KR20210002528A - Glass bottle inspection method and glass bottle manufacturing method - Google Patents

Abstract

The present invention provides an inspection method and a glass bottle manufacturing method for automatically determining the presence or absence of defects in a glass bottle having a piece on its surface from an image.
One aspect of the glass bottle inspection method is a glass bottle inspection method having a piece on the surface of the body portion, an image acquisition step (S12) of capturing a rotating glass bottle to acquire an image of the entire body portion, and a piece from the image. And a masking step (S18) of masking the engraving area including the derived pattern, and a determination step (S20) of determining the presence or absence of a defect except for the masked engraving area of the image.

Description

Glass bottle inspection method and glass bottle manufacturing method

The present invention relates to a method for inspecting a glass bottle having a piece on the surface of the body and a method for manufacturing the glass bottle.

Glass bottles with uneven surfaces are known. Sculpted glass bottles impress consumers with their originality and luxury.

Patent Document 1, for example, has been proposed as a method for inspecting glass bottles with irregularities on the surface. In the invention of Patent Document 1, an engraving surface due to irregularities and a smooth surface without irregularities are optically determined.

Japanese Patent Publication No. 58-216906

However, in the invention of Patent Literature 1, only the side with and without the irregularities was determined, and the defect (defect) of the glass bottle was not inspected. In order to optically inspect a vial with irregularities, it is difficult to determine whether it is a shadow due to irregularities or a shadow due to defects such as scratches or air bubbles. Even now, the presence or absence of defects such as wounds and air bubbles in uneven glass bottles relies solely on visual inspection.

Accordingly, the present invention provides an inspection method for automatically determining the presence or absence of a defect (defect) in a glass bottle having a piece on its surface from an image, and a method for manufacturing a glass bottle.

The present invention has been made to solve at least some of the above-described problems, and can be realized as the following aspects or application examples.

In addition, in the description below, "sculpting" is a design caused by irregularities on the surface of a glass bottle, and "pattern or pattern" is a contrast density due to "sculpture" appearing in an image obtained by imaging a glass bottle. It is a change of.

[1] One aspect of the method for inspecting a glass bottle according to the present invention is a method for inspecting a glass bottle having a piece on the surface of the body, and the entire circumference of the body portion is imaged (photographed) by photographing the rotating glass bottle. An image acquisition process of acquiring a finished image, a masking process of masking a engraving area including a pattern originating from the engraving in the image, and a determination process of determining the presence or absence of a defect except for the engraving area masked with respect to the image It characterized in that it comprises a.

According to an aspect of the glass bottle inspection method, it is possible to automatically determine the presence or absence of a defect in a glass bottle having a fragment on its surface to determine the presence or absence of a defect excluding the engraving area.

[2] In one aspect of the glass bottle inspection method, the defect (defect) includes at least a surface defect, and in the image acquisition process, the transmitted light transmitted through the body portion may be captured by a line sensor.

According to an aspect of the glass bottle inspection method, by imaging transmitted light with a line sensor, even defects in which shadows such as surface bubbles are difficult to appear can be automatically determined from the image.

[3] In one aspect of the glass bottle inspection method, pattern position detection is performed by performing a pattern search on the image acquired in the image acquisition process using a pattern registration image previously created based on the shape of the piece. A process may be further included, and the masking process may mask the engraving area including the pattern detected by the pattern position detection process.

According to an aspect of the glass bottle inspection method, a pattern is detected using a pattern registration image created based on the shape of the piece, so that even if the pattern is light, the position of the pattern can be stably detected.

[4] In one aspect of the glass bottle inspection method, the pattern position detection step may perform a pattern search for a predetermined height range in the image.

According to one aspect of the glass bottle inspection method, since the height at which the pattern appears in the image is almost constant, it is possible to reduce the load on the inspection apparatus by pattern searching for a predetermined height range.

[5] In one aspect of the method for manufacturing a glass bottle according to the present invention, a parison is molded from a gob in a rough mold, and the parison is formed in a finishing mold. It is characterized in that the glass bottle is molded into the glass bottle, and an aspect of the glass bottle inspection method is performed on the glass bottle to obtain a glass bottle determined to be free from defects.

According to an aspect of the glass bottle manufacturing method, even a glass bottle having a piece can automatically determine a defect, and thus a glass bottle without defects can be manufactured.

According to an aspect of the glass bottle inspection method according to the present invention, it is possible to automatically determine the presence or absence of a defect in a glass bottle having fragments on its surface from an image. According to an aspect of the method for manufacturing a glass bottle according to the present invention, even a glass bottle having fragments can be manufactured without defects.

1 is a side view schematically showing an inspection device.
2 is a plan view schematically showing an inspection device.
3 is a flowchart of the inspection method according to the present embodiment.
4 is an example of an image.
5 is a diagram for explaining an image processing, a detection process, and a masking process.
6 is a diagram for explaining image processing and masking steps.
7 is a diagram illustrating a determination process.

Hereinafter, preferred embodiments of the present invention will be described in detail with reference to the drawings. In addition, the embodiment described below does not unreasonably limit the content of the present invention described in the claims. In addition, not all of the configurations described below are essential components of the present invention.

The glass bottle inspection method according to the present embodiment is a glass bottle inspection method having a piece on the surface of the body portion, an acquisition step of capturing an image of the rotating glass bottle to acquire an image of the entire circumference of the body portion; and the image And a masking process of masking an engraving area including a pattern derived from the engraving, and a determination process of determining the presence or absence of a defect except for the engraving area masked with respect to the image.

In the glass bottle manufacturing method according to the present embodiment, a parison is molded from a gob with a rough mold, the parison is molded into a glass bottle through a finishing mold, and an aspect of the glass bottle inspection method for the glass bottle. To obtain a glass bottle determined to be free from the above defects (defects).

1. Inspection device

The inspection apparatus 1 of the glass bottle 10 will be described in detail with reference to FIGS. 1 and 2. 1 is a side view schematically showing an inspection apparatus 1 used in the inspection method according to the present embodiment, and FIG. 2 is a plan view schematically showing the inspection apparatus 1.

The inspection apparatus 1 shown in FIGS. 1 and 2 is an inspection apparatus 1 for a glass bottle 10 having a piece 15 on its surface. The inspection device 1 is built as part of a manufacturing line (not shown) for manufacturing the glass bottle 10, and after molding, the slowly cooled glass bottle 10 is conveyed to the inspection device 1 The glass bottle 10 after inspection is conveyed to the next process.

The inspection device 1 includes a light emitting unit 20 having a light emitting surface 22 for irradiating light to the glass bottle 10, and disposed opposite the light emitting unit 20 with the glass bottle 10 interposed therebetween. Includes a control unit 50 including an image pickup unit 40 and a determination unit 52 that determines the presence or absence of a defect based on the image 80 (Fig. 4) of the glass bottle 10 captured by the image pickup unit 40. do.

Here, as shown in FIG. 1, the glass bottle 10 is inspected in a vertically erected state, that is, a central axis 12 along a vertical direction. The vertical direction is the direction of gravity and the horizontal direction is the direction orthogonal to the vertical direction.

The inspection device 1 includes a mounting table 30 that supports the glass bottle 10 while rotating it around the central axis 12, and a side roller that rotates the glass bottle 10 while contacting the side surface of the glass bottle 10. It includes (32). In FIG. 1, the side roller 32 is shown as if it is between the glass bottle 10 and the image pickup unit 40. However, as a convenience for explaining the side roller 32, the side roller 32 is an image pickup unit ( It is arranged at a position that does not interfere with the imaging of the glass bottle 10 at 40).

The glass bottle 10 is transparent or translucent. The translucency is a degree of transparency in which defects of the body portion 13 of the glass bottle 10, for example, the surface air bubbles 18, can be determined by light from the light emitting portion 20 that has passed through the glass bottle 10. The glass bottle 10 is, for example, a large-inlet bottle having a head portion 11 and a body portion 13 and a bottom portion 14 having a cross-sectional shape. The cross-sectional shape of the glass bottle 10 may be polygonal. The glass bottle 10 has a piece 15 on its surface. The pieces 15 are irregularities formed on the surface of the glass bottle 10 and are formed by, for example, irregularities engraved on the surface of a mold during molding.

The side roller 32 contacts the body 13 and rotates the glass bottle 10 around the central axis 12. The central axis 12 is an imaginary line that becomes the rotation center axis in which the glass bottle 10 rotates. The side roller 32 controls the driving force of the motor 60 according to the command of the rotation control unit 62 to the belt 35 or the like. By passing through the glass bottle 10, the glass bottle 10 is rotated. The side roller 32 rotates the glass bottle 10 by a predetermined amount at a predetermined speed. The predetermined amount of rotation is an amount sufficient to image the entire circumference of the glass bottle 10. The predetermined amount of rotation is set to, for example, 1.5 rotations or more so that the entire detection object can be grasped with one image data. The rotation detection unit 54 may be a rotary encoder installed directly or indirectly on the motor 60. In response to the pulse output of the rotation detection unit 54, the imaging unit 40 captures images of the glass bottle 10 a predetermined number of times.

The light-emitting unit 20 is a light source that illuminates the glass bottle 10. The light-emitting unit 20 has a light-emitting surface 22 on the side of the glass bottle 10, and is a surface light source capable of illuminating the glass bottle 10 from the opposite side of the imaging unit 40. The light emitting portion 20 is set to a height capable of illuminating the entire glass bottle 10 to be inspected by the inspection apparatus 1. As shown in FIG. 2, the overall width W2 of the light emitting surface 22 is narrower than the overall width W1 of the glass bottle 10. By making the overall width W2 narrower than the overall width W1, it becomes possible to clearly image the shadow of the surface air bubbles 18. The total widths W1 and W2 are the total widths of the glass bottle 10 and the light emitting surface 22 when the inspection device 1 is viewed in plan.

As shown in Fig. 2, the light emitting surface 22 has, for example, a rectangular shape, and almost all of it emits light. The light emitting surface 22 faces the glass bottle 10 and the image pickup unit 40 in front, and is disposed so that the light transmitted through the glass bottle 10 reaches the image pickup unit 40.

As the light source of the light emitting portion 20, for example, a known light source such as LED or organic EL can be used. The light-emitting unit 20 is a diffused light, and when an LED is used, a uniform light may be irradiated to the glass bottle 10 by using a diffuser plate on the light-emitting surface 22. As the diffuser plate, a known one that diffuses light from a light source such as LED and emits it to the outside may be used. Since the light is diffused by the diffusion plate, when a plurality of light sources are used, the imbalance with the portion where the light source does not exist can be reduced.

The imaging unit 40 is disposed to face the light emitting unit 20 with the glass bottle 10 interposed therebetween. The imaging unit 40 is disposed to capture an image of the surface of the glass bottle 10 on an extension line of the central axis 12. The imaging unit 40 can image at least a portion of the glass bottle 10 to be inspected, and in this case, the entire vertical direction of the body portion 13 of the glass bottle 10 is arranged so that the entire vertical direction is within the field of view of the imaging unit 40 do.

The imaging unit 40 may capture an image including a detection body (eg, including the surface bubbles 18) by light from the light emitting unit 20 that has passed through the glass bottle 10. The imaging unit 40 may use, for example, a known line sensor camera. The image pickup unit 40 captures an image according to the rotation of the side roller 32 by the output of the rotation detection unit 54, so that even if the rotational speed changes for some reason, the image 80 is not affected.

The imaging unit 40 captures an image of the electric pole of the body 13 and transmits the data to the image processing unit 53 of the control unit 50.

The control unit 50 includes a determination unit 52 and an image processing unit 53. The control unit 50 includes a processor such as a CPU (Central Processing Unit) or a Graphics Processing Unit (GPU), a hard disk drive (HDD), a solid state drive (SSD), a read-only memory (ROM), a random access memory (RAM). It is composed of a storage device such as a memory device, an input device such as a keyboard, a mouse, and a touch pad, a display device such as a liquid crystal display and an organic EL (Electrocess) display, and a digital input/output board such as an I/O board. The control unit 50 executes a process of inspecting the glass bottle 10. The processing in which the inspection apparatus 1 intermittently conveys the glass bottle 10 at a predetermined speed is executed by a control unit different from the control unit 50, but may be configured to be executed by the control unit 50.

The determination unit 52 determines the presence or absence of a defect based on the image acquired from the imaging unit 40. The defect (defect) determined by the determination unit 52 is, for example, a surface defect. Surface defects are surface air bubbles 18, contamination, and foreign matter present on the inner or outer surface of the glass bottle 10. In addition to the surface defect, the determination unit 52 may determine, for example, a defect inside the glass bottle 10. It is preferable that the determination unit 52 judges the surface bubbles 18 having a length of 3.0 mm or more, a width of 1.0 mm or more, and a depth of 0.05 mm or more as a defect. Further, it is preferable that the determination unit 52 does not erroneously determine the pattern in the image derived from the piece 15 as a defect.

The control unit 50 outputs the result of the determination of the determination unit 52 to the outside for each glass bottle 10, so that, for example, the glass bottle 10 determined to be defective in the line after the discharge unit of the inspection device 1 Exclude. The specific processing in the control unit 50 will be described in "3. Inspection method" below.

2. Manufacturing method

A method of manufacturing the glass bottle 10 according to the present embodiment will be described. The glass bottle 10 first molds a parison from a gob with a rough mold. The parison is molded into a bottomed cylindrical shape by blowing compressed air into a hot gob placed in a rough mold. You may also use a plunger with compressed air. Next, the parison is transferred to the finishing mold, and compressed air is blown into the parison in the finishing mold to form the product glass bottle 10. The glass bottle 10 immediately after molding is high temperature, so it is slowly cooled by moving it to a slow cooling furnace. Perform the following inspection method for the glass bottle (10) from the slow cooling furnace. Then, by executing the following inspection method, the glass bottle 10 determined to be free from defects is obtained as a genuine product.

Thus, according to the manufacturing method of the glass bottle 10 according to the present embodiment, the defect can be automatically determined even in the glass bottle 10 having the pieces 15, so that the glass bottle 10 without defects can be manufactured. .

3. Inspection method

A method of inspecting the glass bottle 10 according to the present embodiment using the inspection apparatus 1 in FIGS. 1 and 2 will be described with reference to FIGS. 3 to 7. 3 is a flowchart of the inspection method according to the present embodiment, FIG. 4 is a diagram for explaining an image preprocessing (S14), a pattern position detection process (S16), and a masking process (S18), and FIG. 6 is an image preprocessing (S14). And a masking step (S18), and FIG. 7 is a diagram illustrating a determination step (S20).

As shown in Fig. 3, the inspection method according to the present embodiment is an inspection method of a glass bottle 10 having a piece 15 on the surface of the body 13, at least an image acquisition process (S12) and a masking process ( S18) and a determination process (S20) are included. The inspection method according to the present embodiment may further include a step (S10) of starting imaging before S12, may include an image preprocessing (S14) for performing a predetermined process on the image after S12, or You may further include a pattern position detection process (S16) later. Each process will be described in the following order with reference to FIGS. 1 and 2.

S10: The control unit 50 instructs the imaging unit 40 to start imaging. The imaging unit 40 captures the transmitted light transmitted through the body 13 of the glass bottle 10 rotating around the central axis 12 according to the command of the control unit 50 with a line sensor. At this time, the control unit 50 calculates the rotation angle of the glass bottle 10 based on the output from the rotation detection unit 54, and continuously performs 1.5 turns (for example, 360°×1.5 = 540°). Take an image. The defect shown in FIG. 1 is the surface air bubble 18, for example. By imaging the transmitted light with a line sensor, even defects such as the surface air bubbles 18 that are difficult to be shaded can be automatically determined from the image. The captured image data is transmitted from the imaging unit 40 to the control unit 50.

S12: The control unit 50 executes an image acquisition process (S12) of acquiring an image 80 (Fig. 4) in which the electric pole of the body 13 transmitted from the imaging unit 40 is captured. The image 80 is stored in a storage device (not shown) of the control unit 50. In the image 80, an image of at least 1.5 turns of the body 13 may be captured, and additionally, an image of 1.5 turns of the head 11 may be captured. Since the image 80 has more than 1.5 turns of the body part 13, the control unit 50 cuts off a plurality of inspection areas 82 to 84, 88, and 89 equivalent to one turn of the body part 13 It can be placed on the image 80 without it.

The image 80 shown in Fig. 4 shows a state in which the pattern 15a derived from the piece 15, the alignment line 16 extending in the vertical direction, and the surface air bubbles 18a are captured in a dark shadow. The alignment line 16 is a shadow generated by a step formed by a mold used when molding the glass bottle 10. In addition to the surface air bubbles 18a derived from the surface air bubbles 18, the shadows determined as defects in the image 80 may include shadows derived from internal air bubbles, white stone, foreign matter, and the like. In order to clearly distinguish these shadows from the pattern 15a or the alignment line 16 and determine them as defects, a plurality of rectangular inspection areas 82 to 84 in the portion where the body portion 13 is imaged in the image 80 88, 89) are installed to execute a preset inspection algorithm for each inspection area. In Fig. 4, each inspection area 82 to 84, 88, and 89 is indicated by a broken line.

S14: The image processing unit 53 executes image preprocessing (S14) on the pattern 15a in order to more reliably execute the pattern position detection process (S16). In the image preprocessing S14, for example, "vignetting processing" can be performed in order to perform a pattern search on the pattern 15a in an abstract shape. "Gradation processing" can be performed, for example, by an averaging filter, and the averaging filter is a two-dimensional output by replacing the pixel value of the target pixel with the average value of all pixel values within the filter size range. It is a filter.

Further, as the image preprocessing (S14), in addition to the "gradation process", for example, a "expansion process" for expanding the black color of the shadow may be employed.

S16: As shown in Fig. 5, the determination unit 52 executes a pattern position detection step (S16). Before executing the pattern position detection step S16, as shown in Fig. 5A, the operator prepares a pattern registration image 86 created in advance based on the appearance of the piece 15. The pattern registration image 86 may be created based on the external shape of the pattern 15a derived from the piece 15. The pattern registration image 86 is a frame slightly larger than the piece 15 and may be set to have the same size as the rectangular first inspection area 82. The pattern registration image 86 is stored in a storage device (not shown) of the control unit 50.

Next, the determination unit 52 executes a pattern position detection step (S16). As shown in Figs. 5B and 5C, the pattern position detection step S16 is a pattern registration image previously created based on the appearance of the piece 15 with respect to the image 80 acquired in the image acquisition step S12. The pattern is searched using (86) to detect the pattern 15a. In the pattern search, a detection object suitable for the pattern registration image 86 is searched in the image 80, and the detection object is converted into a pattern 15a when the pattern registration image 86 matches the outer shape of the pattern 15a to a certain extent. To detect. Since the pattern 15a is detected using the pattern registration image 86 created based on the outer shape of the piece 15, the position of the pattern 15a can be stably detected even if the shade of the pattern 15a is light.

The pattern position detection process S16 may perform a pattern search for a predetermined height range in the image 80. As shown in Fig. 1, the glass bottle 10 is on the mounting table 30, and the height at which the pattern 15a appears in the captured image 80 is almost constant. Therefore, as shown in Fig. 4, the horizontal range of the second height H2 is set to the lower limit with the first height H1 from the lower end of the image 80 as the pattern search area 85 (a rectangle surrounded by a dashed-dotted line). Area), and pattern search is performed within the pattern search area 85. In this way, pattern search for a predetermined height range can reduce the processing load of the inspection apparatus 1.

S18: The image processing unit 53 executes a masking process (S18). As shown in Figs. 5D and 5E, in the masking step S18, the engraving area 81 including the pattern 15a originating from the engraving 15 in the image 80 is removed by the mask 87. Masking. The engraving area 81 is an area including the entire pattern 15a. The engraving area 81 may be an area surrounded by the pattern registration image 86, or a range slightly narrower than the pattern registration image 86 and close to the pattern 15a as the engraving area 81. The mask 87 is the same as the engraving area 81. The mask 87 is created in advance as a set with the pattern registration image 86. The arrangement of the mask 87 and the pattern registration image 86 can also be set in advance. Thereby, when the pattern registration image 86 is arranged at an appropriate position with respect to the image 80 by pattern search, the engraving area 81 is set as the image 80 and at the same time is masked by the mask 87. Further, the image processing unit 53 arranges a plurality of inspection areas 82 to 84, 88 and 89 of a predetermined shape on the image 80 based on the position of the pattern 15a detected by the pattern search. By presetting the position of the pattern 15a and the position of each inspection area 82 to 84, 88, and 89, when the position of the pattern 15a is determined, a plurality of inspection areas 82 to 84, 88, and 89 ) The position can be automatically laid out on the image 80.

In addition to the above-described S14 to S18, other image processing may be employed as, for example, S14. For example, the image processing unit 53 thickens the pattern 15a by subjecting the image 80 acquired in the image acquisition step S12 of FIG. 6A to the expansion process a plurality of times as shown in (b). The image 80 is subjected to binarization processing as in (c), and the determination unit 52 executes the pattern position detection step (S16), and the engraving area 81 is applied to the detected pattern 15a. ) May be set, and the image processing unit 53 may perform the masking process (S18) as in (d). In this case, in the pattern position detection step S16, the pattern 15a can be detected by the center and area of the pattern 15a obtained by the binarization process.

S20: The determination unit 52 executes a determination process (S20). The determination step S20 determines the presence or absence of a defect (defect) except for the engraving area 81 masked for the image 80. Since the presence or absence of a defect is determined except for the engraving area 81, the presence or absence of a defect in the glass bottle 10 having the engraving 15 on the surface can be automatically determined from the image 80. The determination process S20 includes a first inspection area 82, a second inspection area 83, a third inspection area 84, a fourth inspection area 88, and a fifth inspection area 89 set in the image 80. For ), a predetermined inspection algorithm is executed to determine the presence or absence of a defect. Each inspection area is preset as, for example, a rectangular shape of a predetermined size.

The first inspection area 82 is an area surrounding the engraving area 81 and is narrower than the second inspection area 83. In the first inspection area 82, except for the area covered with the mask 87, the image processing unit 53 is an example of the surface bubbles 18 with thin (or thin) shadow lines as shown in Fig. 7(a). For example, if vertical and horizontal enhancement processing is performed and then binarization processing is performed, as shown in (b), the line of the shadow becomes a continuous body that does not break. After the binarization process, the determination unit 52 determines whether the detection object is defective or not based on the area and maximum length of the detection object, and if the detection object is not defective, it determines as "no defect" and the control unit 50 (10) is treated as a genuine product (good product) (S22). In addition, when the detection object is defective, it is determined as "defective" and the control unit 50 treats the glass bottle 10 as a defective product (S24).

The second inspection area 83 surrounds the first inspection area 82 and extends from the lower end of the image 80 to the lower end of the head 11. The second inspection area 83 is a portion excluding the first inspection area 82. The image processing unit 53 performs image processing similar to that of the first inspection area 82 on the second inspection area 83 to determine whether or not the determination unit 52 is a defect with respect to the detection object (S20). Since the second inspection area 83 is less likely to erroneously determine the pattern 15a as a defect, the determination unit 52 is a detection body of the second inspection area 83 with higher precision than the first inspection area 82. Can be determined.

The two third inspection areas 84 are disposed on the left and right outer sides of the second inspection area 83 and extend from the lower end of the image 80 to the lower end of the head 11. The third inspection area 84 is disposed at a portion in the image 80 where the alignment line 16 appears. The inspection algorithm in the third inspection area 84 needs to distinguish the detection object from the alignment line 16. For example, as shown in (c) and (d) of Figs. 7 (c) and (d), the distance between two lines in the horizontal direction within the divided frame by dividing a plurality of detection objects in the vertical direction is measured. When the distance D1 between the teeth is wider than a predetermined width and there are, for example, three or more detection objects in the upper and lower edges, the determination unit 52 determines it as the surface bubble 18, and the distance D2 between the teeth If is narrower than the predetermined width, it is determined as the alignment line 16. In addition, the image processing unit 53 removes the vertical shadow (fit line, 16) from the image in Fig. 7(e) and performs binarization processing as shown in (f), and then the determination unit 52 Determine whether it is a defect or not by the area of. In the image processing at this time, since the vertical shadow can be erased by emphasizing the change in luminance in the vertical direction, surface bubbles having a horizontally long portion with a change in luminance in the vertical direction remain. By applying the other two inspection algorithms together as described above, a more accurate inspection becomes possible without misrepresenting the alignment line 16 as a defect.

The fourth inspection area 88 is between the second inspection area 83 and the third inspection area 84 and extends from the lower end of the image 80 to the lower end of the head 11. The fourth inspection area 88 is a portion of the body 13 that faces the second inspection area 83. The image processing unit 53 and the determination unit 52 can perform the same processing as the second inspection area 83 on the fourth inspection area 88.

The fifth inspection area 89 is a portion corresponding to the head 11 extending above the second inspection area 83 to the fourth inspection area 88. The fifth inspection area 89 is divided into two and disposed on the image 80 except for the position where the alignment line 16 is imaged. This is because a disturbance shadow orthogonal to the alignment line 16 may be generated at a position sandwiched between the two fifth inspection areas 89. The image processing unit 53 and the determination unit 52 can perform the same processing as the third inspection area 84 on the fifth inspection area 89, for example.

The present invention is not limited to the above-described embodiments, and various modifications are possible, and includes substantially the same configurations as those described in the embodiments. Here, "same configuration" refers to a configuration having the same function, method, and result, or a configuration having the same purpose and effect. In addition, the present invention includes a configuration in which a non-essential part of the configuration described in the embodiment is substituted. Further, the present invention includes a configuration that exhibits the same effects as the configuration described in the embodiment or a configuration capable of achieving the same object. In addition, the present invention includes a configuration in which a known technique is added to the configuration described in the embodiment.

One… Inspection device, 10... Glass bottle, 11... Head, 12... Central axis, 13… Torso, 14... Bottom, 15… Piece, 15a... Pattern, 16... Surface air bubbles, 18a... Surface air bubbles, 19... Bend, 20... Light emitting part, 22... Light emitting surface, 30... Wit, 32... Side roller, 35... Belt, 40... Imaging unit, 52... Control unit, 52... Judgment section, 53... Image processing unit, 54... Rotation detection unit, 60... Motor, 62... Rotation control unit, 80... Video, 81... Engraving area, 82… First inspection area, 83... Second inspection area, 84... 3rd inspection area, 85... Pattern search area, 86... Pattern registration image, 87... Mask, 88... 4th inspection area, D1, D2... Distance between two lines, H1… 1st height, H2... 2nd height, W1, W2... Overall width

Claims (5)

This is a method of inspecting glass bottles with pieces on the surface of the body
An image acquisition process of photographing the rotating glass bottle to obtain an image of the entire circumference of the body portion;
A masking process of masking a engraving area including a pattern derived from the engraving in the image,
And a determination step of determining the presence or absence of a defect excluding the engraving area masked with respect to the image. The method of claim 1,
The defect comprises at least a surface defect,
In the image acquisition process, the glass bottle inspection method, characterized in that the transmitted light transmitted through the body is captured by a line sensor. The method according to claim 1 or 2,
Further comprising a pattern position detection step of detecting the pattern by performing a pattern search on the image acquired in the image acquisition step using a pattern registration image previously created based on the outer shape of the piece,
The masking process is a glass bottle inspection method, characterized in that the masking of the engraving area including the pattern detected by the pattern position detection process. The method of claim 3,
The pattern position detection process is a glass bottle inspection method, characterized in that the pattern search for a predetermined height range in the image. A parison is molded from a gob in a rough mold, the parison is molded into a glass bottle in a finishing mold, and the glass bottle according to any one of claims 1 to 4 for the glass bottle. A glass bottle manufacturing method, characterized in that, by performing an inspection method, a glass bottle determined to be free from the defect is obtained.

Images