JPH071935B2 - Package abnormality inspection device - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION [Industrial field of application] The present invention inspects abnormalities in a package, such as breakage of a heat-sealed portion, which is a seam between films, when a product is packaged with a heat-shrinkable film. Regarding the device.

[Conventional technology]

This type of film packaging is usually performed by welding two heat-shrinkable films, which are applied from above and below the product or from the left and right, to each other with a high-frequency sewing machine or the like. However, if the welding work is not performed properly, the heat seal portion may be broken. If such a heat-sealed portion is torn, the value of the product may be reduced. Therefore, conventionally, a person visually identifies in the final step of the production line to eliminate what is considered defective.

[Problems to be Solved by the Invention]

However, since the packaging film is generally transparent and extremely thin, it is difficult to accurately determine the defective product by the visual inspection as described above, and the defective product is shipped as it is. It was

The present invention has been made to eliminate the above-mentioned problems, and an object of the present invention is to provide a package abnormality inspection apparatus that inspects a package for abnormality.

[Means for Solving the Problems]

The package abnormality inspection of the present invention is, as shown in FIG. 1, an apparatus for inspecting a seal portion which is a film joint when a product is packaged with a heat-shrinkable transparent film for any abnormality such as breakage. And a camera (B) that captures an image of the product (A) to be inspected
And a master image data storage means (C) for pre-storing image data of a film-packaged product having no seal portion as master image data, and in order to eliminate the influence of the brightness of the design of the product (A),
Image difference calculation means for calculating the difference between the data of each pixel corresponding to the master image data read from the master image data storage means (C) and the image data of the product (A) obtained by the camera (B). (D) and a binarization process for binarizing the difference image data obtained by the image difference calculation means (D) in order to extract only the image data of the seal portion that was in the product (A). Means (E)
When the direction orthogonal to the extending direction of the seal part is defined as the y direction,
An abnormality detecting means (F) for detecting whether or not there is an abnormality such as breakage of the seal portion by checking the change of the y coordinate value of the pixel data of the seal portion obtained by the binarization processing means (E). It is characterized by having.

[Action]

Image difference calculation means for the image data of the inspection object A having the seal portion, which is captured by the camera B, and the master image data of the standard inspection object, which is registered in the master image data storage means C in advance. With D, the difference in brightness is calculated for each pixel of both data. By this difference calculation, the brightness data for the same pattern in both image data is canceled, so that the brightness data of each pixel mainly for the sticker portion is extracted, and this brightness data is next binarized by the binarization processing means E. It is binarized. The brightness data includes brightness data for a pattern that cannot be excluded by the image difference calculation and brightness data for the seal portion. However, since the brightness data for the seal portion is due to diffused reflection of the seal portion, a large brightness is obtained. To have. Therefore, by adopting an appropriate threshold value in the binarization process, the luminance data for the pattern that cannot be excluded by the image difference calculation is removed as noise, and thus the pixel corresponding to the seal part is " The data of 1 "is set, and the data of" 0 "is set for each pixel in the other area.

Next, the abnormality detecting means F checks whether or not the pixel data of "1" for the seal portion is linearly continuous.
It is detected whether there is any abnormality such as tear of the seal part.

Since it is inconvenient to separately prepare a film-wrapped product to obtain master image data, it is inconvenient to obtain a master image from a product to be inspected having a seal portion by the method described in claim 2. I'm getting the data.

〔Example〕

FIG. 2 shows an embodiment of the package abnormality inspection of the present invention.

Reference numeral 1 denotes a product to be inspected, and four products are packed in a film as one pack.

Reference numeral 2 is a stopper for stopping the product 1 at a predetermined inspection position on the line, and reference numeral 3 is a sensor for detecting whether or not the product 1 is actually present at the inspection position. 4 is an illumination lamp for illuminating the product 1 in the inspection position,
A high-frequency fluorescent lamp is used to eliminate the flicker associated with commercial frequencies. 5 is a camera using a CCD image sensor,
An image of one side of the product 1 in the inspection position is captured via the mirror 6. The mirror 6 is provided here because the camera 5 requires a certain distance to capture the entire image of the product 1, which makes installation of the camera 5 difficult. This is because the reflected image of the product 1 according to (3) is captured by the camera 5 having a line below, so that the inspection device can be saved in space.

Reference numeral 7 denotes a CPU (central processing unit) for centrally controlling this apparatus, and an image processing unit 7a which will be described in detail later is provided inside.
Have. 8 stores a control program executed by the CPU 7.
ROM (Read Only Memory). Reference numeral 9 is a signal input section, and the detection signals from the stopper 2 and the sensor 3 are taken into the CPU 7 via the signal input section 9.
Reference numeral 10 is an image memory for storing a video signal from the camera 5, and 11 is an image monitor display for monitoring an image pickup screen by the camera 5. Reference numeral 12 is a keyboard for inputting various setting conditions and the like. 13
Is a discharge device for discharging the product 1 in which an abnormality is detected in the package by the inspection of this device from the flow of the line, and 14 is a signal output unit of the CPU 7 for the discharge device 13.

Reference numeral 15 is a measurement result display for displaying the inspection result of this apparatus, and 16 is a floppy disk device for storing the inspection result as data.

Next, the operation of the apparatus having the above configuration will be described with reference to the flowchart of FIG.

First, in step S1, it is determined whether the master image creation mode is the defect inspection mode. At first, since master image creation needs to be performed, step S2 and the subsequent steps are performed. This master image is an image based on the product 1 that has been film-wrapped and does not have a heat-sealed portion in the inspection area. For this purpose, the film-wrapped product 1 that does not include heat-sealing in the inspection area is taken by the camera 5 In this embodiment, the master image is obtained by using the minimum value synthesizing method described below because it is inefficient to take an image.

First, in step S2, the image data 101 for the first product shown in FIG.
It is stored in the memory area I of the image memory 10. In step S3, the image data 102 of the next product is stored in the memory area II of the image memory 10. In step S4, the brightness data B ₁ and B ₂ are compared for each corresponding pixel in the memory areas I and II, and when there is a relation of B ₁ > B ₂ in the next step S5, the corresponding brightness data in the memory area I is satisfied. The pixel brightness data B ₁ is updated with B ₂ . Step
In S6, it is judged whether or not the image capturing is finished,
If it is to continue, the process returns to step S3, and similarly, the image data captured next is sequentially compared with the image data stored in the memory area I, and the image data 110 of the tenth product is captured, and the process ends. And By capturing the plurality of images, the minimum brightness data of the corresponding pixels of the image data 101 to 110 is stored in the brightness data B of each pixel in the memory area I.

By the way, the heat seal part Y shown in the image data 101 to 110 is diffusely reflected at that position and detected as a high brightness part. However, since the position of the heat seal part Y is different for each product, the brightness data described above is used. The luminance data for the heat-sealed portion Y is removed by comparing and updating the above, and as a result, the master image data for the film-wrapped product having no heat-sealed portion Y as shown by 100M in the memory area I is obtained. can get. This master image data is compared with the image data to be inspected as described below, but the maximum value known to the master image data ( MAX) Filtered.

This master image data is taken in by the CPU 7, but can be registered in the floppy disk device 16 in step S7 if necessary. This completes the master image creation 100M, after which the process proceeds to step S8 and subsequent steps, which are the processes performed by the image processing unit 7a.

In capturing the image data in step S8, the image data is captured twice for the same product 1 and the two image data are combined so that more faithful image data can be obtained.

In the next step S9, as shown in FIG. 5, the difference between the brightness data on the surface of each pixel between the image data 100 of the product 1 to be inspected and the master image 100M is obtained. By this difference calculation, the brightness data for the common pattern in both image data 100 and 100M is canceled, and only the brightness data for the heat seal portion Y is extracted.
Is obtained.

In the next step S10, the luminance data of the differential image data 150 is binarized to generate noise (differential image data 150
Image data 15 having only the heat-sealed portion Y as shown in FIG.
0'is obtained. The area Q ₁ shown here is the area to be scanned described below.

In the next step S11, as shown in FIG. 7, the image data 150 'is transferred in the horizontal direction (x) which is the extending direction of the heat seal portion Y.
Direction is sequentially scanned for each pixel row, pixel data (0 or 1) is accumulated for each pixel row, and the pixel row having the maximum accumulated value is determined to be the location where the heat seal portion Y exists. To be done. When the position of the heat seal portion Y is detected in this manner, an abnormality such as breakage of the heat seal portion Y is then checked in step S12 depending on whether or not the heat seal portion Y is linearly continuous. . Therefore, as shown in FIG. 8, the scanning area Q ₁ is narrowed to Q ₂ by including the heat seal portion Y in the middle, and in this area Q ₂ , the position change of the heat seal portion Y in the y direction is suppressed. Find out.

That is, as shown in FIG. 8, the data is sequentially read out for each pixel row in the y direction, and the y coordinate value when the pixel data of “1” is read for each pixel row is detected, FIG. 9 shows the first-order differential dy / dx as the position change of the y-coordinate value. A maximum Y ₁ ′ and a minimum Y ₂ ′ are generated in FIG. 9 corresponding to the beginning and end of the portion Y ₀ which is not linearly continuous in the heat seal portion Y of FIG. This maximum Y ₁ ′,
By knowing the minimum Y ₂ ′, the existence of the linearly discontinuous portion Y ₀ shown in FIG. 8 can be determined, and the portion Y ₀ is broken at the heat-sealed portion Y as shown in FIG. To detect as.

When a tear is detected, the judgment in step S13 results in
In the next step S14, by outputting a predetermined discharge command signal, the product 1 is discharged from the line by the discharging device 13 and the status of occurrence of defects is displayed on the measurement result display display 15. After that, the process returns to step S8, and the same inspection as described above is performed on the next product 1.

If the film-wrapped product 1 is soiled or double-wrapped, those parts appear in the difference image 150 as a result of the difference calculation in step S9. Can be detected.

[Effect of the Invention] As described above, according to the present invention, the seal portion that matches the product is detected based on the difference between each pixel of the master image data and the product image data, and the continuity of the seal portion Since the abnormality such as the breakage in the seal portion is detected by the inspection, any abnormality in the package such as the breakage of the seal portion or the distortion of the seal line can be accurately detected, and the labor can be saved.

[Brief description of drawings]

FIG. 1 is a diagram corresponding to the claims of the package abnormality inspection of the present invention, FIG. 2 is a system diagram showing an embodiment of the present invention, FIG. 3 is a flow chart showing the control operation of the apparatus shown in FIG.
FIG. 5 is a diagram for explaining the minimum value synthesizing method for obtaining the master image data, FIG. 5 is a diagram for explaining an image difference performed for extracting the heat seal portion, and FIG. 6 is a binary value. FIG. 7 is a diagram showing image data from which noise has been removed by the image processing, FIG. 7 is a diagram for explaining a method for detecting the position of the heat seal portion in the image data, and FIG. 8 is a heat seal in which the position is detected. Image data showing only a portion, FIG. 9 is a diagram showing a change of the heat seal portion from the image data shown in FIG.
FIG. 10 is a diagram showing the presence / absence of abnormality in the heat seal portion based on the change in the heat seal portion in FIG. 1 ... Product, 2 ... Stopper, 3 ... Sensor, 4 ... Illumination lamp, 5 ... Camera, 6 ... Mirror, 7 ... CPU, 7a ... Image processing unit, 8 ... ROM, 10 ... … Image memory, 11 …… Image monitor display, 12 …… Keyboard, 13 …… Ejection device, 15 …… Measurement result display, 16 …… Floppy disk device.

Claims (2)

[Claims] 1. An apparatus for inspecting a seal portion, which is a film joint, for abnormality such as breakage when a product is wrapped with a heat-shrinkable transparent film, and takes an image of the product to be inspected. A camera, a master image data storage means for pre-storing image data of a film-wrapped product without a seal portion as master image data, and the master image data storage for eliminating the influence of the brightness due to the design of the product. Image difference calculation means for obtaining the difference between the data of each corresponding pixel with respect to the master image data read out from the means and the image data of the product obtained by the camera, and only the image data of the seal part that matches the product. In order to extract, the binarization processing unit that binarizes the difference image data obtained by the image difference calculation unit, and the extension of the seal portion. When the direction perpendicular to the direction is the y direction,
An abnormality detecting unit for detecting whether or not there is an abnormality such as breakage of the seal portion by checking a change in y coordinate value of pixel data of the seal portion obtained by the binarization processing unit. Abnormality inspection device for packaging. 2. The package according to claim 1, wherein the master image data is obtained by selecting, from a plurality of image data to be inspected, data having a minimum luminance between corresponding pixels. Abnormality inspection device.