular, comprises a light source positioned on one side of the container and a camera positioned on the other side of the container. When capturing an image, the container is driven to rotate around its vertical axis. Such a machine is used for horizontal molds. seamIt does not appear to be designed to detect all defects. Furthermore, the monitoring rate of this machine is limited, so it cannot monitor containers at known production rates on the container production line.

[0007] Publication WO 2013 / 128538 describes horizontal molds. seam A method for detecting defects in the finish of a glass container is described. This method aims to ensure that the container itself is rotated so that a light ray is projected perpendicularly onto the finish. A linear camera collects the light reflected by the finish. This method includes analyzing the reflected light to infer defects if the contour of the reflected light changes. Although this method has a very high level of reliability in detecting small defects, it can only detect flanges that reflect light in the direction of the camera, and cannot quantify the dimensions of the defects, nor can it detect the horizontal shape of the container mold. seam It is not possible to detect various types of defects in this context.

[0008] Therefore, it is possible to detect small defects, typically less than 1 mm in size, at a high rate, while maintaining a very high level of reliability for horizontal detection of the container mold. seam While various defects can be detected in such containers, the horizontal of the mold of such containers seam If the required quality standards are met, there appears to be a need for technology that can avoid considering defective containers. [Overview of the project] [Problems that the invention aims to solve]

[0009] Therefore, the object of the present invention is to achieve a high rate for the finishing part of a glass container using a horizontal mold. seam This aims to satisfy this need by proposing a method for detecting defects in the horizontal of the mold of such a container. seam If it actually meets the required quality standards, then the horizontal mold will be used, while avoiding considering defective containers. seamIt has a very high level of reliability for detecting small defects in [the area]. [Means for solving the problem]

[0010] To achieve this objective, the present invention provides a horizontal mold for the finishing portion of a plurality of glass containers, each having a vertical axis. seam We propose a method for detecting defects in the following way, and the method for detecting defects in each container is as follows: The container is placed between the light source and the image capture camera, where the observation optical axis of the image capture camera is substantially perpendicular to an axis parallel to the vertical axis of the container, and the field of view of the image capture camera is horizontal to the mold. seam Including at least a portion of the left end or the right end of the finished portion, Ensure that the container rotates along the vertical axis so that it makes at least one rotation. The camera acquires images at each increment of the container's rotation so that the number of images per rotation is greater than 36, and Analyze the captured images for each container. Includes, The image above is Horizontal mold seam A finishing inspection zone extending to the height of the finishing section, including at least a portion of it, is defined in the image as follows: The contour of the edge of the finished part is detected in the image inspection zone. The image is configured to compare the contours of the edges of the finished portion with reference contours of the edges of the finished portion, and to detect the deviation between these contours of the edges of the finished portion and the reference contours of the edges of the finished portion, and Horizontal mold for containers seam A defect in the container is detected when at least one image of the container has a deviation. It was captured.

[0011] According to an advantageous modification of the embodiment, the camera has a field of view that is horizontal. seam At least a portion of and horizontal mold seamIt is arranged to include a reference whose height position is known with respect to it, and to define a finished part inspection zone in the image, and a horizontal mold seam A reference whose height position is known with respect to it is specified, and the finished part inspection zone is such that the finished part inspection zone is a horizontal mold seam It is positioned with respect to the reference so as to extend over a height including it.

[0012] For example, all or part of the surface of the finished part or the surface of the unfinished part is specified as a reference in the image.

[0013] Typically, there is an image capture camera whose horizontal width of the field of view is from 5 mm to 130 mm and whose height of the field of view is from 3 mm to 20 mm.

[0014] For example, there is an image capture camera having a resolution greater than 25 pixels / mm.

[0015] According to one preferred feature of the implementation, the image capture camera is arranged such that its observation optical axis substantially touches the left end or the right end of the finished part.

[0016] In the method according to the present invention, a camera equipped with a telecentric lens can be used.

[0017] According to one advantageous feature of the embodiment, there is a light source having vertical and horizontal illumination dimensions that are 100% to 200%, preferably 100% to 120%, of the value obtained by multiplying the dimensions of the camera's field of view by the distance between the light source and the camera's lens and dividing by the distance between the axis of the container and the camera's lens.

[0018] According to another embodiment, there is a telecentric light source, and its illumination area has dimensions greater than or equal to the dimensions of the camera's field of view.

[0019] According to one modified embodiment, the contour of the edge of the finished portion of the image is compared to a reference contour of the edge of the finished portion by using at least one contour obtained for at least one reference container to be fitted as a reference contour, and a deviation is detected.

[0020] According to another modification of the embodiment, the contour of the edge of the finished portion of the image is compared to the reference contour of the edge of the finished portion by using as the reference contour an average contour calculated over several rotational increment values ​​of at least one reference container to be fitted, and a deviation is detected.

[0021] According to another modification of the embodiment, the contour of the edge of the finished part of the image is compared with the reference contour of the edge of the finished part, using the contour of the edge of the finished part to which a low-pass filter has been applied as a reference contour, and the deviation is detected.

[0022] According to another modification of the embodiment, the contour of the edge of the finished part of the image is compared with a reference contour of the edge of the finished part by applying a high-pass filter to the contour of the edge of the finished part, and a deviation is detected.

[0023] Advantageously, for each image, the contour of the edge of the finished part and the reference contour of the edge of the finished part are compared by comparing at least one of the area, amplitude, and / or gradient measurements to a threshold for at least several elevations in the inspection zone, and a deviation is detected if at least one of these measurements exceeds this threshold.

[0024] Generally, the above method involves a horizontal mold with at least one defect selected from knockout, flange, and overhang / overmatch defects. seam This includes detecting defects.

[0025] According to one preferred embodiment, at least one detection criterion is defined for each of the following defects, namely knockout, flange, and overhang / overmatch, and the method includes identifying a detected defect from among these three defects by using at least one of these detection criteria.

[0026] According to one feature of the implementation, for at least one image, the shape of the deviation detected for several elevations in the inspection zone is selected as the detection criterion.

[0027] According to another feature of the implementation, the observation angle range of the deviation corresponding to the number of consecutive images in which the deviation is detected is selected as the detection criterion.

[0028] To identify overhang / overmatch defects in addition to knockout and flange defects, the shape of the deviations detected at several heights in the inspection zone is analyzed.

[0029] To distinguish knockout defects from flange defects, the observation angle range of the deviation is analyzed by considering a knockout defect to be detected when the observation angle range of the deviation falls below a certain maximum value.

[0030] To distinguish knockout defects from flange defects, the observation angle range of the deviation is analyzed by considering a flange defect to be detected when the observation angle range of the deviation exceeds a certain minimum value.

[0031] Conventionally, horizontal molds seam For each defective container, a signal is sent to remove that container from the production line.

[0032] According to one advantageous feature of the implementation, the container rotates along its vertical axis so as to complete at least one rotation during a maximum duration of 200 milliseconds.

[0033] Another object of the present invention is a horizontal mold for the finishing part of a glass container. seam The objective is to propose an inspection apparatus that implements a method for detecting defects in the present invention. [Brief explanation of the drawing]

[0034] [Figure 1] Figure 1 is an enlarged view showing the finished part of the glass container, the horizontal mold seam, and each of the vertical mold seams.

[0035] [Figure 2] Figure 2 is a schematic elevation view showing an apparatus for carrying out the method according to the invention for detecting horizontal mold seam defects in the finished portion of a glass container.

[0036] [Figure 3A] Figure 3A is a side view of the end of the finished portion of a glass container having a knockout-type horizontal mold seam defect. [Figure 3B] Figure 3B is a top view of the finished surface of a glass container having a knockout-type horizontal mold seam defect.

[0037] [Figure 4A] Figure 4A is a side view of the end of the finished portion of a glass container having a flange-type horizontal mold seam defect. [Figure 4B] Figure 4B is a top view of the finished portion of a glass container with a flange-type horizontal mold seam defect.

[0038] [Figure 5A] Figure 5A is a side view of the end of the finished portion of a glass container having an overhang / overmatch type horizontal mold seam defect. [Figure 5B] Figure 5B is a top view of the finished surface of a glass container having an overhang / overmatch type horizontal mold seam defect.

[0039] [Figure 6] Figure 6 is an example image of the rightmost edge of the finished portion of a glass container that has a defect in the horizontal mold seam.

[0040] [Figure 7] Figure 7 illustrates the extraction of an image of the contour of the end of the finished portion of a container having a defect in the horizontal mold seam.

[0041] [Figure 8] Figure 8 illustrates the extraction of the reference contour image of the edge of the finished portion from the compatible container.

[0042] [Figure 9] Figure 9 illustrates a comparison between the contour of the end of the finished portion of a container having a flange or knockout-type horizontal mold seam defect and the reference contour of the end of the finished portion.

[0043] [Figure 10] Figure 10 illustrates a comparison between the edge contour of the finished portion of a container having a horizontal mold seam defect of the container overhang / overmatch type and the reference edge contour of the finished portion.

[0044] [Figure 11] Figure 11 illustrates the extraction of the reference contour image of the end of a linearly shaped finished portion of a threaded finish.

[0045] [Figure 12] Figure 12 is an example image of the rightmost edge of the finished surface of a container with a knockout defect, showing the container's contour and reference contour. [Modes for carrying out the invention]

[0046] As can be understood more specifically from Figures 1 and 2, the object of the present invention is a horizontal mold in a glass container R. seamThe present invention relates to an apparatus 1 for carrying out a method for detecting JH defects. Conventionally, a container R has a bottom F from which a vertical cylindrical wall V rises along a vertical axis Z and ends in a portion called a finishing portion B. In the case of a bottle-type container R, the vertical cylindrical wall V emerges from the bottom F and is the portion that forms the body of the bottle, connected to a collar C via a shoulder portion E.

[0047] Each container R, as shown in Figure 1, has a horizontal mold corresponding to the interface between the mold portion called the ring and the two half-molds that form the vertical cylindrical wall V of the container. seam Includes JH. By definition, horizontal mold seam JH used the container seam It is located slightly below the surface S of the finishing part, which corresponds to the plane for which it is made. seam JH extends around the entire circumference of the finished section.

[0048] According to the present invention, the above method involves selecting a horizontal mold from among the defects of knockout JHK, flange JHF, and overhang / overmatch JHO. seam This includes detecting defects in each container at JH.

[0049] As can be understood more concretely from Figures 3A and 3B, the knockout JHK defect is a horizontal mold seam JH and the vertical mold between the ring-shaped mold section and the two half-molds that form the vertical wall. seam This corresponds to a glass flange at one of the T-shaped intersections of the joint. This defect corresponds to a glass flange caused by a blunt bulge at the interface of the three molds or due to poor adjustment between these molds. This knockout JHK takes the form of a tip-like or needle-like (often triangular in cross-section) glass projection, which is thin in thickness along the vertical axis Z and has a small horizontal spread, often less than 1 millimeter. As a result, it has a small circumferential angular range θ around the vertical axis Z. Typically, the knockout JHK has a small circumferential angular range θ of less than a few degrees.

[0050] As can be understood more concretely from Figures 4A and 4B, the flange JHF defect corresponds to excess glass resulting from a misalignment between the mold ring and the two half-molds that form the vertical wall. This flange JHF takes the form of a thin glass slide along the direction of the vertical axis Z. Furthermore, this flange JHF has a considerable angular range θ around the vertical axis Z. Typically, the flange JHF has a considerable angular range θ of 10° or more, but easily exceeding 30°. It can even extend further up to 180°. Rarely, the flange may be present around the entire circumference.

[0051] As can be seen more concretely from Figures 5A and 5B, the overhang / overmatch JHO defect corresponds to excess glass resulting from a misalignment between the mold ring and the two half-molds that form the vertical wall. This overhang / overmatch JHO takes the form of a very thick glass protrusion along the vertical axis Z and has a considerable angular range θ around the vertical axis Z. Typically, the ring offset can appear to exceed 180°, but the resulting step is generally noticeable over an angular range θ of 90° or more.

[0052] As will be described in detail below, the object of the present invention is a horizontal mold that has one or all of the following defects: knockout JHK, flange JHF, and overhang / overmatch JHO. seam The objective is to detect a container having a defect in JH. According to an advantageous modification of the embodiment, the object of the present invention is to select a horizontal mold defect from among knockout JHK, flange JHF, and overhang / overmatch JHO defects. seam The purpose is to identify defects in JH.

[0053] For this purpose, horizontal mold seam Apparatus 1 for implementing a method for detecting defects in JH includes a fixed light source 3 located on one side of container R and a fixed image-capturing camera 4 located on the other side of container R. This camera 4, including a lens 4a, captures the horizontal mold of container R. seamThe camera 4 is adapted to capture an image in which at least a portion of JH is visible. The camera 4 analyzes the captured image and the horizontal mold of container R. seam It is connected to an analysis processing unit 5 configured to detect defects in JH.

[0054] The image of container R is captured while container R rotates at least once along its vertical axis Z, thereby creating a horizontal mold. seam The entire JH can travel in front of the camera. Each container R is supported by a rotating system 6. For example, the rotating system 6 includes a sliding or laying surface 7 for the bottom F of the container R, as well as all types of drive systems 8 known to be used. The drive system 8 is operated so that each container R stays in place for the time necessary to make at least one rotation between the light source 3 and the camera 4, during which time images are captured by the camera 4, as will be described later.

[0055] For example, the rotating system 6 includes a wheel or spinner 8 as a drive system that rotates the container by friction, while the container is supported by at least two free casters or rollers 9. The rollers, for example, constitute part of a transport star wheel 10 that transports the container R along a circular path, continuously supporting and sliding the container R on the laying surface 7 in front of the detection device 1. Advantageously, this transport star wheel 10 constitutes part of an inspection machine that inspects more than 150 containers per minute and includes one or more stations for inspecting containers in motion at the exit of the production line. In other words, the detection device 1 according to the present invention can be implemented without reducing the run rate of containers at stations that integrate the rotation of containers for this inspection, in addition to inspection stations that were conventionally installed to inspect containers on the production line.

[0056] According to the features of the present invention, the inline inspection machine inspects 50 to 500 containers per minute, typically 150 to 450 containers per minute. For each container, the method includes a transport step, which involves transporting the container R and positioning it between the light source 3 and the image acquisition camera 4, and a step, which involves ensuring that the container R rotates along the vertical axis Z to at least one rotation, during which an image is acquired. It should be noted that with the transport star wheel 10, a new container to be inspected is brought in while previously inspected containers are transported to the inspection station. Therefore, the inline inspection machine is equipped with high-rate handling means. At a rate of 150 containers per minute, the duration of the rotation and inspection steps is up to 200 milliseconds. According to one preferred feature of the present invention, a complete rotation of the container lasts up to 100 milliseconds. Rotating the container 1.5 times is also a common choice.

[0057] In other words, the method according to the present invention involves placing a container R between a light source 3 and an image capture camera 4, where the observation optical axis of the image capture camera 4 is substantially perpendicular to an axis parallel to the vertical axis Z of the container, and the field of view of the image capture camera 4 is horizontal. seam This includes ensuring that the container R rotates along the vertical axis Z so that it completes at least one rotation in less than 200 milliseconds, and that the left end or right end of the finishing portion includes at least a portion of the container, and that the camera 4 acquires an image at each rotation increment of the container such that the number of images per rotation is greater than 36.

[0058] According to one feature of the present invention, the camera 4 is operated so that an image is captured at each rotation increment of the container such that the number of images per rotation is greater than 36. In other words, the method according to the present invention aims to acquire at least one image for every 10° rotation of the container R. For example, the number of images over 360° of the container R is between 36 and 96. The rotation increment of the container between each captured image represents the angular sector through which the finishing section passes, for example, in the range of 10° to less than 1°. Of course, with a high-speed camera with an integration time of less than 500 microseconds and a readout time of 0.5 milliseconds, it is possible to acquire 400 images per rotation in 200 milliseconds. Increasing the acquisition frequency leads to improved accuracy, particularly in estimating the radial length of the knockout. As a result, the cost of the system increases due to the price of the high-speed camera 4 and the computing power that the analysis processing unit 5 must possess.

[0059] According to the present invention, the light source 3 and camera 4 capture the horizontal mold of the container R during the rotation of the container around the vertical axis Z. seam Image I, in which at least a portion of JH is visible, is adapted so that the camera can acquire it. As shown in Figure 2, conventionally, the vertical axis Z of the container is considered to be parallel to the vertical direction z of the orthogonal reference frame x, y, z. This vertical direction z is perpendicular to the transverse direction y, which passes through the light source 3 and camera 4, while the transverse direction x is perpendicular to the transverse direction y and the vertical direction z.

[0060] The container R is located between the light source 3 and the camera 4, and the observation optical axis Y of the camera 4 extends parallel to the transverse direction y, i.e., substantially perpendicular to the vertical direction z. The camera 4 has a field of view that extends laterally along the transverse direction x, which is perpendicular to the vertical direction z and the observation optical axis Y. Thus, the field of view of the camera 4 extends within a plane defined by the vertical direction z and the transverse direction x. In a preferred example, the camera 4 is positioned so that its optical axis is substantially tangent to the left or right end of the finished part.

[0061] Following the vertical direction z, the field of view of camera 4 includes at least the horizontal mold. seamJH is included. The field of view of camera 4 includes at least the left edge or the right edge of the finishing section B. According to a modification of one preferred embodiment, the field of view of camera 4 includes only either the left edge or the right edge of the finishing section B in order to obtain good resolution. Of course, the method according to the present invention can be carried out using a camera whose field of view is not limited to the right edge or the left edge. In this preferred modification, the inline inspection machine rotates the container at least once, and observation of one side is of the horizontal mold of the finishing section. seam It has been shown to be sufficient to ensure the observation of defects in JH, making it particularly suitable for high rates. Furthermore, one-sided observation allows for optimal use of the camera's field of view, and its optical axis can be positioned to touch the edge of the finished part being observed.

[0062] Camera 4, like a matrix camera, transmits a horizontal projection of the finished portion for each rotation increment to the analysis processing unit 5, in which at least one end of the finished portion, referred to as the left end or right end, is prominent, taking into account their positions as they appear in each image captured by the camera (Figure 6). Since the rotating container does not strictly include either the left or right side, the conditions of left or right are given from the viewpoint of observation in all shooting. In the example shown in Figure 6, image I corresponds to the right end of container R.

[0063] Therefore, each image captured by the camera shows at least the left end or the right end of the finishing section B, which indicates the horizontal mold. seam It becomes possible to visualize at least a portion of JH and at least the contour P of the end of the finished portion. In other words, as can be seen in Figure 6, each image I is a horizontal mold at an altitude or height along a direction parallel to the vertical z. seam It must include at least a portion of the outer shape or contour P of the end of the finished part, which includes at least a portion of JH. Therefore, in each image, the horizontal mold seam An inspection zone ZI is defined that corresponds to a horizontal band-shaped portion of the image having a height determined to include a portion of the contour P at the end of the finished part, which includes a part of JH.

[0064] According to the advantageous features of the implementation, camera 4 has a horizontal field of view. seam At least a portion of JH and horizontal molds seam It is positioned to include a reference whose altitude position is known relative to JH. Therefore, after finding the reference in the image, the finishing inspection zone is definitely horizontal. seam It is possible to position the finishing inspection zone ZI relative to the reference so as to extend over a height that includes the reference area.

[0065] Of course, this image capture takes into account the variation in the positioning of container R relative to camera 4 and the manufacturing tolerances of container R, using this standard and horizontal mold. seam The process is executed with an added margin to visualize a portion of the JH. If the container height can vary by + / - 2 mm, this margin is, for example, at least + / - 2 mm.

[0066] It must be understood that the criteria presented by container R are, for example, notable elements such as contours, angles, and shapes that can be shown in the image. These notable elements can also be seen in all images captured during the rotation of the article, and their vertical position in each image can be uniquely determined, as can the horizontal mold. seam Since the vertical distance Dv to JH is known in each image, it is preferable that this distance be constant. This reference location in the image allows for the assignment of a positional reference frame to each image, which can then be used for image analysis.

[0067] Typically, for example, all or part of the finished surface S or the unfinished surface CB may be used as a reference if the container includes it. When the finished surface S is used as the reference, the camera 4 is adjusted so that the height of its field of view includes a margin to account for variations in container positioning and manufacturing tolerances in addition to the finished surface S. Similarly, the camera 4 is adjusted so that the height of its field of view includes a horizontal mold. seam It is adjusted to include a margin in addition to JH.

[0068] The horizontal width of the field of view of camera 4 is adapted to include the contour P of the edge of the finished part with a margin and a reference, i.e., a portion of the surface of the finished part in the example considered. According to one preferred modification, the width of the field of view is selected to include only the necessary margin in addition to the edge of the finished part. For example, camera 4 preferably has a horizontal field of view width of 5 mm to 80 mm and a field of view height of 3 mm to 20 mm. This allows for optimization of the camera's field of view, i.e., it is possible to limit the unnecessary area of ​​the sensor. According to another modification, the field of view may reach 130 mm in width to observe a finished part with a diameter of 120 mm and a margin of 10 mm, with the two left and right edges of the finished part included in the field of view.

[0069] For example, camera 4 is a matrix camera combined with an optical lens, making it possible to observe the inspection zone with a resolution greater than 25 pixels / mm overall.

[0070] Light source 3 is made in any suitable manner to provide backlight for container R, which is adapted to capture images by camera 4.

[0071] According to one preferred feature of the embodiment, the light source 3 has determined vertical and horizontal illumination dimensions. Conventionally, the vertical dimension is taken along the vertical direction z, and the horizontal dimension is parallel to the transverse direction x. In a non-telecentric optical system, the dimension DL of the light source 3 is of the type DL = CH × DI / DC; where CH is the field of view of the camera 4, DI is the distance between the light source 3 and the camera lens 4a, and DC is the distance between the inspected zone and the camera lens, more specifically, DC is considered as the distance between the camera lens (e.g., the optical center of a non-telecentric lens) and the focal plane, where the focal plane includes the vertical axis Z of the container and the point of contact between the end of the finished part of the ray passing through the optical center. The vertical and horizontal illumination dimensions are the dimensions of the effective area of ​​the light source 3 observable by the camera 4.

[0072] According to one feature of the embodiment, the light source 3 has vertical and horizontal illumination dimensions of 100% to 200%, preferably 100% to 120%, of the value obtained by multiplying the dimensions of the field of view of the camera 4 by the distance between the light source 3 and the lens 4a of the camera, and dividing that by the distance between the vertical axis Z of the container and the lens 4a of the camera.

[0073] According to one modification of the embodiment, the light source 3 may be a telecentric light source whose illumination area is larger than the dimensions of the camera's field of view. Therefore, the light source 3 generates a beam in which its average rays are parallel to the camera's optical axis relative to the camera's field of view. Note that in this case, the camera's lens 4a is a telecentric lens.

[0074] The images captured by camera 4 for each container R are of the horizontal mold of container R. seam The horizontal mold is analyzed by an analysis processing unit 5 configured to detect defects in the JH. seam The method for detecting defects in JH involves, for each container, the captured image and, in particular, the horizontal mold in each image. seam This method includes analyzing a finishing inspection zone ZI extending over the height of the finishing portion, which includes at least a portion of the finished portion. This analysis method includes detecting the contour P of the edge of the finishing portion in the image inspection zone, and comparing the contour P of the edge of the finishing portion in the image with a reference contour Pf of the edge of the finishing portion to detect the deviation between these contours P of the edge of the finishing portion and the reference contour Pf of the edge of the finishing portion. Horizontal mold for container seam A defect in the horizontal mold is detected if at least one image of the container R has a deviation. seam For each container having a JH defect, the analytical processing unit 5 transmits a signal indicating the container's poor quality, and such a signal can control the discharge device for removing the container from the production line.

[0075] The detection of the contour P at the edge of the finished part in each captured image can be performed by any appropriate image processing method.

[0076] As explained above, it is preferable that the image analysis be limited to the finishing inspection zone ZI, where the horizontal mold is located. seam A portion of the contour P of the finished part, including the part, is visible. Advantageously, the inspection zone is positioned relative to the container reference shown in the image, and its elevation is horizontal. seam This is known for this. In a modified version of this implementation, the analysis of the contour of the finished part is performed on the horizontal mold. seam While limiting it to the parts necessary to characterize the horizontal mold, seam Other contours that could be confused with it, such as the contours containing screw threads, have been eliminated.

[0077] According to this method, horizontal mold seam In relation to that reference, a known altitude position is identified in each image, and the finishing inspection zone is always horizontal in relation to that reference. seam It is positioned to extend across the height including JH.

[0078] Figure 7 shows, as an example, the detection of the contour P at the right edge of the finished portion in image I, where the finished portion surface S is shown as a reference. The first step involves identifying the height, i.e., the position of the finished portion surface S along the vertical direction z. For example, along a vertical line continuing from the left edge of the image, the change in the vertical gray level corresponding to the finished portion surface S is detected. Next, the image inspection is performed on the horizontal mold. seam Based on this criterion, inspection zone ZI can be positioned so that JH is inevitably taken into consideration.

[0079] Of course, the method for searching for the criterion can vary. For example, if the criterion is an unfinished area, the search for the largest two-dimensional correlation can be performed between the image and a criterion window containing a trained model of the unfinished area, or the visible right or left edge of the unfinished area.

[0080] The second step is, for example, from the surface S of the finishing part, to a horizontal mold that is below the surface of the finishing part. seamThis includes extracting the outer shape or contour P of the end of the finished portion up to the lower limit of inspection zone ZI, which is located beyond ZH. As shown in Figure 7, the contour P of the end of the finished portion is a horizontal mold. seam This includes JH. Searching for the first contour point P at the end of the finished portion may involve viewing the image horizontally at a predetermined height relative to a reference until a change in the horizontal gray level is detected.

[0081] Extraction of the contour P of the edge of the finished part in each image can be performed using any image processing method. For example, it may be provided in the inspection zone ZI to find the outer edge of the finished part by searching for black / white transitions Ti along the vertical z direction for each elevation, or by searching for white / black transitions from the outside. The position of these transitions Ti along the lateral x direction is determined with respect to the vertical z direction. This search for transitions Ti is performed at each elevation across the entire elevation of the inspection zone ZI. As shown on the right of Figure 7, for each image, the contour P of the edge of the finished part may be obtained in the form of a curve referenced by planes z, x. For example, the contour P consists of all transitions Ti, and the coordinates x and z of each transition are known.

[0082] The next step involves comparing the contour P of the end of the finished portion with a reference contour Pf of the end of the finished portion, obtained by various possible means. According to one preferred modification of the present invention, as understood in Figure 8, the extraction of the reference contour Pf from the end of the finished portion is performed in the same manner as the extraction of the contour P of the end of the finished portion of the container shown in Figure 7. For this purpose, a defect-free horizontal reference mold is used. seam When an image is captured of a reference container that includes a reference contour Pf at the end of the finished portion, which also includes a portion of the finished portion, it becomes possible to visualize this reference contour Pf at the end of the finished portion, and the surface of the finished portion as a reference. This reference image is advantageously created under the same conditions as the image of the container under inspection by using the inspection apparatus by placing the reference container between the camera 4 and the light source 3. By analyzing this image in the manner described above, a defect-free horizontal mold is created. seam This makes it possible to extract the reference contour Pf of the edge of the finished part where the characteristic appears.

[0083] According to another modification of the embodiment, the reference contour Pf of the edge of the finished part is obtained by learning from several images of the container, or from images of several containers to be fitted. For example, any type of mathematical operation can be performed on the images or on the contours extracted from the images. Thus, one or more contours obtained for a reference container to be fitted, or for several reference containers to be fitted, may be selected as the reference contour Pf. Similarly, the average contour calculated over several rotation increments of a reference container to be fitted, or for several reference containers to be fitted, may be selected as the reference contour Pf. According to another modification of the embodiment, the reference contour Pf of the edge of the finished part may also be obtained from a manufacturing drawing or from an arbitrary geometric model of the container.

[0084] According to another modification of the present invention shown in Figure 11, for example, in the case of the threaded finish shown in Figure 1, the end of the finish includes a cylindrical portion, so the contour of the end of the finish includes the right portion PD. In this case, the reference contour Pf of the end of the finish can be a simple reference straight line D that best passes through the straight portion PD of the contour P of the end of the finish, and thus passes through a point Ti in this portion of the outer shape of the finish, or a straight line parallel thereto.

[0085] According to another modification of the embodiment shown in Figure 12, the present invention aims to use the contour of the edge of a finished part to which a low-pass filter has been applied as the reference contour Pf. In other words, the reference contour Pf corresponds to the contour of the edge of a finished part to which filtering or smoothing has been applied. In the example in Figure 12, the contour P (white) of the edge of the finished part corresponds to the raw contour extracted from the image, and the reference contour Pf (gray) corresponds to the moving average of contour P. Of course, various types of low-pass filters can be applied, such as averaging filters, Gaussian filters, median filters, or phase-shifted median filters.

[0086] It should be noted that, according to another modification of the embodiment, a high-pass filter may be applied to the contour of the edge of the finished part in order to compare the contour P of the edge of the finished part in the image with a reference contour Pf of the edge of the finished part. Applying a high-pass filter, for example a gradient type, makes it possible to highlight deviations, i.e., strong or rapid fluctuations in the derivative of the contour of the edge of the finished part, which reflect the presence of local irregularities. The presence of such deviations is a horizontal mold seam This corresponds to a defect in [the relevant field].

[0087] A comparison step for each image between the contour P of the finished edge and the reference contour Pf of the finished edge leads to the detection of whether there is a deviation between these contours of the finished edge and the reference contour of the finished edge. Horizontal mold for container seam A defect in the container is detected if at least one image of the container has a deviation.

[0088] Horizontal mold seam This comparison, which leads to the observation of whether or not defects are present, can be carried out according to various analytical methods.

[0089] For each image, the contour P of the finished edge and the reference contour Pf of the finished edge are compared for at least several elevations, preferably all elevations in the inspection zone ZI. This step aims to compare at least one of the area, amplitude, and / or slope measurements to a threshold, and a deviation is detected if at least one of these measurements exceeds this threshold. Typically, the area measurement may correspond to the area between the contour P of the finished edge and the reference contour Pf of the finished edge. The amplitude measurement may correspond to the difference between the contour P of the finished edge and the reference contour Pf of the finished edge along the lateral x direction. The slope measurement is obtained by subtracting the contour P of the finished edge and the reference contour Pf of the finished edge. This could correspond to a measurement of the slope of the curve. Another possible measurement is the difference between two amplitudes: the contour P of the edge of the finished part and the reference contour Pf of the edge of the finished part, at the same or different altitudes.

[0090] According to one modification of the embodiment, in order to compare the contour P of the end of the finished portion with a reference contour Pf of the end of the finished portion, the actual position of the end of the finished portion is taken into consideration, which includes the tilt in the image related to the tilt of the finished portion with respect to the Z axis due to operational interference during monitoring or lack of verticality of the container. Such tilt of the finished portion is shown, for example, in Figure 11. Note that in this simplified modification, if the reference contour Pf of the end of the finished portion is a straight line D passing through a straight portion PD of the contour of the finished portion, the deviation can be determined by measuring the distance between multiple points on the contour P of the finished portion and the reference straight line D.

[0091] In the example shown in Figure 9, curve CEc is a curve corresponding to the deviation between the contour P of the end of the finished part and the reference contour Pf of the end of the finished part, and this curve gradually develops along the vertical axis z, i.e., according to the altitude in the inspection zone. Various examples of possible measurements of the deviation curve CEc make it possible to estimate the size and / or shape of the deviation, and at the same time, it is possible to estimate the presence of a defect, and in some cases its type and risk. For example, the point of maximum amplitude of this curve corresponding to the edge of the projection furthest from the vertical axis Z of the vessel can be considered. This maximum amplitude corresponds to the height of the peak of the deviation curve CEc, which gives an accurate estimate of the radial length of the knockout or flange and constitutes a very good physical criterion for the risk of a defect. This amplitude can be compared to a threshold, and if this measurement exceeds this threshold, a deviation (or defect) is detected.

[0092] The threshold is adjustable to parameterize the strictness of the test; for example, the operator can decide that a slightly marked knockout is acceptable.

[0093] If the measurement that enables the detection of deviation is the surface area obtained by subtracting the contour of the edge of the finished part from the reference contour of the edge of the finished part, the measurement is, for example, the integral of the deviation curve CEc. If the measurement is a gradient, it may be the maximum gradient of the deviation curve CEc, or the deviation between two consecutive gradients of the deviation curve CEc on either the peak or threshold crossing side.

[0094] From the above description, the method according to the present invention relates to a horizontal mold that has been selected from defects such as knockout JHK, flange JHF, and overhang / overmatch JHO. seam It is clear that this includes detecting JH defects.

[0095] According to an advantageous modification of the embodiment, the method according to the present invention makes it possible to identify defects detected from among these three defects, namely knockout JHK, flange JHF, and overhang / overmatch JHO. For this purpose, at least one detection criterion, namely knockout, flange, and overhang / overmatch, is defined for each defect, and at least one of these detection criteria is used to identify the defect. Thus, it is possible to identify the type of defect and improve the accuracy of container inspection.

[0096] According to one exemplary embodiment, the shape of the deviation detected at several elevations in the inspection zone is selected as a detection criterion for at least one image. For example, the shape of the deviation can be described in various ways, for example, using gradient or amplitude measurements at several consecutive elevations. This shape can be observed in the deviation curve CEc. In Figure 9, the deviation curve CEc has a peak characterizing a knockout or flange, but if the defect is an overhang / overmatch as in Figure 5A, the deviation curve CEc is as shown in Figure 10. It will take on a staircase shape.

[0097] According to another exemplary embodiment, an observation angle range of the deviation corresponding to a number of consecutive images is selected as the detection criterion, where the deviation is detected, i.e., at the rotation angle of the container, while the deviation is detected in the image. As can be clearly seen in Figures 3B, 4B, and 5B, the observation angle range of the deviation for flange or overhang / overmatch type defects is very close to the circumferential angle range of the defect viewed from above. For knockouts, the observation angle range is larger than its circumferential angle range. For example, if the rotation increment between two image acquisitions is 4°, and the knockout has a very small circumferential angle range of 2°, and it is long enough, then the observation angle range will be 8° or 12°, respectively, as it can be seen in several consecutive images, each divided in time by one or more rotation increments, into two or three sections.

[0098] Of course, these standards can be implemented independently or in combination sequentially, and one standard can be applied before another, and vice versa.

[0099] In one example of implementation, the shape of the deviations detected at several heights in the inspection zone is analyzed to identify overhang / overmatch JHO defects in contrast to knockout JHK and flange JHF defects. Indeed, as can be clearly seen from Figure 5A, overhang / overmatch JHO defects are distinguished from knockout (Figure 3A) and flange (Figure 4A) defects, which are characterized by sharp protrusions, particularly by their stepped or wall-like shape.

[0100] In another example of implementation, the observation angle range of the deviation is analyzed, and knockout JHK defects are identified as flange JHF defects, taking into account that knockout defects are detected when the observation angle range of the deviation falls below a maximum value, for example, defined as 30°. In other words, knockout JHK defects are detected when the number of consecutive images in which the deviation is detected corresponds to an observation angle range below a maximum value, for example, defined as 30°.

[0101] In another example of implementation, the observation angle range of the deviation is analyzed, and a knockout JHK defect is identified as a flange JHF defect, taking into account that a flange defect is detected when the observation angle range of the deviation exceeds a minimum value defined, for example, 30°. In other words, a flange JHF defect is detected when the number of consecutive images in which the deviation is detected corresponds to an observation angle range that exceeds a minimum value defined, for example, 30°.

[0102] According to the present invention, it is preferable that identification criteria, such as thresholds applied to measurements to detect deviations, those applied to comparisons of contour shapes, and / or those applied as minimum or maximum values ​​of the observation angle range, are stored in the analysis processing unit 5 and are adjustable by the operator according to the inspected container and the required quality.

Claims

1. A method for detecting defects in the horizontal mold seam (JH) of the finished portion (B) of a plurality of glass containers (R), each having a vertical axis (Z), wherein the method for detecting the defect for each container comprises the following steps: The container (R) is positioned between the light source (3) and the image capture camera (4), wherein the observation optical axis of the image capture camera is substantially perpendicular to the axis parallel to the vertical axis (Z) of the container, and the field of view of the image capture camera includes at least a portion of the horizontal mold seam, and at least the left end or the right end of the finished portion. Ensure that the container (R) rotates along the vertical axis (Z) to at least one rotation. Images are acquired by the image acquisition camera (4) at each rotation increment of the container such that the number of images per rotation is greater than 36, and The captured images are analyzed for each container. Includes, The aforementioned image is, The finishing inspection zone (ZI), which extends up to the height of the finishing portion and includes at least a portion of the horizontal mold seam, is defined in the image as follows: The contour (P) of the end of the finished portion is detected in the finished portion inspection zone (ZI), The contour (P) of the end of the finished portion in the image is compared with a reference contour (Pf) of the end of the finished portion to detect the deviation between these contours of the end of the finished portion and the reference contour of the end of the finished portion, and A defect in the horizontal mold seam (JH) of the container is detected when at least one image of the container has a deviation. It was captured, The method includes identifying defects detected from among knockouts (JHK) and flanges (JHF) based on at least one detection criterion, wherein the at least one detection criterion is an observation angle range of the deviation corresponding to the number of consecutive images in which the deviation was detected. method.

2. The method according to claim 1, wherein the image capturing camera (4) is positioned such that its field of view includes at least a portion of the horizontal mold seam (JH) and a reference whose height position relative to the horizontal mold seam is known, and defines the finishing inspection zone (ZI) in the image, the reference whose height position relative to the horizontal mold seam is known is identified, and the finishing inspection zone (ZI) is positioned relative to the reference such that the finishing inspection zone (ZI) extends over a height including the horizontal mold seam.

3. The method according to claim 2, wherein all or part of the surface (S) of the finished portion or the surface (CB) of the unfinished portion is identified in the image as a reference.

4. The method according to any one of claims 1 to 3, wherein the image acquisition camera (4) has a horizontal field of view of 5 mm to 130 mm and a height of the field of view of 3 mm to 20 mm.

5. The method according to any one of claims 1 to 4, wherein the image capturing camera (4) is positioned such that its observation optical axis is substantially in contact with the left end or the right end of the finishing portion.

6. The method according to any one of claims 1 to 5, wherein the light source (3) has vertical and horizontal illumination dimensions of 100% to 200%, preferably 100% to 120%, of the value obtained by multiplying the dimensions of the field of view of the image capturing camera (4) by the distance between the light source and the lens (4a) of the image capturing camera (4), and dividing that by the distance between the axis of the container and the lens of the image capturing camera (4).

7. The method according to any one of claims 1 to 6, wherein the contour (P) of the end of the finished portion in the image is compared with the reference contour (Pf) of the end of the finished portion by using at least one contour obtained for at least one reference container to be fitted as a reference contour (Pf), and a deviation is detected.

8. The method according to any one of claims 1 to 6, wherein the contour (P) of the end of the finished portion in the image is compared to the reference contour (Pf) of the end of the finished portion by using the average contour (Pf) calculated over several rotational increment values ​​of at least one reference container to be fitted as the reference contour (Pf), and a deviation is detected.

9. The method according to any one of claims 1 to 8, wherein the contour (P) of the end of the finished portion in the aforementioned image is compared with the reference contour (Pf) of the end of the finished portion by using the contour of the end of the finished portion to which a low-pass filter has been applied as a reference contour (Pf), and a deviation is detected.

10. The method according to any one of claims 1 to 6, wherein the contour (P) of the end of the finished portion in the aforementioned image is compared with a reference contour (Pf) of the end of the finished portion by applying a high-pass filter to the contour of the end of the finished portion, and a deviation is detected.

11. The method according to any one of claims 1 to 10, wherein for each image, the contour (P) of the end of the finished portion and the reference contour (Pf) of the end of the finished portion are compared by comparing at least one of the area, amplitude, and / or gradient measurements with a threshold for at least some elevations in the finished portion inspection zone (ZI), and a deviation is detected if at least one of these measurements exceeds the threshold.

12. The method according to any one of claims 1 to 11, wherein for at least one image, the shape of the deviation detected for several heights in the finishing inspection zone (ZI) is selected as a detection criterion.

13. The method according to any one of claims 1 to 12, wherein the shape of the deviations detected for several heights in the finish inspection zone (ZI) is analyzed in order to distinguish overhang / overmatch (JHO) defects from knockout (JHK) and flange (JHF) defects.

14. The method according to any one of claims 1 to 13, wherein the observation angle range of the deviation is analyzed by considering that a knockout defect is detected when the observation angle range of the deviation falls below a certain maximum value, in order to distinguish the knockout (JHK) defect from the flange (JHF) defect.

15. The method according to any one of claims 1 to 14, wherein the observation angle range of the deviation is analyzed by considering that a flange defect is detected when the observation angle range of the deviation exceeds a certain minimum value, in order to distinguish the knockout (JHK) defect from the flange (JHF) defect.

16. An inspection apparatus configured to perform the method according to any one of claims 1 to 15 for detecting defects in the horizontal mold seam (JH) of the finished portion of the glass container.

Images