JP7671038B2 - Product Inspection Equipment - Google Patents

Description

The present invention relates to an object inspection device that can inspect the formation state of an object being inspected, for example, an object inspection device that can inspect the contour lines, ruled lines, cut lines, half-cut lines, embossed lines, etc. of unfolded paper packaging containers in the most suitable manner for each.

Conventionally, when inspecting the formation state of an object to be inspected, image data acquired from the object to be inspected is compared with inspection reference data, but when inspecting using such inspection reference data, methods using image data determined to be good by visual inspection, or methods using design data such as CAD data used at the time of design are used (e.g., Patent Document 1). Of these, the method using the design data used at the time of design has the advantage of being able to perform more accurate inspection than when using inspection reference data generated by visual inspection.

JP 2020-173129 A

However, when performing inspection using such design data, the following problems arise:

That is, because design data such as CAD data is composed of lines, it is effective for inspecting the position of lines and contour patterns formed on the surface of an object to be inspected, but it is not possible to inspect the state of parts according to the type of lines on the design data. Specifically, on the surface of paper packaging containers such as thick paper and cardboard, not only are there contour lines, which are the outline parts, but various lines are formed, such as groove-like lines embossed in advance on the folded part, continuous cut lines on the flap for opening, half-cut lines that cut the thick paper in half to make it easier to fold, and convex embossed lines that display the product logo, etc. However, because all of these lines are represented as simple lines in the design data, it is not possible to inspect the contour lines, ruled lines, cut lines, half-cut lines, and embossed lines by changing the program or inspection standard data for each of them.

In addition, when inspecting not only paper packaging containers but also other objects, there is a problem that accurate inspection cannot be performed using design data if there are characteristics in the various lines and patterns formed on the object to be inspected.

In order to solve the above problems, the present invention aims to provide an object inspection device that can inspect each part formed on an object when inspecting the forming state of the object using design data.

In other words, in order to solve the above-mentioned problems, the present invention comprises an image acquisition means for acquiring an image of an object to be inspected, a line type information extraction means for extracting line attributes such as line layer, thickness, solid line, dashed line, dotted line, double dotted line, and color in design data used when designing the object to be inspected, a storage means for storing inspection programs and parameters which differ depending on the line attributes, and an inspection means for reading out from the storage means the inspection programs and parameters corresponding to the line attributes extracted by the line type information extraction means, and comparing them with image data based on the image acquired by the image acquisition means for inspection.

In addition, in such an invention, the lines of the design data are divided into at least two or more types corresponding to contour lines, ruled lines, cut lines, half-cut lines, and embossed lines, and the attributes of the lines are assigned to the lines.

The present invention makes it possible to perform inspections using the most appropriate method for each area based on the attributes of the lines drawn for each area to be inspected.

FIG. 1 is an external view of an article inspection device according to an embodiment of the present invention. Functional block diagram of the same configuration FIG. 1 shows a product package that is an object to be inspected in the same manner. FIG. 1 shows design data in the same form. FIG. 2 shows a memory means in the embodiment. FIG. 13 is a flowchart showing the inspection in the same embodiment.

One embodiment of the present invention will be described below with reference to the drawings.

The product inspection device 1 in this embodiment is capable of inspecting the quality of the formed state of an inspection object, which is an object, using design data such as CAD. As shown in FIG. 2, it is configured to include an image acquisition means 2 that acquires a surface image of the inspection object, and an inspection means 4 that compares image data based on the image acquired by the image acquisition means 2 with the design data to inspect the quality. Characteristically, when the design data is created, lines with different attributes are created for each part to be inspected, and during inspection, as shown in FIG. 4, the inspection method can be changed according to the attribute of the line. The present embodiment will be described in detail below. Note that in this embodiment, an object to be inspected is described using an object such as a product package that is unfolded in a flat surface as an example, and the contour lines, ruled lines, cut lines, half-cut lines, embossed lines, patterns, colors, etc. formed on the surface of the product package are inspected.

First, the image acquisition means 2 for acquiring a surface image of the inspection object uses a camera 21 to acquire a surface image of the inspection object as shown in FIG. 1, and acquires an image of the inspection target area formed on the surface by receiving reflected light from the light irradiated from the lighting device 22. Depending on the inspection target area, it may be better to change the light irradiation angle or color for inspection, so the light irradiation angle and color are made changeable. The image acquired in this way is then stored as image data in the storage means 3.

This storage means 3 stores not only image data acquired from the object to be inspected, but also different inspection programs and parameters according to the attributes of the lines, as shown in FIG. 5, so that the acquired image data of the object to be inspected can be inspected using the inspection program and parameters corresponding to the attributes of the lines in the design data. At this time, for example, parameters according to the attributes are stored, such as the "thickening width" of the line used when generating the inspection area according to the attributes of the line, the "search distance" for inspecting whether or not there is a pixel corresponding to each pixel of the master image according to the attributes of the line, the "threshold value" of the brightness that is to be considered as a good product, the "expansion amount" of the defective area, and the "threshold value" of the area or length that is to be determined as defective.

The inspection means 4 is executed by an inspection program, and as shown in FIG. 5, inspection is performed using inspection programs and parameters stored corresponding to the attributes of the lines. These inspection programs may each have a different inspection algorithm, or the program itself may be common and inspection may be performed using different inspection methods using parameters. When performing such an inspection, the contour line of a previously acquired image and the contour line of the design data are aligned using pattern matching or reference marks, and in this state, the contour line, ruled line, cut line, half-cut line, embossed line, pattern, and color are inspected. An example of this inspection method is described below.

First, when performing inspection, attribute information such as line type, color, thickness, etc. is extracted from the design data. At this time, if multiple types of lines are used, attribute information for all lines is extracted.

Then, for the extracted line, parameters are read from the storage means 3 based on the attribute information of the line, and the line is thickened to set the inspection area. At this time, for example, the "thickening area" is made small for contour lines, and the "thickening area" is made large for parts with irregularities such as embossed lines.

When performing an inspection by setting an inspection area in this way, the color of the master image and the color of the object being inspected (RGB brightness) are compared for each pixel, and pixels that do not belong to the "threshold" that corresponds to the line attribute are extracted.

Then, the extracted pixels (bad pixels) are expanded by a preset "amount of expansion of the bad area" to prevent them from being judged as good pixels due to discontinuities such as blurring.

Then, a labeling process is performed on the defective pixels extracted in this way, and areas that are clusters of defective pixels are extracted, and it is determined whether or not they correspond to a defective area using "threshold values" of area and length that are registered in advance in correspondence with the line attributes.

Then, these processes are performed by shifting the image forward, backward, left, and right within the range of the search amount registered corresponding to the line attribute, and if it is OK, the area is deemed OK and a judgment process is performed. Conversely, if it is not good, the area is deemed not good and a judgment process is performed.

The design data used during this inspection is designed with different line attributes for each part to be inspected. For example, as shown in Figure 4, contour lines are recorded on the first layer of CAD, ruled lines on the second layer, cut lines and half-cut lines with different line types and thicknesses on the third layer, and embossed lines and patterns on the fourth layer as contours.

The line type information extraction means 5 is capable of extracting the attributes of lines extracted from the design data, such as the layer corresponding to the part to be inspected, the line thickness, type (solid line, dashed line, dashed line with one dot, dashed line with two dots, etc.), and color.

The output means 6 also outputs the inspection results. If the product is found to be good, it outputs a message to that effect, and if the product is defective, it outputs not only the result that the product is "defective" but also information about the defective area, etc.

The process of inspection using the thus configured product inspection device 1 will be described with reference to the flowchart in Figure 6. Note that, before carrying out the inspection, it is assumed that design data is generated by changing the line attributes for each contour line, ruled line, cut line, half-cut line, embossed line, pattern, and color.

When inspecting an object to be inspected, the object to be inspected is positioned in front of the camera 21, and its surface image is acquired by the image acquisition means 2 (step S1). At this time, if it is better to inspect each part to be inspected with different light irradiation angles or with different colors, light is irradiated with light at multiple irradiation angles and colors, and the images are acquired and stored in the storage means 3.

Next, after acquiring the image in this manner, the design data is read from the storage means 3 (step S2), and from that design data, the line attributes of the design data corresponding to the part to be inspected are extracted using the line type information extraction means 5 (step S3). Then, the inspection program and parameters corresponding to the extracted line attributes are read from the storage means 3 (step S4).

Then, using the inspection program and parameters thus read out, it is judged whether the inspection target part corresponding to the line is proper or defective (step S5). Specifically, for the extracted line, an inspection area is set for thickening the line and inspecting it based on the parameters of the set attribute information. Then, the thickened inspection area is inspected, and at this time, the color of the master image that serves as the reference for inspection and the color (RGB brightness) of the acquired inspection target are compared for each pixel, and pixels that do not belong to the "threshold value" stored corresponding to the attribute of the line are extracted. Then, the extracted pixel (defective pixel) is expanded by the "expansion amount of the defective area" previously set as the attribute of the line, so that it is not judged as a good pixel due to discontinuities such as blurring. Then, a labeling process is performed on the pixel to extract the defective area, and the defective area is judged by the threshold value of the area and length registered in advance. Then, these processes are performed by shifting the image back and forth and left and right within the range of the search amount registered corresponding to the attribute of the line, and if it is OK, the area is treated as OK and processed.

If the inspection of all the target parts is judged to be proper, it is judged to be a "good product" and this is output to the output means 6 (step S7). On the other hand, if it is judged to be defective, a message indicating that it is "defective" is output together with the defective parts, so that visual inspection, etc. can be performed (step S6).

As described above, according to the embodiment, there is provided an image acquisition means 2 for acquiring an image of the object to be inspected, a line type information extraction means 5 for extracting line attributes (layer, thickness, solid line, dashed line, dotted line, double dotted line, color ) in the design data used when designing the object to be inspected, a storage means 3 for storing different inspection programs and parameters according to the line attributes, and an inspection means 4 for reading out the inspection program and parameters corresponding to the line attributes extracted by the line type information extraction means 5 from the storage means 3 and comparing them with image data based on the image acquired by the image acquisition means 2 for inspection. Therefore, it becomes possible to inspect the object to be inspected in the most appropriate manner according to the part, based on the attributes of the lines drawn for each part.

The present invention is not limited to the above embodiment and can be implemented in various ways.

For example, the above embodiment has been described with reference to an example of inspecting the formation state of product packaging, but the present invention can also be applied to inspecting the formation state of other flat objects.

In the above embodiment, the lines of the design data are assumed to be straight lines and curves, but they may also be rectangular, circular, or polygonal. In addition, the attributes have been described using layer, line thickness, line type, and color as examples, but other attributes may also be straight lines, curves, single points, rectangles, circles, and polygonal shapes. In addition, if the design data includes auxiliary lines such as center lines, the line types may be excluded from the inspection target.

In addition, in the above embodiment, the parameters are changed to perform the inspection, but it is also possible to provide separate inspection programs for each line and perform the inspection using an inspection program that corresponds to the attributes of the line.

Reference Signs List 1: Article inspection device 2: Image acquisition means 21: Camera 22: Lighting device 3: Storage means 4: Inspection means 5: Line type information extraction means 6: Output means

Claims (4)

An image acquisition means for acquiring an image of an object to be inspected;
a line type information extraction means for extracting attributes of lines, such as layer, thickness, solid line, dashed line, one-dot chain line, two-dot chain line, and color, from design data used in designing the inspection object;
A storage means for storing different inspection programs and parameters according to the attributes of the line;
an inspection means for reading out from said storage means an inspection program and parameters corresponding to the line attributes extracted by said line type information extraction means, and comparing said program and parameters with image data based on the image acquired by said image acquisition means for inspection;
An article inspection device comprising: The object inspection device according to claim 1, wherein the lines of the design data are divided into at least two or more types corresponding to contour lines, ruled lines, cut lines, half-cut lines, and embossed lines, and are provided with attributes of the lines. acquiring an image of an inspection object;
a step of storing different inspection programs and parameters according to attributes of lines in design data used when designing the inspection object, such as layer, thickness, solid line, dashed line, one-dot chain line, two-dot chain line, and color , reading out the inspection program and parameters corresponding to the stored line attributes, and comparing them with the acquired image of the inspection object to inspect the line;
An article inspection method comprising: The object inspection method according to claim 3, wherein the lines of the design data are divided into at least two of contour lines, ruled lines, cut lines, half-cut lines, and embossed lines, and are provided with attributes of the lines.

Images