JPH10129630A - Device, method and system for discriminating application adhesive member during production of packaging container and record medium of program for execution of the discrimination - Google Patents

Abstract

PROBLEM TO BE SOLVED: To accurately know the application position of an adhesive member for production of packaging containers and accurately discriminate the application position of the adhesive member from the allowable value preset in advance. SOLUTION: The applied face of an adhesion member is taken in by an image-input means and changed to electric signals by an image-processing means to take in a memory. The gray lever data of multiple values are processed to calculate a detected value by use of the standard deviation, and it is discrimination whether it is adaptable for a preset allowable value or not. In this way, the position and the quantity of the adhesive member can be clearly discriminated by the discriminating device of the application condition of the adhesive member during the production of packaging containers, and the contrasted difference to the substrate color can be accurately detected. And hence the degree of freedom of a light source can be extremely increased.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a coating position and a coating amount of an adhesive member applied to various parts of a packaging container for making a box when applying the adhesive member to the packaging container during the box making. For judging whether or not an adhesive member is at a fixed position or a fixed amount of an adhesive member is used. The present invention relates to a determination system or a medium on which a program for performing determination of an adhesive member application state is recorded.

[0002]

2. Description of the Related Art Conventionally, in order to determine whether or not a coating position and a coating amount of a coated adhesive member for producing a packaging container are at a fixed position or whether a fixed amount of the adhesive member is used. In addition, an inspection by human eyes was performed.
Further, a discriminating device for discriminating articles, which is generally used frequently, that is, an image captured by an image input means is processed as an electric signal, and then the density data of the electric signal is binarized to discriminate the applied state of the adhesive member is used. It is also possible.

[0003]

In this case, in the inspection by human eyes, not only the possibility of oversight but also labor is required, and improvement in work efficiency is desired. Next, in a commonly used discriminating device, that is, a device for detecting the presence or absence of the applied state of the adhesive member by binarizing the grayscale data of the applied state of the adhesive member, in particular, the adhesive member must be translucent. There is no contrast with the background, and it cannot be accurately binarized from the grayscale data. That is, even in the grayscale data, image data can be extracted only in a state where the ground color of the packaging container body is partially dull, and it is difficult to accurately detect the presence or absence of the adhesive member even when binarization is performed.

In order to improve the detection accuracy of the adhesive member based on the grayscale data, it is conceivable to use illumination with extremely high brightness. However, since the difference in contrast with the background is extremely small, accurate binarization is required. Accordingly, it is difficult to accurately determine the presence or absence of the adhesive member, and handling of the light source, which is illumination, requires extreme care. Therefore, it is not possible to accurately grasp the adhesion state of the adhesive member for box making of the packaging container at the time of box making. A solution is needed.

[0005]

SUMMARY OF THE INVENTION The present invention has been made to solve the above-mentioned drawbacks, and provides an accurate grasp of the state of application of the adhesive member for box making of a packaging container in box production and a coating of the adhesive member from a preset allowable value. An image input means for capturing an adhesive member application surface of a packaging container in a box manufacturing process according to the first aspect of the present invention in order to accurately determine a loading position and a coating amount. Image processing means to be loaded into the memory instead of the memory, and the gradation data by the electric signal of the adhesive member applied state taken into the memory is extracted for each pixel within a certain range or for each pixel, and the arithmetic operation is performed, and then the applied state of the adhesive member is calculated. An adhesive member application state determination device during packaging of a packaging container, comprising: determination means for determining whether the multi-valued grayscale data of the adhesive member application state taken into the memory is standard deviation. Use Provided is a device for judging the applied state of an adhesive member during packaging of a packaging container, which determines a coated state of an adhesive member by calculating a detection value and judging whether or not a predetermined acceptable value is met. Can be solved.

Next, in the method for judging the adhesion state of an adhesive member on a packaging container during box making according to the second aspect of the present invention, the image of the adhesive member application surface is taken into an image and converted into an electric signal to obtain multi-valued grayscale data. None, a detection value is calculated using a standard deviation for the density data, and whether or not the detection value is compatible with a preset tolerance value indicating that the coating state of the adhesive member is appropriate is determined. The problem can also be solved by using a method of determining the applied state of the adhesive member of the packaging container during the box making for determining the applied state of the adhesive member.

According to a third aspect of the present invention, there is provided a multi-valued window in which the multi-valued gray-scale data of the adhesion state of the adhesive member taken in the memory is divided into windows which are divided in the memory. The detected value of the adhesive member application state in the entire memory or in an arbitrary range is calculated by calculating the detection value using the standard deviation for each of the obtained grayscale data, and the preset application state of the adhesive member is appropriate. The packaging container according to claim 1, further comprising a determination unit having a determination unit configured to determine a coating state of the adhesive member based on whether each of the multi-valued detection values matches an allowable value indicating that This is a system for judging the adhesion state of the adhesive member during box making, and can also be solved by using this system.

Next, in the method for judging the adhesion state of an adhesive member on a packaging container during box making according to the fourth aspect of the present invention, the surface on which the adhesive member is applied is taken in an image and converted into electric signals to obtain multi-valued grayscale data. A window in which the multi-valued gray data is divided into arbitrary ranges is provided, and the gray data in these windows is sequentially calculated using the standard deviation to calculate a detection value, and a detection value for each window is calculated. Adhesion of packaging containers during box making to determine the application state of the adhesive member based on whether or not the above-mentioned detection value matches the allowable value indicating that the application state of the adhesive member set in advance is appropriate. A member coating state determination method may be used.

[0009] Next, an image input means for taking in the image the adhesive member coated surface of the packaging container during the box making according to the fifth aspect of the present invention, and the end or bent edge portion of the packaging container during the box making into the image. Image input means to be captured, image processing means to capture the captured image into an electric signal instead of an electric signal, and multi-valued shading based on the state of the adhesive member applied to the memory and the electric signal at the end or bent edge of the packaging container An adhesive member application state determination apparatus during box making of a packaging container, comprising: determination means for extracting and calculating data to determine the application state and position of the adhesive member, wherein the determination means is loaded into a memory. The detection value is calculated using the standard deviation for the multi-valued grayscale data of the edge or bent edge of the packaging container, and the contact of the packaging container is determined based on whether or not it has a preset allowable value. Determines whether the adhesive member application position at the time of member application is in a preset state, and calculates the detection value using the standard deviation for the multi-valued grayscale data of the adhesive member application state taken into memory Then, by judging whether or not a predetermined allowable value is met, the applied state of the adhesive member is determined, and the applied state and position of the adhesive member are determined. The coating state determination system can also detect not only the coating state of the bonding member, but also the bonding member coating position when the bonding member is coated on the packaging container, and also detect the bonding member coating state, This can be solved by using this system.

[0010] Finally, the invention according to claim 6, that is, a medium recording a program for judging the adhesion state of the adhesive member of the packaging container during the box making, wherein the program is the adhesive member application state captured in the image The image data of the surface is output as multi-level grayscale data converted into an electric signal, the detection value is calculated using the standard deviation for the multi-level grayscale data, and the application state of the adhesive member set in advance is set. Using a medium in which a program for executing the determination of the adhesion state of the adhesive member of the packaging container during the box making to determine whether or not the above-mentioned detection value is compatible with the allowable value indicating the appropriateness is used. This can solve the problem.

[0011]

DESCRIPTION OF THE PREFERRED EMBODIMENTS FIG. 1 is a basic schematic block diagram of the present invention, and FIG. 2 is a third embodiment of the present invention. FIG. 3 is a diagram showing a basic embodiment of the present invention. First, a basic embodiment of the present invention will be described with reference to FIGS. 1, 2 and 3. First, in the manufacturing process of the packaging container 2,
An adhesive member 3 is applied to a predetermined portion for the bonding of 2.In this case, an image input device 4a, which is a device for inputting an image, such as a CCD camera, removes the applied portion of the adhesive member. Shoot.

An image input device 4b, which is also a device for inputting an image, such as a CCD camera, takes an image of an end portion or a bent edge portion of the packaging container. In this case, the image input of the necessary part is appropriately performed by the sensor 5 or the like as in the known means, and the determination as to whether or not the packaging container 2 is at the fixed position may be performed using the known means. .

Next, the input image data is sent to an image processing device 7 comprising image processing means, which converts the image data into electric signals to form gray-scale data, and then takes them into a memory. The grayscale data captured in this memory includes a search area 12 which is a range to be particularly detected.
Can be displayed on the display screen of the monitor 11. Multivalued grayscale data is extracted from the pixels in the search area 12 and / or the area to be calculated.

Here, the search area 12 is an area for judging whether or not the adhesion state of the adhesive member is appropriate. The area to be calculated is the density used for the calculation when calculating using the standard deviation from the density data. Refers to the data area. The grayscale data may be extracted for each pixel, or may be extracted by including a plurality of pixels within a certain range. Next, a detection value is calculated for the grayscale data in the search area 12 and / or the area to be calculated using the standard deviation. The formula for this standard deviation is given by
Can be calculated by the following formula.

[0015]

(Equation 1) Thereby, a detection value is calculated from the multi-valued grayscale data. The region to be calculated may be the search region, or may be a wider or narrower range. It is sufficient that the region to be calculated is the most appropriate in calculating the detection value using the standard deviation. Therefore, when the detection value is calculated using the standard deviation, if the contrast is significantly different from the average density value in the region to be calculated, a high numerical value is calculated, and the same value as the average density value in the region to be calculated is calculated. In the case of contrast, only very low values are calculated.

By using the multi-valued gray scale data, the difference in gray scale, that is, the difference in contrast is displayed by individual detection values. In contrast, packaging containers
The allowable values to be set in advance by preliminarily planning the application location of the adhesive member 3 and the amount of the adhesive member 3 required in the box making process 2 and replacing these required application locations with numerical values. Is determined. The allowable value is not necessarily limited to a fixed numerical value, but may be a numerical value group, or may be defined as a range between a fixed numerical value and a different numerical value. When setting the allowable value, input as a setting area 13 for a range in which determination corresponding to the search area 12 is desired,
A permissible value expressing a desired coating state of the adhesive member 3 in a numerical value is set in advance. Therefore, a determination is made between the detection value of the search area 12 calculated as described above and the allowable value.

Thus, when the detected value satisfies the allowable value, the adhesion state of the adhesive member 3 is good, and when the detected value does not satisfy the allowable value, the coating of the adhesive member 3 is performed. The applied state becomes poor, and the applied state can be determined. In particular, since this detection value is calculated from the multi-valued grayscale data of the region to be calculated, small differences such as those with large, medium, and small contrast differences can be expressed as numerical values. In addition, not only can the presence or absence of the adhesive member 3 be accurately determined, but also the difference in the amount can be determined.

At the same time, since the detection value is calculated using the standard deviation from the grayscale data in the region to be calculated, a constant detection value is calculated if the coating amount of the adhesive member 3 in the search region 12 is an appropriate amount. However, it is also possible to determine whether or not the required amount of the adhesive member 3 is in the coating position. That is, if a fixed amount of the adhesive member 3 is applied to the search area where the adhesive member is to be applied and the search area for determining the adhesive member application state is determined, a certain detection value is output. The coating shape of the adhesive member 3 is circular or elliptical as long as the adhesive member 3 is coated within the permissible range of the position of the adhesive member coating portion without being confused by the fine coating position and coating form. A fixed detection value is calculated regardless of whether it is applied to a shape such as a rod or the like. This makes it possible to detect a necessary amount of the adhesive member 3 required for producing the packaging container 2 in a box.

As described above, according to the first aspect of the present invention,
The adhesive member is applied by an image input device 4a which is an image input means for capturing the adhesive member coating portion 14 of the packaging container 2 in the box making into an image.
The photographed portion 3 is photographed, the captured image is replaced with an electric signal and stored in a memory, and the image data is converted into an electric signal by the image processing device 7 to obtain multi-level grayscale data. The multi-valued grayscale data is used to calculate a detection value using a standard deviation and to determine whether or not the detection value has a preset allowable value. It is possible to accurately determine whether or not the set amount of the adhesive member is applied in the set position by the device.

The invention according to claim 2 is a method for judging the adhesion state of the adhesive member during the box making, wherein the adhesive member application part 14 is taken into an image and converted into multi-valued grayscale data by the image processing device 7. After calculating a detection value using the standard deviation for the grayscale data, it is determined whether or not the detection value corresponds to a preset allowable value indicating that the coating state of the adhesive member is appropriate. It is also possible to accurately determine whether or not the set amount of the adhesive member 3 has been applied within the set position by using the adhesive member application state determination method.

Next, according to the third and fourth aspects of the present invention, as shown in the block diagram of FIG. 2, after passing through the image input means and the image processing means, the discriminating means applies the adhesive member applied to the memory. The multi-level gray data of the state is divided into windows 15 in the memory,
The detection value is calculated for each of the multi-valued grayscale data of the sequentially divided window 15 using the standard deviation, thereby calculating the detection value of the adhesive member application state in the entire memory or in an arbitrary range in the memory.

The window 15 is one section obtained by dividing the search area 12 into a plurality of sections. When the area to be searched covers a wide range, or when the detection value of each window 15 which is a section in the search area 12 is detected to detect a detailed adhesive member application state or the like, the search area 12 Is divided into the window 15 and the area to be operated and the window 15
After extracting the multi-valued grayscale data for each, the detected value of each window 15 is calculated for these grayscale data using the standard deviation. As a result, the detection value for each window 15 in the search area 12 is output, so that the window 15
Can be calculated, and the detailed adhesive member application state is clearly shown. Accordingly, in response to this, in the setting area 13 corresponding to the search area 12 composed of each window 15, the allowable value group is input in advance in the setting area 13 corresponding to the detection value group of each window 15. Detailed determination becomes possible.

In the case where the packaging container 2 has the adhesive member coating portion 14 on the surface on which the packaging container 2 is folded, for example, a wide range of density data including the overlapping portion can be calculated as the detection value. . This method uses the light and shade data as a so-called shadow part, which is different from the applied state of the adhesive member 3 and also different from the ground color part, for the end portion of the overlapping packaging container 2. A detection value different from the detection value indicating the presence / absence state of the image is output. Therefore,
A portion having a high density in the allowable value indicates a state of contrast different from that of the other portions, and it can be set in advance that the detection value is different from the presence or absence of the application state of the adhesive member 3, and thereby the packaging container 2 can be set. Inspection of the state of itself can be performed.

According to a fifth aspect of the present invention, the multivalued grayscale data of the adhesive member coated surface 14 of the packaging container 2 during the box production is used to calculate a detection value using a standard deviation, and in addition to the above, the detection value is calculated.
The image data of the end portion or the bent edge portion 16 is taken in, converted into multi-valued grayscale data, a detection value is calculated for the grayscale data using the standard deviation, and the position of the packaging container 2 at the time of applying the adhesive member is determined. It is to detect. Therefore, if the position of the packaging container 2 at the time of applying the adhesive member is at the set position, the adhesive member is applied at a predetermined position also at the adhesive member application location, and it is possible to determine whether or not the predetermined position. Becomes

Next, according to the sixth aspect of the present invention, there is provided a medium recording a program for judging an adhesion state of the adhesive member of the packaging container 2 during the box making, wherein the image of the adhesive member application section 14 is stored. A medium in which a program for determining whether or not the detected value corresponds to an allowable value by causing data to be output as multi-valued grayscale data converted to an electric signal and calculating a detected value using a standard deviation is recorded. It is. Therefore, a recording medium comprising a program for calculating a detection value using the standard deviation after converting the image data into multi-valued grayscale data and performing a comparison with the allowable value, which has the allowable value In such a case, the package can be determined to be appropriate. On the other hand, if the package is inappropriate, the packaging container is found to be defective. In the medium on which the program according to the present invention is recorded, even when the search area 12 is divided into windows 15 and sequentially divided, the detection values are calculated using the standard deviation for the multi-valued gray data for each window 15. It is possible to do.

In the present invention according to claims 1 to 6, the adhesive member 3 is an adhesive made of a synthetic resin. Refers to a member having a thermoplastic property or the like to be achieved. The adhesive member 3 made of a synthetic resin adhesive or glue or a member having thermoplastic characteristics is often transparent or translucent, but the color is not particularly limited. In particular, in the case of an adhesive member made of a transparent or translucent material, it is possible to accurately determine the application state of the adhesive member in such a state. By mixing the colored paint, the accuracy of the discrimination in the discrimination device based on the multi-value processing of the present invention is enhanced.

Next, a description will be given of a basic embodiment of the apparatus for judging the applied state of the adhesive member during the production of the packaging container 2 with reference to FIG. Reference numeral 1 denotes a device for producing a packaging container, in which a packaging container 2 being produced on a conveyor 1a is flowing.
The adhesive member 3 is applied to the packaging container 2 by the adhesive member applying nozzle 1b. 1c and 1d are folding arms, which are used to make boxes. In this case, after the adhesive member 3 has been applied, an area of the adhesive member application section 14 is photographed by the image input device 4a, for example, a CCD camera or the like. The range to be photographed is input as an image in a range wider than the region to be calculated and the search region 12, but is not limited thereto, and image data in the range of the region to be calculated and the search region 12 may be input. .

If necessary, the image input device 4b also photographs an end portion of the packaging container 2 or a whole area of the bent edge portion 16. As a result, the image data is taken in by the image input device 4a for the adhesive member application portion 14, and the image data is inputted by the image input device 4b for the end portion or the bent edge portion 16 of the packaging container 2. In this case, the image input devices 4a and 4b may be configured to perform image input by one image input device 4. In this image input, reference numeral 5 denotes a sensor, and reference numeral 6 denotes a light source serving as illumination for inputting image data. The image data thus photographed is sent to the image processing device 7 having image processing means.

This image processing means converts image data into electric signals in the range of the area to be calculated and the search area 12 or in a wider range and converts the image data into grayscale data and takes it into the upper memory. The processing is performed by the processing device 7. The image processing device 7 is connected to a discriminating device 8 having a discriminating means. That is, the determining means determines the applied state of the adhesive member 3 by extracting density data based on the electric signal of the applied state of the adhesive member taken into the memory for each pixel within a predetermined range or for each pixel, and calculating the extracted data. This is performed by the discriminating device 8. In this calculation, a detection value is calculated using the standard deviation according to Equation 1 with respect to the multivalued grayscale data of the adhesion state of the adhesive member taken in the memory, and is set in advance after calculating the detection value. The application state of the bonding member 3 is determined by determining whether or not the adhesive member 3 is compatible with the allowable value.
In addition, the input of the end portion of the packaging container 2 or the bent edge portion 16
The same applies to processing and determination.

Although the image processing device 7 and the discriminating device 8 of the adhesive member coating portion 14 and the end portion or the bent edge portion 16 of the packaging container 2 are shown as different devices in FIG. 3, the present invention is not limited to this. The image processing means and the determination means may be performed by a device. The discriminating device 8 is provided with a buzzer 9 and a light 10 when a defective product is found, and starts operation of a device for removing the defective product by discriminating the defective product. Further, a monitor 11 for monitoring the application state and the discrimination state of the adhesive member 3 is also provided. Although not shown, the discriminating device 8 naturally has an input section for inputting an allowable value.

FIG. 4 is a view showing a search area 12 for the packaging container. A predetermined range for detecting the adhesive member-coated portion 14 of the packaging container 2 is set as a search area 12a, and detection is performed for the bent edge portion 16. A certain range is defined as the search area 12b. In this case, when the position of the packaging container 2 is determined, the position is not limited to the bent edge portion 16 but may be an end portion or may include both.

FIG. 5 is a partially enlarged view of FIG.
FIG. 6 is a view showing a search area 12a of a second adhesive member application section 14. Accordingly, one detection value is calculated for the entire range, and it is possible to determine the presence / absence of the amount of the adhesive member 3 and its suitability. In addition, in the case of performing detection using the window 15 in the search area 12a, for example, the windows 15 in the search area 12a are extracted one by one or a necessary portion, and the image data is sequentially extracted for each window 15. After being converted into multi-valued gray data, the detected values are detected using the standard deviation. The window 15 may be for each pixel.

FIGS. 6 to 8 show an example of image data of the adhesive member application section 14 according to the first and second aspects of the present invention, and are diagrams showing image data of the search area 12a. FIG. 6 shows three types of image data, each having an elliptical adhesive member applying portion 17, an irregularly shaped adhesive member applying portion 18, and a star-shaped adhesive member applying portion 19. These have different shapes, but have the same amount and the same area to calculate the same detection value.

FIG. 7 also shows three types of image data, and the circular adhesive member coating sections 20 are provided at the upper right, center, and lower left of the search area 12, respectively. Even in this case, the same detection value is calculated because the coating position is different but provided in the search area and has a common area with a common amount.

FIG. 8 also shows three types of image data, each of which is a circular adhesive member applying section 20, an adhesive member applying section 21 surrounded by an outer frame and a circle, and a circular adhesive member applying section 22. It has. In this case, the circular adhesive member applying portion 20 and the outer frame and the adhesive member applying portion 21 surrounded by a circle have different shapes, but have the same detection value by having a common area with a common amount. calculate. In addition, regarding the circular adhesive member coating portion 22, the detection value is calculated based on the difference between the contrast inside the circular shape and the contrast between the outer frame and the circular portion. If the difference in contrast with the part 20 or the difference in contrast between the ground color and the adhesive member coating part 21 surrounded by the outer frame and the circle is constant, the same detection value is calculated.

In the present invention, a constant detection value is calculated from the difference in density between the so-called ground color and the portion to be detected. If the difference in density is constant, a constant detection value is obtained. As a result, even when there is a difference in the brightness of the light source or a difference in the way of illuminating the image input, the difference between the ground color and the density of the detected portion is constant, and an accurate detection value is calculated regardless of the difference in the brightness. I can do it.

FIG. 9 is a partially enlarged view of FIG.
3 is a diagram showing a search area 12b of a bent edge portion 16 with respect to FIG. In this case, by performing detection using the window 15 in the search area 12b, the detection value of each window 15 can be sequentially detected, and a specific detection value of the bent edge portion 16 can be calculated. That is, the shaded portion generated by overlapping the paper sheets of the packaging container 2 can be extracted as grayscale data, and a higher numerical value is detected in the window 15 of the overlapping portion than in the other portions. Therefore, it is possible to detect the state of the bending. On the contrary,
By appropriately inputting an allowable value in an appropriate state as the allowable value, it is possible to determine whether or not the adhesive member 3 is applied in a correct form. In addition, this allows the packaging container 2
It is of course possible to determine whether or not the folding back is correctly performed.

As described above, in this determination, when the adhesive member 3 is not applied to a necessary portion, the detected value is a value of the ground color and there is no contrast, and the detected value is different from the allowable value. When a numerical value is indicated, it is possible to determine that the adhesive member 3 is not present at the corresponding location. Also, an adhesive member
If the amount of 3 is less than a certain amount, the contrast with the ground color of the packaging container 2 is reduced, and a low numerical value is detected, so that it is possible to see the defective product without reaching the allowable value.
On the other hand, when the amount of the adhesive member 3 is larger than expected, the contrast is larger than when the appropriate amount of the adhesive member 3 is applied, and the detected value differs from the appropriate detection value. There is a difference between the values and it can be determined that an appropriate amount of the adhesive member has not been applied. This is possible because the present invention uses multi-valued grayscale data, and the detected value also has a small difference as a numerical value.

Further, when detecting the end portion or the bent edge portion 16 of the packaging container 2, the detection value obtained from the density data of each of the ground color of the packaging container 2 or the color of the adhesive member 3 applied thereto is used. In comparison, different detection values based on having different contrasts due to so-called shadow portions or the like are calculated, and by predicting such detection values in advance, it is also possible to determine the bending state or the like and the adhesion state of the adhesive member. In addition, when the adhesive member coating portion is provided at a specific position in correlation with the bending position, it is possible to determine the coating position relationship. That is, since different values are calculated based on the difference in the contrast of the detected value, it is possible to set different allowable values for the difference in the values.

FIG. 10 is a diagram showing an embodiment in which a plurality of windows 15 are arbitrarily set in order to determine the applied state of the adhesive member 3. In addition, regarding the detection of the bent edge portion 16 of the packaging container 2 of the present invention, it is also possible to detect only the bent edge portion 16 in the manufacturing process of the packaging container.
This is because the image data of the adhesive member coating portion is not taken in at all, and only the image data of the folded edge portion 16 of the packaging container 2 is converted into multi-valued grayscale data. This is because it is possible to calculate the detection value of whether or not it is bent, and by calculating the detection value of each window 15 using the window 15,
It is possible to determine a specific bent state. At the same time, the present invention similarly inputs the image data of the coated surface of the adhesive member, without inputting the image data, and converts the image data into multi-valued gray-scale data after inputting the image data as an image of the inspection object such as a packaging container, and detects the gray-scale data using the standard deviation formula The value is calculated and discriminated from a predetermined allowable value. This is because, for example, in a case where dirt or the like is attached to the inspection object, a value higher than a predetermined allowable value is calculated, and a defective product or the like can be detected.

[0041]

【The invention's effect】

First, the first effect of the present invention is that, at the time of box making, accurate determination of the coating position of the adhesive member for box making of a packaging container and accurate determination of the coating position of the adhesive member from an allowable value set in advance. Can be performed. In other words, the glue application state is read as grayscale data, and it is possible to determine whether or not glue has been applied to a predetermined location based on the difference in contrast. This is because the detection of a transparent or translucent adhesive member, which could not be discriminated by conventional binarization, enables detection of fine differences by calculating a detection value based on a standard deviation from multi-level grayscale data. It is.

2. As a second effect of the present invention, a detection value is derived from the contrast of the grayscale data. Regarding the light source at the time of capturing the image data, the lightness, the uniformity of the light source, and the light source as in the related art are considered. It is not necessary to have a constant direction of application, the allowable range of the light source is widened, and detection errors due to the condition of the light source are eliminated. In particular, there are many problems related to the light source in the detection based on the conventional binarization, and in particular, the brightness of the light source is indispensable for accurate detection, and attention must be paid to the application of the light source. However, the present invention dramatically increases the degree of freedom of the light source without affecting the accuracy of the determination.

3. As a third effect of the present invention, it is possible to accurately grasp the applied amount of the adhesive member, and to judge whether the amount of the adhesive member is too small, appropriate or too large based on a preset allowable value. is there. Again, in the conventional binarization, only the difference of whether or not the adhesive member is applied can be determined, whereas the determination that the appropriate amount of the adhesive member is applied is determined. It is very effective for proper box making. In particular, according to the first and second aspects of the present invention, if a predetermined amount of the adhesive member is applied within a predetermined range, the detection value and the allowable value match, so that the fine adhesive member is used. Irrespective of the state of application of the adhesive member, it is possible to determine that the necessary amount of the adhesive member is at the required position.

[Brief description of the drawings]

FIG. 1 is a basic schematic block diagram of the present invention.

FIG. 2 is a basic schematic block diagram of the present invention according to claims 3 and 4;

FIG. 3 is a diagram showing a basic embodiment of the present invention.

FIG. 4 is a diagram showing image data in a search area.

FIG. 5 is a partially enlarged view of FIG. 4 showing a search area of an adhesive member application portion of the packaging container.

FIG. 6 is a diagram showing an example of image data of an adhesive member-coated portion according to the first and second aspects of the present invention, showing image data of an adhesive member-coated portion for calculating a predetermined detection value;

FIG. 7 is a diagram showing an example of image data of an adhesive member-coated portion according to the first and second aspects of the present invention, showing image data of an adhesive member-coated portion for calculating a predetermined detection value;

FIG. 8 is a diagram showing an example of image data of an adhesive member-coated portion according to the first and second aspects of the present invention, showing image data of an adhesive member-coated portion for calculating a predetermined detection value;

9 is a partially enlarged view of FIG. 4 showing a search area of a bent edge portion of the packaging container.

FIG. 10 is a diagram showing an embodiment in which a plurality of windows are arbitrarily set;

[Explanation of symbols]

 DESCRIPTION OF REFERENCE NUMERALS 1 Packaging container manufacturing equipment 1a Conveyor 1b Adhesive member application nozzle 1c Folding arm 1d Folding arm 2 Packaging container 3 Adhesive member 4 Image input device 5 Sensor 6 Light source 7 Image processing device 8 Discriminator 9 Buzzer 10 Light DESCRIPTION OF SYMBOLS 11 Monitor 12 Search area 13 Setting area 14 Adhesive member coating part 15 Window 16 Folded edge part 17 Elliptical adhesive member coating part 18 Non-fixed-shaped adhesive member coating part 19 Star-shaped adhesive member coating part 20 yen Adhesive member application part 21 having a shape 21 Adhesive member application part surrounded by an outer frame and a circle 22 Adhesive member application part having a circular shape

Claims (6)

[Claims] 1. An image input means for taking an image of a surface of a packaging container on which an adhesive member is applied in a box, an image processing means for taking the captured image into a memory instead of an electric signal, and an adhesive member applied state taken into a memory. A method of extracting the grayscale data by the electric signal for each pixel within a predetermined range or for each pixel and calculating and judging the applied state of the adhesive member. A state discriminating device, wherein the discriminating means calculates a detection value using a standard deviation with respect to the multi-valued grayscale data of the adhesion state of the adhesive member taken into the memory, and conforms to a preset allowable value. An adhesive member applied state determination apparatus during packaging of a packaging container, wherein the applied state of the adhesive member is determined by determining whether or not the adhesive member is applied. 2. A method for judging the adhesion state of an adhesive member applied to a packaging container in a box making process, wherein the image of the adhesive member application surface is taken into an image and converted into an electric signal to obtain multi-valued grayscale data. A detection value is calculated for the data by using the standard deviation, and the coating value of the adhesive member is determined based on whether or not the detection value matches a preset tolerance value indicating that the coating state of the bonding member is appropriate. A method for judging the adhesion state of the adhesive member of the packaging container during the box making for judging the adhesion state. 3. An image input means for taking in an image the adhesive member application surface of a packaging container during box production, an image processing means for taking in the captured image into a memory instead of an electric signal, and applying the adhesive member taken in the memory. A method of extracting density data based on the electrical signal of the state for each pixel within a certain range or for each pixel, and calculating and judging the applied state of the adhesive member. An attachment state determination device, wherein the determination means divides the multi-valued grayscale data of the adhesion state of the adhesive member taken into the memory into windows divided in the memory, and sequentially multi-values the divided windows. The detection value of the adhesive member application state in the entire memory or in an arbitrary range is calculated by calculating the detection value using the standard deviation for each of the obtained grayscale data, and is set in advance. Determining means for determining a coating state of the adhesive member based on whether or not each of the multi-valued detection values matches an allowable value indicating that the coating state of the adhesive member is appropriate; An adhesive member application state determination system during packaging of a packaging container. 4. A method for judging the adhesion state of an adhesive member applied to a packaging container in a box making process, wherein the adhesive member applied surface is taken into an image and converted into an electric signal to obtain multi-valued grayscale data, and the multi-valued data taken into the image is obtained. A window in which the grayscale data is divided into arbitrary ranges is provided, and the grayscale data in these windows is sequentially calculated using the standard deviation to calculate a detection value. Calculate the detection value corresponding to each window indicating the adhesive member application state in the range, and determine whether the detection value matches a preset allowable value indicating that the adhesive member application state is appropriate. A method for determining the applied state of the adhesive member of the packaging container during the box making, wherein the applied state of the adhesive member is determined based on whether or not the adhesive member is applied. 5. An image input means for taking in an image the adhesive member-coated surface of a packaging container being made in a box, and an image input means for taking in an image the end or bent edge of a packaging container being made in a box. Image processing means that takes in the memory instead of electrical signals, and extracts and applies multi-valued grayscale data based on the applied state of the adhesive member taken into the memory and the electric signal of the end or bent edge of the packaging container, and then calculates and bonds An adhesive member application state determination device during packaging of a packaging container, comprising: a determination unit for determining a coating state and a position of a member, wherein the determination unit includes an end portion or a bent edge of the packaging container taken into the memory. The detection value is calculated using the standard deviation for the multi-valued grayscale data of the part, and the adhesive member application position at the time of applying the adhesive member of the packaging container according to whether or not it has a preset allowable value. Determines whether the state is set in advance or not, calculates the detection value using the standard deviation for the multi-valued grayscale data of the adhesion state of the adhesive member loaded in the memory, and conforms to the preset allowable value An adhesive member application state determination system during packaging of a packaging container that determines the application state of the adhesive member by determining whether or not to perform the application, and determines the application state and position of the adhesive member. 6. A medium in which a program for determining an adhesive member application state of a packaging container being manufactured is recorded, wherein the program converts image data of an adhesive member application surface captured in an image into an electric signal. The changed multi-valued grayscale data is output, a detection value is calculated using the standard deviation for the multi-valued grayscale data, and a tolerance indicating that the preset coating state of the adhesive member is appropriate is indicated. A medium in which a program for executing a determination of a state of application of an adhesive member of a packaging container during box making for determining whether or not the detected value matches the detected value is recorded.