JPH01147680A - Piston type discriminating device - Google Patents

Abstract

PURPOSE:To accurately decide the type of a piston by identifying the unevenness of the shape of the piston as a binarized shadow pattern. CONSTITUTION:A piston P and a cylinder C are illuminated by an illumination means 2 and the images or upper faces are obtained by a camera 3 for the piston and the cylinder. These obtained images are outputted and stored in a frame memory 6 via an image input board 5. Data on the memory 6 is sent to an arithmetic board 7 via a CPU 12. Thus the area of the shadow part is obtained for the data received by the board 7. A stamp mark is put on the upper face of the piston P for elimination of noise via an expansion/contraction board. Thus the type of the piston is discriminated by a stamp mark and the shadow with high accuracy.

Description

DETAILED DESCRIPTION OF THE INVENTION [Industrial Application Field] The present invention relates to a piston type discriminating device, and is particularly useful as a device for inspecting whether the correct type of piston is assembled in an engine assembly process.

[Problem of invention]

Although the types of pistons differ depending on the type of engine, it is possible to install different types of pistons in the same cylinder block, so it is possible to install the wrong type I piston in the engine assembly process.

Therefore, in order to prevent such errors or to detect errors through inspection, humans visually distinguish the type of piston, but there is a problem that it is not always possible to distinguish the type of piston reliably. Ta.

It has also been proposed to attach a stamp mark to the top surface of the piston and recognize the stamp mark with a visual recognition device to determine the type of piston, but as the number of types of piston increases, the patterns of the stamp mark also become more diverse. Moreover, the stamp marks are not necessarily clear, making it difficult to accurately determine the type of piston.

Therefore, it is an object of the present invention to provide a piston type discrimination device that can accurately discriminate the M class of a piston by recognizing the shape of the piston.

[Structure of the invention]

The piston type discriminating device of the present invention includes: illumination means for illuminating the top surface of the piston so as to produce shadows due to the unevenness of the top surface of the piston; camera means for obtaining an image of the top surface of the piston; a shade extraction binarization means for obtaining a binarized shade image for shade extraction; a shade identification means for identifying a shade pattern from the binarized shade image; and a shade identification means based on the shade identification result. Piston's! tW4
The configuration is characterized in that it includes a discriminating means for discriminating.

[Effect]

In the piston type discrimination device of the present invention, the pi type of the piston is discriminated by identifying the shape of the shading on the upper surface of the piston. Therefore, errors caused by human visual perception and blurred stamp marks can be eliminated, and the type of piston can be reliably determined.

Note that if the identification of the shape of the shading on the top surface of the piston and the identification of the stamp mark according to the present invention are used together, the number of identifications to be made will be reduced for both the identification of the shading shape and the identification of the stamp mark, and the identification that is likely to be mistaken will be avoided. You will be able to avoid doing it. In other words, pistons with different shading shapes are classified by 1-1 based on the shading shape, and pistons with no difference in shading shape are identified by stamp marks. Since there are differences in shape, there is no misidentification.

Next, since stamp marks need only be used for identification among types with common shading shapes, the number of types is reduced, and only stamp marks with different patterns can be used. . Therefore, errors in identifying stamp marks are also prevented.

As a result, the type of piston can be determined much more accurately than in the past.

〔Example〕

Hereinafter, the present invention will be explained in more detail based on embodiments shown in the drawings. Here, FIG. 1 is a block diagram of the configuration of a piston type discriminating device according to an embodiment of the present invention, FIG. 2 is a schematic flowchart of the operation of the piston type discriminating device shown in FIG. 1, and FIG. FIG. 4 is an example of a binary shading image, FIG. 5 is an example of a window set in a shaded area, and FIG. 6 is an example of a binary stamp mark image and a window. Illustrative diagram, No. 71!
l is an explanatory diagram of hole-filling processing, Fig. 8 is an example diagram of a binarized stamp mark image subjected to hole-filling processing, and Fig. 9 is a binarized stamp mark image in a window and black pixels appearing in the X direction. 10 is an example of a histogram of a binarized stamp mark image cut out in the X direction and a black pixel appearing in the Y direction. FIG. 11 is an example of a histogram of a binarized stamp mark image cut out in the FIG. 12 is an explanatory diagram of the noise erasing process. FIG. 13 is an example of a binarized stamp mark image that has been subjected to the noise erasing process. In the case of three points lined up in the direction, the corresponding figure in Figure 9, Figure 15
The figure is a diagram corresponding to FIG. 1θ of the same, and FIG. 16 is a diagram corresponding to FIG. 13 of the same. Note that the present invention is not limited to the following examples.

The piston type discriminating device 1 shown in FIG.
, a frame memory 6 , an arithmetic board 7 , an expansion/deflation board 8 , an input/output board 9 , an operation surface 10 , a sequencer 11 , and a CPU 12 .

The illumination means 2 shines illumination light from diagonally upward onto the top surface of the piston P assembled in the cylinder block C to create shadows due to the unevenness of the top surface of the piston P, and also to mark the stamp mark attached to the top surface of the piston P. It's lit.

The camera means 3 is, for example, a COD camera, and captures the upper surface of the piston P and the cylinder block C around it as an image.

The captured image is input to the frame memory 6 via the image input board 5 and stored.

5. The image can be viewed by the operator through the monitor means 4.

The calculation board 7 is a part that performs shadow area calculation and histogram calculation at high speed.

The expansion/contraction board 8 is a part that performs hole filling processing and noise erasing processing on the binarized stamp mark image at high speed.

The input/output board 9 is a sequencer that transmits operator instructions inputted on the operation surface 10 to the CPU 12, and automatically removes the engine when it is determined that the wrong type of piston is installed. 11, CPU
This is the part that gives instructions from I2.

The CPU 12 is the center of all control, and its operation will be understood from the following description.

FIG. 2 is a schematic flowchart of the operation of the piston type discriminating device 1, and the operation will be explained in this order.

For convenience of explanation, four types of pistons P l + p, l P co, and P piston are considered as shown in FIG. Due to the difference in the concave shape of the top surface of the piston, it can be divided into a group of p,, p, and a group of +P3 + p, and the former is called A.
type, and the latter is designated as type B. Also, due to the difference in the types of cars used, there are two groups: p,, p, and P,.

The former group is called the X-type, and the latter is called the Y-type.

Type A and type B can be distinguished by the shading 21 because there is a difference in the recess on the top surface of the piston, but the shading 21
In some cases, it is not possible to distinguish between the X series and the Y series. Therefore, a stamp mark 22 is attached to distinguish between the X series and the Y series.

There are four types of pistons, but there are two stamp marks.
Just as the N class is fine and the difference is large, the X series has one nine points,
The Y system uses three nine points.

Now, in FIG. 2, an image corresponding to the brightness of the upper surface of the piston P is obtained and stored in the frame memory 6 (Sl).
.

Next, the CPU 12 binarizes the previously stored image using a darkness value that allows the shadow 21 to be clearly extracted (32
), whereby a binarized shadow image as shown in FIG. 4 is obtained and stored in the frame memory 6.

When the binarized shadow image is obtained, the CPU 12 sets the window W as shown in FIG.
The area of the shaded part (black part) within is calculated (33).

If the area of the calculated shadow is 20% or more of the area of window W, CPU 2 identifies it as type A;
If it is less than 0%, it is identified as type B (S4).

Next, the CPU 2 binarizes the previously stored image using a darkness value that allows the stamp mark 22 to be clearly extracted (S5).

As a result, a binarized stamp mark image as shown in FIG. 6 is obtained.

A window W is set for this binarized stamp mark image, and hole filling processing is performed within the window W.
That is, since the stamp mark 22 is not necessarily completely marked and may have some blurring as shown in FIG. 7, first check the run length of white pixels in the If there are consecutive pixels, the white pixels are replaced with black pixels. That is, 23 shown in FIG.
It is converted into a stamp mark such as (S6), and furthermore, Y
If the same process is performed for the direction, 24 as shown in FIG.
A stamp mark such as the following is obtained (S7). Such processing is performed using the calculation board 7 and the expansion/contraction board 8.

FIG. 8 shows the binarized stamp mark image after performing the above-mentioned hole-filling process.

Next, the CPU 2 uses the calculation board 7 to perform a process of cutting out the stamp mark 24 in the window W.
That is, as shown in FIG.
A histogram of black pixels arranged in the direction is calculated in the Y direction (38), and a binarized stamp mark image is cut out using the smallest value and the largest value among the Y coordinates of the minimum point of this histogram H1 (S9) .

The cut out binarized stamp mark image is as shown in Fig. 10, but when the histogram of black pixels in the Y direction is calculated in the X direction, it becomes H2 (510
), if we perform the same extraction as above (Sll)
, an area 26 slightly larger than the stamp mark 24 can be obtained, as shown in FIG.

The CPU 12 performs noise erasing processing in this area 26. That is, as shown in FIG. 12, the CPU 12 calculates the run length of black pixels in the , Convert to white pixel 0 For example, if the number of black pixels exceeds the range of 5 dots to 20 dots, convert to white pixel. If this noise cancellation is performed in the X direction and in the X direction (S12.5
13), for example, a noise component part 25 as shown in FIG.
Even if there is a stamp mark, it is removed and a binarized stamp mark image as shown in FIG. 13 is obtained.

The CPU 2 counts the number of stamp marks from the binarized stamp mark image after noise cancellation (S14), that is, checks the continuity of black pixels in the X direction,
When the sequence of black pixels is interrupted, the sequence of black pixels in the X direction is checked, and when the sequence of black pixels is interrupted, the count is set to 1. If this is done for the entire area 26, the number of stamp marks within the area 26 can be counted.

If the count number of stamp marks is 1, it is identified as Y-series, and if it is 3, it is identified as Y-series (S
15).

Thus, there is an illumination means for illuminating the top surface of the piston so as to create a shadow due to the unevenness of the top surface of the piston and to make the stamp mark attached to the top surface of the piston visible, a camera means for obtaining an image of the top surface of the piston, and a binarization process for the image. a shadow extraction binarization means for obtaining a binarized shadow image for extracting the shadow; a shadow identification means for identifying a shadow pattern from the binarized shadow image; a binarization means for stamp mark extraction to obtain a binarized stamp mark image for extracting the stamp mark;
a stamp mark identification means for identifying the pattern of the stamp mark from the disulfide stamp mark image; ! The piston type discriminating device 1, which is equipped with a discriminating means for discriminating the type, can identify whether the piston is A type or B type, and whether it is Y type or Y type. It can be determined (S16).

9, FIG. 10, and FIG. 13 show the case where there is one stamp mark, but the case where there are three stamp marks is as shown in FIG. 14, FIG. 15, and FIG. 16.

〔Effect of the invention〕

According to the present invention, there is provided an illumination means for illuminating the upper surface of the piston so as to produce a shadow due to the unevenness of the upper surface of the piston, a camera means for obtaining an image of the upper surface of the piston, and a method for binarizing the image to extract the shadow. binarization means for extracting shadows to obtain a binarized shadow image; shadow identification means for identifying a pattern of shadows from the binarized shadow image; and discrimination of the type of piston based on the result of identifying the shadows. A piston type discriminating device is provided, which is characterized by comprising a discriminating means, which makes it possible to accurately discriminate the type of piston. Therefore, it is extremely useful as a piston type inspection device in the engine assembly process. ′

[Brief explanation of the drawing]

Fig. 1 is a block diagram of the configuration of a piston type discrimination device according to an embodiment of the present invention, Fig. 2 is a schematic flowchart of the operation of the piston ffi classification device shown in Fig. 1, and Fig. 3 is a diagram showing piston types and shading. and illustrations of stamp marks, 4th v.
l! J is an example of a binarized shadow image, Fig. 5 is an example of a window set in a shaded area, Fig. 6 is an example of a binarized stamp mark image and window, and Fig. 7 is a hole-filling process. 8th vgJ is an example of a binarized stamp mark image that has been subjected to hole-filling processing.
An example diagram of a digitized stamp mark image and a histogram of black pixels appearing in the X direction, FIG. 10 is an illustrative diagram of a binarized stamp mark image cut out in the X direction and a histogram of black pixels appearing in the X direction, FIG. 11 is an example diagram of a binarized stamp mark image cut out in the X direction and the X direction, FIG. 12 is an explanatory diagram of noise elimination processing, and FIG.
The figure is an example of a binarized stamp mark image that has been subjected to noise elimination processing, and Fig. 14 shows three stamp marks arranged in the Y direction.
9, FIG. 15 is a diagram equivalent to FIG. 1O, and FIG. 16 is a diagram equivalent to FIG. 13. [Explanation of symbols] 1... Piston type identification device 2... Illumination means 3... Camera means 4...
Monitoring means 5... Image input board... Frame memory 7... Arithmetic board 8... Expansion/contraction board 12... CPU 21... Shading 22.23
．． 24... Stamp mark 26... Cut out area P... Piston C... Cylinder lock W
...Window H,, H2...Histogram. Applicant Daihatsu Motor Co., Ltd.

Claims (1)

[Claims] 1. Illumination means for illuminating the top surface of the piston so as to create shadows due to the unevenness of the top surface of the piston, camera means for obtaining an image of the top surface of the piston, and binarization processing of the image to extract the shadow. Binarizing means for extracting shadows to obtain a shadow image; shadow identifying means for identifying a pattern of shadows from the binary shadow image; and determining means for determining the type of piston based on the result of identifying the shadows. A piston type discrimination device comprising: