JP3871167B2 - Method for distinguishing workpiece types - Google Patents

Description

[0001]
BACKGROUND OF THE INVENTION
The present invention relates to a method for discriminating types of cylindrical workpieces having the same design, such as aluminum wheels, and having different sizes and offset amounts.
[0002]
[Prior art]
There are various materials and structures for automobile road wheels, but the ratio of mounting aluminum wheels is increasing for the purpose of reducing the weight of automobiles and improving the appearance and design. This aluminum wheel is manufactured through steps such as melting, casting or forging, heat treatment, machining, painting, inspection, and packing of an Al—Si—Mg alloy. In this manufacturing process, for example, painted aluminum wheels are discriminated from each other, sorted into several groups based on the results, and then transported to the next process.
[0003]
In this aluminum wheel, a plurality of design holes are formed on the design surface of the disk portion in response to a request for appearance and the like, and the shape and arrangement of the design holes differ depending on the type of the aluminum wheel. Therefore, usually, the shape of the aluminum wheel that has been photographed is determined by photographing the shape of the design surface of the aluminum wheel and comparing the obtained image with a reference image. For example, in the conventional image processing method, as described in JP-A-8-44871, a plurality of detection lines (concentric circles with the wheel center) set on the element shape (design hole) on the surface of the aluminum wheel; The feature line of the element shape is determined from the position of the feature point that is the intersection with the element shape, and based on the relationship between the reference point of the element shape (the center of gravity of the design hole) and the reference line passing through the reference point (center) of the wheel The feature amount of the element shape (inclination and length of the design hole) was calculated and compared with the reference value.
[0004]
[Problems to be solved by the invention]
However, the processed aluminum wheel has a through hole (hereinafter referred to as a mounting hole) for mounting the wheel provided on the disk portion to the wheel even if the size (outer diameter, width, offset) and design hole shape and arrangement are the same. There are different numbers and arrangements. In addition, even if these parameters are the same, the surface processing / painting state of the design part may be different. Therefore, in the conventional image processing method, it is extremely difficult to efficiently discriminate the types of aluminum wheels that differ only in the number and arrangement of mounting holes or the surface state of the design portion.
[0005]
Accordingly, an object of the present invention is to provide a method capable of efficiently and accurately discriminating types of cylindrical workpieces having similar surface shapes or dimensions.
[0006]
[Means for Solving the Problems]
In order to achieve the above object, in the present invention, the size of the bottomed cylindrical workpiece and the number of holes having a plurality of shape patterns separated on the end face and a plurality of holes arranged concentrically on the inner circumference thereof. Based on the measurement results, the group to which the workpiece belongs is selected, the end face of the workpiece is photographed, the image signal is converted into multi-values, and then the polar coordinates are converted, and the image is within the selected group. A workpiece having the same size and design is selected by selecting an image having the most correlation with the registered image, and a predetermined annular region including a part of the shape pattern on the end surface is selected for the workpiece. By calculating the average luminance of the area having the luminance level and comparing the value with the reference value, the workpiece having the same surface condition of the end face is selected, and the hole arrangement is determined for the workpiece. By comparing with a reference value, to determine the varieties of the work, employing the technical means of.
[0007]
DETAILED DESCRIPTION OF THE INVENTION
Details of the present invention will be described below with reference to the drawings. FIG. 1 is a cross-sectional view (a) and a front view (b) of an aluminum wheel that is an object of the present invention. The aluminum wheel 100 has a hub portion 101 to which an axle is mounted, a disc portion 102 formed around the hub portion, and a rim portion 103 formed on the outer peripheral portion thereof. A plurality of design holes 104 are formed in the disk portion 102 at equal intervals along the circumferential direction, and a plurality of mounting holes 106 are formed on the inner peripheral side, and 105 is a design surface. In the figure, D, W, and S indicate the diameter, width, and offset amount of the aluminum wheel. The offset amount S indicates the distance from the position W / 2 on the inner diameter side of the aluminum wheel to the hub portion 101. Aluminum wheels for automobiles may have the same design, for example, D may be 12 ″ to 18 ″ (7 sizes) and W may be 3.5 to 10.5J (15 sizes). There are some with different offset amounts, and moreover, 300 kinds of products may be mixed in the production line considering the difference in the surface condition caused by the number of mounting holes, their arrangement or processing or painting. Thus, in order to quickly and accurately discriminate among the various types of aluminum wheels, the present invention uses the apparatus shown in FIGS.
FIG. 2 is a plan view showing the configuration of the type discriminating apparatus, and FIG. 3 is a cross-sectional view taken along line AA of FIG. This type discrimination device is composed of a size detection unit 1, an offset detection unit 2, and an image processing unit 3. The configuration of each unit is as follows.
[0008]
The size detection unit 1 includes a stopper 10 for stopping an aluminum wheel (hereinafter referred to as a workpiece) conveyed in the direction of the arrow by the conveyor 4 at a predetermined position, and a cylinder with an encoder disposed above the workpiece to detect the width of the workpiece. 11 and a phototube 12 composed of a light-emitting tube 12a and a light-receiving tube 12b disposed on both sides of the workpiece.
The offset detection unit 2 includes a stopper 20 that stops a workpiece at a predetermined position, a clamp member 22 having a pair of hands 21 that handle the workpiece, and a sensor that is installed on the lower surface of the workpiece and detects displacement. 23.
The image processing unit 3 includes an illumination device 30 that irradiates a work with light, a camera 31 that captures the workpiece, an image processing device 32 that inputs a captured image and performs product type discrimination, and a monitor 33 that displays the image. Have
[0009]
FIG. 4 is a functional block diagram illustrating the configuration of the image processing apparatus. In the drawing, an analog signal output from an image capturing unit (CCD camera) 31 is converted into a digital signal by an A / D conversion unit 320. Next, digitized image data is stored in the read image data storage unit 321. An image processing unit 322 performs processing of the image data, and three types of image processing are performed.
In the image processing unit 322a, 1) 2) the number of mounting holes and 3) the arrangement thereof are calculated by binarization processing of image data.
In the image processing unit 322b, multi-value processing of image data, polar coordinate conversion, and phase correction are performed.
1) Multi-value conversion processing is processing for converting each pixel of image data into 256 gradation luminances and position information (X, Y coordinates). 2) Polar coordinate conversion is image data position coordinates (X, Y). ) Is converted to (r, θ) using polar coordinate transformation. Further, 3) phase correction is a process of matching the work phase to obtain an accurate correlation value when comparing the registered image for each work data stored in the registered data storage unit 324 with the image captured by the camera. . By the above processing, it is possible to cope with the type discrimination of workpieces that are randomly conveyed on the conveyor.
The image processing unit 322c performs 1) multi-value processing similar to the above, 2) setting of a region (luminance region) for detecting a difference in surface processing / painting of a workpiece, and 3) calculation of average luminance.
The post-processing data storage unit 323 stores data of the above three image processing units.
On the other hand, the diameter and width data of the workpiece detected from the size data detection unit 1 is stored in the read size data storage unit 325. Next, the registered image search for each work size in the registered data storage unit 324 is compared in the size data comparison unit 326 and compared with the processed image stored in the processed data storage unit 323 in the registered search Gr determination unit 327. A group is determined.
In response to the processing, the image data comparison unit 328 compares the image data registered for each size in the registered data image storage unit 324 and the processed image stored in the processed data storage unit 323 (cross-correlation calculation). Is called. As a result, an image having the maximum correlation value is selected and sent to the work type final determination unit 331.
At the time of work data registration, 31 to 323 are processed in the same manner as at the time of product type discrimination, and the work image is processed. Then, the size data is input to the read size data storage unit 325 and registered in the registration data storage unit 324 as image data for each work size.
The offset amount detected by the offset detection unit 2 is stored in the read offset data storage unit 329 and then sent to the offset data comparison unit 330.
[0010]
Next, the operation of the present invention will be described with reference to the flowchart shown in FIG. First, the size detector 1 measures the size (D, W) of the workpiece (S1). That is, W is calculated by detecting the displacement of the cylinder 11 until the cylinder 11 is lowered toward the work and the tip of the cylinder 11 comes into contact with the surface of the rim portion 103. Further, D is detected by projecting light from each arc tube 13a and counting the number of photoreceivers 13b received. The workpiece whose W and D have been measured is conveyed to the offset detection unit 2 and the offset amount is detected by raising the cylinder 23 (S2). This offset amount is stored in the read offset data storage unit 329. Based on the value D, a region Z1 (see FIG. 6) for detecting the number of mounting holes and the pitch circle (PCD) is set (S3). Next, the work is sent to the image processing unit 3, the work is photographed by the camera 31, and the image is taken into the image processing device 32 (S4).
The detected W and D are stored in the read size data storage unit 325 of the image processing device 32, and are compared with the reference size by the size data comparison unit 326.
[0011]
The image captured by the image processing device 32 is converted into an electrical signal by the A / D conversion unit 320 and then stored in the read image data storage unit 321. The image processing unit 322 calculates the number and arrangement (PCD and hole angle) of the mounting holes 106 in the detection region Z1 (inside D1) set according to D as shown in FIG. (S5). This data is stored in the post-processing data storage unit 323. A registered image search Gr determination unit 327 determines a search group of registered images from the size data (D, W) in the read size data storage unit 325 and the number of attachment holes in the post-processing data storage unit 323 (S6). Data of the read size data storage unit 325, the post-processing data storage unit 323 and the read offset data storage unit 323 are stored in the registration data storage unit 324. The image processing unit 322 multi-values the image as shown in FIG. 9A (S7), and then polar coordinates are converted as shown in FIGS. 9B and 9C (S8). This converted image is phase-corrected to create a search image (FIG. 10A), and this search image is compared with the registered image (FIG. 10B) extracted from the registered data storage unit 324. (S9). The image data comparison unit 328 calculates a correlation value between the search image and the registered image (the ratio between the black part and the white part of the design hole), and when the value is less than the reference value (for example, the three values shown in FIG. 9B). The upper and lower images) are processed as the product type is not determined (S23). On the other hand, when the correlation value is greater than or equal to the reference value (the central image in FIG. 9B), it is determined as a product type determination product (S10).
[0012]
The above-mentioned product type determination products have the same size and design, but the surface state is not necessarily the same due to the difference in processing or painting conditions, so the presence or absence of the difference in surface state is detected (S11). If there is no difference in the surface state, it is determined whether there is an offset difference (S12), and if there is no offset difference, it is determined whether there is a difference in the arrangement of the mounting holes (S13). Steps S11 to S13 are performed by the image data comparison unit 328. If there is no difference in the arrangement of the mounting holes, the final product determination unit 331 determines whether it can be determined (S14), and if it can be determined, it is determined that the product is determined (S15).
In the above description, when it is determined in S11 that there is a difference in the surface state, a luminance detection area is detected, and the average luminance in the area is calculated (S16). That is, as shown in FIG. 7, an annular detection region Z2 surrounded by an outer circle D2 and an inner circle D3 is set on the outer peripheral side of the mounting hole. Since the design hole and the spoke part exist in the detection region Z2, an extreme difference in luminance occurs in the detection region Z2. However, if the average luminance of all the detection areas is obtained, the characteristics for each product type cannot be expressed. Therefore, the luminance is divided into a plurality of levels (for example, three levels) as shown in FIG. ) Area average. The calculated average luminance is compared with a reference value (registered data) (S17), and the surface difference is stored in the product type final determination unit 331 (S18). If there is an offset difference, it is compared with registered data (offset) (S19), and the offset difference is stored in the product type final determination unit 331 (S20). If there is a mounting hole arrangement difference, it is compared with registered data (hole arrangement) (S21), and the arrangement difference is stored in the product type final determination unit 331 (S22). In the product final determination unit 331, whether or not the product can be determined is determined from the data stored in S18, S20, and S22 (S14).
In the present invention, since the size detection unit determines the size group to which the work belongs, and the image processing is performed, that is, the reference work is selected from the same design and the same size work, the time required for the image processing is reduced. It can be shortened. Further, by using a cylinder with a built-in encoder as a sensor for detecting the size, it is possible to obtain a more accurate and simplified structure.
In the present invention, the apparatus units can be combined into one according to the conditions of processing time and installation space.
[0013]
【The invention's effect】
As described above, according to the present invention, the size of the workpiece and the number of mounting holes are measured to identify the size group to which the workpiece belongs, and then the reference pattern preset in the image processing unit with the workpiece design is set. Since the comparison is made, it is possible to quickly and accurately discriminate the types of workpieces having the same design and different sizes. In addition, according to the present invention, the type of the workpiece is determined using the workpiece size and the number and arrangement of the mounting holes as parameters, and the type of the workpiece is also determined using the surface state of the workpiece as a parameter. It can be determined accurately. In the present invention, since the image processing is performed by function division, the processing time can be shortened.
[Brief description of the drawings]
FIG. 1A is a front view of an aluminum wheel, and FIG.
FIG. 2 is a plan view showing an apparatus for carrying out the present invention.
3 is a cross-sectional view taken along the line AA in FIG.
FIG. 4 is a block diagram showing a configuration of the present invention.
FIG. 5 is a flowchart of an embodiment of the present invention.
FIG. 6 is a diagram showing a region for detecting the number of mounting holes and PCD in the present invention.
FIG. 7 is a diagram showing a region for detecting a difference in surface state according to the present invention.
FIG. 8 is a diagram illustrating a brightness level of a workpiece.
FIG. 9A is a diagram illustrating an example of an image read by the image processing unit according to the present invention, FIG. 9B is a diagram illustrating an example of a homopolar coordinate conversion image, and FIG.
FIG. 10 is a diagram showing an example of a registered image in the present invention.
[Explanation of symbols]
1 size detector, 2 image processor, 3 offset detector, 31 camera,
32 image processing equipment, 100 aluminum wheels, 105 design

Claims (2)

Measure the size of the bottomed cylindrical workpiece having a plurality of shape patterns separated on the end face and a plurality of holes arranged concentrically on the inner periphery thereof, and the number of holes based on the measurement results. Select the group to which the image belongs, photograph the end face of the workpiece, multi-value the image signal, convert it to polar coordinates, and compare the converted image with the registered images in the selected group. Selecting a workpiece having the same size and design, and calculating an average luminance of a region having a predetermined luminance level of an annular region including a part of the shape pattern of the end surface for the workpiece, By comparing the value with a reference value, a workpiece having the same surface condition of the end face is selected, and for the workpiece, by comparing the hole arrangement with the reference value, Varieties discriminating method characterized by determining the species. 2. The method for discriminating workpiece types according to claim 1, wherein the workpiece is an aluminum wheel, and the sizes are an outer diameter, a width and an offset.

Images