JPH06229930A - Method and apparatus fro inspecting concrete block product - Google Patents

Abstract

PURPOSE:To enable accurate determination of acceptance or reject of a product by executing a processing on the basis of a binary image and by conducting comparison with predetermined reference values in respect to the length, the number and the width of a fissure. CONSTITUTION:An image obtained by picking up a surface image of a product by a camera 4 is binarized by an image processing means 3 and first the length of a fissure is judged. In this case, the area of a place where a value corresponding to the fissure in a binary image, e.g. '1', gathers is compared with a second reference value and a reference point of the place of which the area is equal to or larger than the reference value is thereby determined and then expanded at a prescribed ratio. By this expansion, the places which are apart from each other on the binary image although the fissure occurs actually are made continuous. Then, a distance between the reference points of these continuous places is determined and compared with a first reference value and thereby the length is judged. In the case when the length of the fissue is small, the number of fissures or the with thereof is judged.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an inspection method and an inspection apparatus for concrete block products capable of automatically inspecting cracks which are one of surface defects of concrete block products.

[0002]

2. Description of the Related Art Conventionally, concrete block products such as interlocking blocks used for laying a pavement on a road or used for construction have been developed from the viewpoint of aesthetics as shown in FIG.
As shown in FIG. 3, a decorative plate 21 having an appropriate pattern is joined to the upper surface of the block body 20. This concrete block product is manufactured, for example, as follows.

That is, a mold for obtaining a plurality of products at the same time is set on the lower iron plate, and concrete is poured into the mold while vibrating the lower iron plate. After that, a material (for example, a powder material) to be the decorative plate 21 is stacked on the concrete upper surface, an upper iron plate having a predetermined pattern is placed on the material, and the iron plate is solidified while applying a predetermined pressure. .

This concrete block product has various defects in the manufacturing process, especially the decorative board 2.
1 Cracks on the surface are likely to occur, and in order to prevent such products from leaking to the market, the inspector visually inspects at the time of product completion or at the shipping stage using inspection standards or limit samples. ing.

[0005]

However, in such an inspection method, the inspection standard is, for example, "length 3".
mm or more and width 0.2 mm or more "
In actual products, there are almost no simple cracks, and there are many cases where the product breaks in the middle or is narrowed to 0.2 mm or less in the middle. Therefore, it is extremely difficult for the inspector to accurately and promptly judge these mass-produced products. In addition, even if you create a limit sample, the probability that cracks of the same shape will occur is extremely low.
The same can be said for the above.

As described above, in the above-mentioned inspection method, which is mainly based on the visual inspection of the inspector, the inspecting method has its own limit, and the inspector may make a variation in the judgment or miss a defective product. There is a problem that it is likely to occur, it is difficult to accurately determine whether the product is pass or fail, and it takes time to pass or fail the determination and the inspection time is likely to be long.
Also, in visual inspection, it is extremely difficult to feed back the inspection results as numerical data to the manufacturing site,
There is a problem that it is difficult to take appropriate measures in the manufacturing process in order to prevent recurrence of defects.

In particular, in the interlocking block product having the aesthetic elements as described above, the cracks on the surface are apt to be a fatal defect, and the improvement of the inspection method is desired.

The present invention has been made in view of the above circumstances, and enables accurate acceptance / rejection of cracks and labor saving of inspection, and can reliably prevent recurrence of defective products in the manufacturing process. An object of the present invention is to provide an inspection method and an inspection device for block products.

[0009]

In order to achieve the above object, the invention of claim 1 binarizes a surface image of a concrete block product based on a predetermined threshold value to obtain a binary image. In the binarized image, a step of obtaining a reference point at a location where values corresponding to cracks are gathered and enlarging the location at a predetermined ratio, and a distance between the reference points of consecutive locations of the enlarged location And comparing the distance with a first reference value to determine pass / fail.

Further, in the invention of claim 2, the area of a place where the values corresponding to the cracks gather is obtained, and the area and the second area are obtained.
It is characterized in that the reference point is obtained after comparing with the reference value of. In addition to the invention of claim 1, the invention of claim 3 counts the number of places where the distance between consecutive reference points is a predetermined value or more, and compares the number with a third reference value to determine whether the result is acceptable or not. It is characterized by including a step of determining. Further, in addition to the invention of claim 1, the invention of claim 4 reduces the enlarged portion at a predetermined ratio and calculates the width of the binary image along a line connecting adjacent reference points of the reduced portion. , And comparing the width with a fourth reference value to determine pass / fail.

Still further, according to the invention of claim 5, the photographing means for photographing the surface image of the concrete block product and the image photographed by the photographing means are determined based on a predetermined threshold value.
Image processing means for performing binarization to obtain a binary image, processing a portion of the binary image where a value corresponding to a crack exists according to a predetermined method, a result processed by the image processing means, and It is characterized by further comprising: a determining unit that compares the set reference value with each other to determine whether the product is acceptable or not.

[0012]

According to the inspection method of the present invention, the "length" of the crack is first judged. In this case, in the binary image, a value corresponding to a crack, for example, the area (number of pixels) of a place where “1” is gathered is compared with a reference value (second reference value), and the area is equal to or larger than the reference value. A reference point (for example, a center point) of the place is obtained, and the place is enlarged at a predetermined ratio. Due to this enlargement, although cracks actually occur, the distant locations on the binary image are continuous. Then, the distance between the reference points at the consecutive points is obtained, and the distance is compared with the reference value (first reference value) to determine the “length”.

In the judgment of "length", when the distance between the reference points is less than the reference value, that is, when the "length" of the crack is short, the "number of cracks" is judged,
Determine the "width" of the crack. The number of cracks is
The number of continuous portions of the enlarged portion is counted, and this number is compared with a reference value (third reference value). Also,
The "width" of the crack is calculated by reducing the enlarged portion and returning it to the original position, connecting the reference points of the respective portions, for example, cutting along a straight line at a regular interval in the right angle direction, and calculating the cutting width. Then, the average value of each section is calculated, and this average value is compared with the reference value (fourth reference value).

Further, according to the inspection apparatus of the present invention, the image taken by the photographing means is binarized by the image processing means, and
A binary image corresponding to a crack is processed according to a predetermined method, and the judgment means compares the processing result with a reference value to judge whether the product is acceptable or not.

[0015]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS An embodiment of the present invention will be described in detail below with reference to the drawings. FIG. 1 shows a basic block diagram of an inspection apparatus for concrete block products according to the present invention. In the figure, the inspection device 1 is a personal computer 2 and an image processing device 3.
It has a camera 4, an image monitor 5, and the like. The personal computer 2 has a CPU (central processing unit), a disk drive device, a RAM, a ROM, etc. (not shown), and for example, control software 6 set in the disk drive device.
Various processing based on the image processing apparatus 3
To control.

The image processing apparatus 3 includes, for example, an image processing processor 7 for performing various image processings and calculations by a control signal from the personal computer 2, an input interface 8 for inputting a captured image of the camera 4, and the input. It has a memory 9 for storing an image input via the interface 8 and a binary image to be described later, a monitor interface 10 for displaying the data processed by the image processing processor 7 on the image monitor 5, and the like.

The camera 4 is, for example, a charge-coupled device (CC
D) has a fixed image element such as
The signal is converted into an electric signal of 2512 pixels of × 512, and this electric signal is output to the input interface 8. Further, the image monitor 5 is composed of a CRT or the like, and the image processing device 3
The various processing results etc. processed by are displayed. Note that the personal computer 2 or the image processing apparatus 3 can be connected to a printer 11 for printing the processing result, the inspection result, and the like, if necessary.

Next, the inspection method by the inspection device 1 will be described with reference to the flowchart of FIG. This flowchart is executed by the personal computer 2 and the image processing device 3 according to the program of the control software 6 set in the personal computer 2.

First, the camera 4 is set above the surface of the concrete block product, and an image of the entire surface of the block product is taken. This image is input as 250,000 pixels to the image processing apparatus 3 via the input interface 8 (S100), is stored in the memory 9, and is subjected to grayscale image processing (S101) by the image processing processor 7. . In this grayscale image processing, for example, processing such as averaging of the input image, shading correction, and contrast enhancement is performed. This process distinguishes from the innumerable irregular shadows on the product surface and black particles contained in the material, which are easily recognized as the same as cracks.

Then, each pixel of the image subjected to the grayscale image processing is binarized into "0" and "1" at a predetermined threshold value (S102) and stored in the memory 9 as a binary image. To do. Next, this binary image has the coordinates (0,
0) and the lower right corner as coordinates (511, 511), and scanning is performed sequentially from left to right from top to bottom, where there is a "1" (hereinafter, this location is referred to as "data location" for convenience of explanation. ) Is extracted and the area is calculated (S103). The calculation of the area is performed by counting the number of pixels at data points. After the area of the data portion is calculated, it is determined whether or not there is a predetermined reference value (second reference value) or more among the areas (S104).

That is, in this judgment, the binary image is scanned as described above, and if the area of the data portion where "1" exists or "1" is gathered is less than the reference value, that value is deleted (2
The value image is set to “0”), and only the data portion having the reference value or more is stored in the memory 9. Therefore, “NO” in step 104.
In the case of, that is, when there is no data portion equal to or greater than the reference value, all the binary images are “0”, and it is determined that there is no crack and “pass” is determined (S119).

On the other hand, if "YES" in the step 104, the reference point of the data portion having the area equal to or larger than the reference value, for example, its center point is calculated (S105). As a method of calculating the center point, for example, as shown in FIG. 3, a line that connects the middle points of the data points D1 of the respective scanning lines xi and yi by scanning the corresponding data point D1 in the X-axis and Y-axis directions. L1, L
2 is obtained, and the intersection of the lines L1 and L2 is set as the center point T1,
The coordinates (x1, y1) of the pixel where the center point T1 is located are stored in the memory 9. If each pixel is weighted, the center of gravity of the data location may be obtained instead of the center point.

When the center point is calculated, the data portion is enlarged in the direction of the solid arrow in FIG. 4 (S106). That is, the periphery of each pixel is enlarged by one pixel (eight pixels for one pixel). Since the insides of the data points overlap, the positions of the pixels that actually contribute to the expansion are mainly the pixels in the outer peripheral part of the data points. Therefore, for example, when there are data locations D2, D3, D4 as shown in FIG. 5, the blank portion K in the figure is enlarged and three data locations D2, D
3 and D4 will be continuous. The meaning of this expansion is
This is so that the cracked portion, for example, a partially constricted portion, will not be overlooked.

When the data portion is enlarged, it is judged whether or not there is a continuous data portion (S107), and in the case of "NO" in this judgment, that is, even if the data portion is enlarged, it is continuous with other data portions. If there is no portion to be processed, it is determined that there is no crack of a predetermined length and it is determined as “pass” (S119). If “YES” in the step 107, that is,
If there are continuous data points, the data points are stored in the memory 9 and the distance between the center points of the continuous data points is calculated (S108). This calculation is performed using the coordinates of both center points. In the case where three or more data points are continuous, for example, as shown in FIG. 5, the distance between the center points of the adjacent two shortest points, that is, the center point T2 (x2, y2) and the center point T3 (x3). , Y3) and between the center point T3 (x3, y3) and the center point T4 (x4, y
4) The distance between them will be calculated respectively.

When the distance between the center points is calculated,
Accumulate all distances between consecutive data points (S109)
Then, it is determined whether or not this cumulative value is equal to or greater than the reference value (first reference value) (S110). If "YES" in this determination, that is, if the cumulative value between the center points of consecutive data points is equal to or greater than the reference value, it is determined that the length of the crack is long,
It is determined as “fail” (S111). In addition, step 1
In 09 and 110, the distances between the center points are accumulated and compared with the reference value. However, for example, each section of the distances between the center points may be compared with the reference value to determine “pass / fail”. On the other hand, if “NO” in the step 110, the process moves to the step of determining the “number” of the next crack.

First, in step 107, the result is "YES", that is, the number of data locations stored in the memory 9, that is,
The number of consecutive points is counted (S112), and it is determined whether or not this counted value is equal to or greater than a reference value (third reference value) (S11).
3) Do. If the result of this determination is "YES", that is, if the number of consecutive data points is greater than or equal to the reference value, it is determined as "fail" because the length is short but the number is large (S11).
1) Do. If “NO” in step 113, that is, if the number is also less than the reference value, the process proceeds to the next step of determining the “width” of the crack.

In this step, the data portion enlarged in step 106 is first reduced in the direction of the dotted arrow in FIG. 4 (S114). Then, along the straight line connecting the central points of the data points, the line is cut at a constant interval in the direction perpendicular to the line (S115), and the cutting length of the cut part, that is, the cutting width is calculated (S116). To do. For example, in the case of the data location D2 shown in FIG. 5, this calculation is performed by enlarging the coordinates of the pixel at which both ends of the cutting line L3 are located, as shown in FIG.
Calculation is performed based on (x5, y5) and q (x6, y6). In the case of the next cutting line L4, the coordinate q (x7, y
7), based on q (x8, y8). After each cutting width is calculated, an average value of each cutting width in each section, that is, between adjacent data points (for example, between T2 and T3 and between T3 and T4 in the case of FIG. 5) is calculated (S117). This average is
It is calculated by dividing the total cutting width by the number of data.

When the average value of each section is calculated,
The average value of each section is compared with the reference value (fourth reference value), and it is determined whether or not there is a portion where the reference value or more continues (S118). That is, in this determination, if only one section is equal to or larger than the reference value, it is ignored, and a portion where the reference value or more is continuous is extracted. If "YES" in this determination, the length of the cracks is short and the number of cracks is small, but the width of the cracks is large and it is determined as "fail" (S11).
1) Then, in the case of "NO", the length, the number, and the width of the cracks are all less than the reference value, and it is determined as "pass" (S119). In this way, the continuous points are extracted in order to extract the cracked portion more accurately, but it is also possible to compare the average value of each section with the reference value to determine the “pass / fail” of the product. Of course, it is possible.

The series of flow charts is completed as described above, that is, the inspection of the product is completed. In this flow chart, up to step 107, it is judged whether or not there is a portion which is considered to be a crack, and up to step 110, there is no crack. This means that the “length”, the “number of cracks” up to step 113, and the “width” of cracks up to step 118 are determined. And step 1
Only in the case of “NO” in 04, 107 and 118, it is judged as “pass”.

As described above, according to the above-described embodiment, the predetermined arithmetic processing is performed on the basis of the binary image, and each item of "length", "number of cracks" and "width" of the crack is a predetermined standard. The "pass / fail" of the product is mechanically determined by comparison with the value. Therefore, the inspection standard can be quantified, and since there is no human action in the pass / fail judgment, variations in the inspection, oversight, etc. are eliminated, and the accuracy of the inspection and the reliability of the inspection can be improved. Further, since the judgment can be made mechanically by the personal computer 2 and the image processing device 3, the inspection can be automated, and the inspection time can be shortened. In addition, for example, the inspection result of each judgment item can be used to express the shape of cracks as numerical data, which makes it easy to feed back the numerical data to the manufacturing site and take appropriate measures to prevent recurrence of defective products. It will be possible.

In particular, in the above embodiment, the "threshold value"
Is set appropriately according to the illumination method, illuminance, brightness of the product surface, etc., and the binary image enlarged in the "width" judgment is enlarged in proportion to the binary image when judging the "length" of the crack. It has been experimentally confirmed that “cracking” of various forms on the surface of concrete block products can be accurately determined with an extremely high probability by performing appropriate processing such as reducing and averaging the cutting width. There is.

In the above embodiment, the inspection device 1
As an example, the image processing unit 3 is used to implement the image processing unit as dedicated hardware, but the present invention is not limited to this. For example, simple image processing functions such as binarization are provided as dedicated hardware, The main image processing may be performed by software on the personal computer 2. In addition, in the above-described example, the crack was inspected, the "number" was determined after the "length" was determined, and then the "width" was determined. For example, the "length" is an important item for determining the crack. "
You may judge the "pass / fail" of the product by only the judgment, you may judge the "width" of the product by judging the "width" after judging the "length", or you may judge the "number of pieces" after judging the "length". It is also possible to judge whether or not the product is “pass / fail”.

Further, it is needless to say that the flow charts in the above embodiments are examples, and various modifications can be made without departing from the gist of the present invention. Further, for example, the enlarging / reducing method in the flow charts, the Other appropriate methods can be adopted as the method of calculating the position of the point and the distance between the center points, the method of calculating the area, the method of calculating the cutting width, and the like. Furthermore, in the above examples, the "crack" of the surface defect was explained, but it can be applied to other defects such as "scratch" similar to the crack, and it is not limited to the interlocking block product, and any other concrete. It can also be applied to block products.

[0034]

As described in detail above, in the concrete block product inspection method and inspection apparatus of the present invention,
It is possible to achieve accurate acceptance / rejection determination and improve the reliability of the inspection, and it is possible to save labor in the inspection. Further, it is possible to appropriately feed back to the manufacturing site, and it is possible to reliably prevent recurrence of defective products in the manufacturing process.

[Brief description of drawings]

FIG. 1 is a basic configuration diagram showing an example of an inspection apparatus for concrete block products according to the present invention.

FIG. 2 is a flowchart showing an example of an inspection method using the same device.

FIG. 3 is an explanatory diagram of the same area calculation method.

FIG. 4 is an explanatory view of the enlarging and reducing method.

FIG. 5 is an explanatory diagram of the enlarged image.

FIG. 6 is an explanatory diagram of a method of calculating the cutting width.

FIG. 7 is a perspective view showing a concrete block product.

[Explanation of symbols]

 1 ... Inspection device 2 ... PC 3 ... Image processing device 4 ... Camera 5 ... Image monitor 7 ... Image processor 8 ... Input interface 9・ ・ ・ ・ Memory 10 ・ ・ ・ Monitor interface 20 ・ ・ ・ Block body 21 ・ ・ ・ Decorative plate

Claims (5)

[Claims] 1. A surface image of a concrete block product,
Binarizing based on a predetermined threshold value to obtain a binary image; determining a reference point of a portion of the binary image where values corresponding to cracks gather and enlarging the portion at a predetermined ratio. And a step of calculating a distance between reference points of consecutive points among the enlarged points and comparing the distance with a first reference value to judge pass / fail. Block product inspection method. 2. The concrete according to claim 1, wherein the area of a portion where the values corresponding to the cracks are gathered is obtained, the area is compared with a second reference value, and then the reference point is obtained. Block product inspection method. 3. A surface image of a concrete block product,
Binarizing based on a predetermined threshold value to obtain a binary image; determining a reference point of a portion of the binary image where values corresponding to cracks gather and enlarging the portion at a predetermined ratio. And a step of calculating the distance between the reference points of the continuous points of the enlarged points and comparing the distance with the first reference value to determine whether the result is acceptable or not. Counting the number of pieces and comparing the number with a third reference value to determine pass / fail;
An inspection method for a concrete block product, comprising: 4. A surface image of a concrete block product,
Binarizing based on a predetermined threshold value to obtain a binary image; determining a reference point of a portion of the binary image where values corresponding to cracks gather and enlarging the portion at a predetermined ratio. And a step of calculating the distance between the reference points of the continuous points among the enlarged points and comparing the distance with the first reference value to determine whether the result is acceptable, and reducing the enlarged point at a predetermined ratio. And a step of calculating the width of the binary image along a line connecting adjacent reference points at the reduced position and comparing the width with a fourth reference value to determine pass / fail. Inspection method for concrete block products characterized by 5. A photographing means for photographing a surface image of a concrete block product, the image photographed by the photographing means is binarized based on a predetermined threshold value to obtain a binary image, and the binary image of the binary image is obtained. home,
Image processing means for processing a portion having a value corresponding to a crack in accordance with a predetermined method, and a result of processing by the image processing means is compared with a preset reference value to determine whether the product is acceptable or not. An apparatus for inspecting a concrete block product, comprising: