JP2000121569A - Apparatus for inspection of flaw on surface of rod - Google Patents

Abstract

PROBLEM TO BE SOLVED: To obtain an inspection apparatus by which a knocked flaw such as an indentation flaw or the like formed on the surface of a rod is detected accurately and by which the quality of the rod is judged with high accuracy on the basis of the size and the depth of the flaw. SOLUTION: This inspection apparatus is constituted of an imaging device 2 which images the surface of a rod 5 as an object to be inspected, over the whole circumferential face of the rod while the rod is being turned. In addition, it is constituted of a flaw judgment part 36 in which the difference in a shade value between adjacent pixels on the surface of the rod is detected on the basis of an output signal from the imaging device 2 or in which an unevenness value is detected on the basis of the difference in the shade value and in which a flaw is discriminated when the shade value or the unevenness value is adjacent to each other at an interval which is set in the circumferential direction or the axial direction. In addition, it is constituted of a flaw discrimination means 3 which calculates the flaw as an area or a volume and which corrects image data when the rod is jumped.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a method for detecting flaws on the surface of a cylindrical object such as a cylinder or a rod used in a shock absorber or the like by using image processing to judge the flaws and to bounce a rod of a rotating mechanism. The present invention relates to a rod surface flaw inspection device that can perform correction by data processing even when there is a problem, specifically, a rod dent flaw and a surface flaw that occur when a rod is dropped or hit during rod manufacture. The present invention relates to a rod surface flaw inspection device which can be detected.

[0002]

2. Description of the Related Art JP-A-49-129585 discloses that
A technique is described in which a laser beam is scanned along a generatrix on a slowly rotating rod, and the reflected light is received by a photodetector to photoelectrically detect a flaw on the rod surface. When receiving light, a bright-field method in which light is received in the specular reflection direction with respect to the rod surface (scratch is detected as a dark area) or a dark-field method in which light is received in a direction different from the specular reflection direction (scratch is detected as a bright light area) ) Can be adopted.

[0003] Japanese Patent Application Laid-Open No. 5-180777 discloses that a cylindrical object to be inspected is rotated, image information obtained by dividing the outer peripheral surface of the object into a plurality of parts in the axial direction is captured by a television camera, and the luminance of each divided image information is obtained. There is described an appearance inspection apparatus for a cylindrical object in which components are added and a portion where the added luminance component exceeds a predetermined threshold value is determined to be defective.

[0004]

Various techniques have been proposed for inspecting a surface to be inspected for flaws based on image information of the surface to be inspected obtained by photoelectrically converting reflected light from the surface to be inspected. However, the conventional technology detects a portion where image information (for example, density value or luminance value) of a surface to be inspected exceeds a reference value (threshold value) as a flaw, and determines based on the number of detected flaws and the area of the flaw. The pass / fail judgment is made.

[0005] Cylinder rods used in shock absorbers and the like may be damaged by dropping of the rods and collision of the rods during manufacturing. A concave portion and a convex portion are formed in the damaged portion. In the case of the shock absorber, if the surface of the cylinder / rod has a protrusion, the protrusion may cause a scratch or a crack in the oil seal, which may cause oil leakage. For this reason,
It is necessary to detect the dent scratches on the rod surface to determine the acceptability. However, it is difficult with the conventional scratch inspection technology to accurately detect the dent scratches and determine the acceptability.

For example, if the reference value (threshold) is set low (if the detection sensitivity of the flaw is set high) so that a shallow flaw can be detected in a conventional inspection apparatus, a shallow spot-shaped flaw other than a dent flaw is formed. All dust and dirt are detected, and a defect is determined based on the detection result of a shallow dot-like flaw. Therefore, if the reference value (threshold value) is set high (if the detection sensitivity of the flaw is set low) so that only deep point-like flaws can be detected, there is a problem that a shallow dent flaw cannot be detected. In addition, there is a problem that it is difficult to accurately detect a flaw due to a variation in brightness of illumination, a variation in an image due to a rod rotation shake, and the like.

SUMMARY OF THE INVENTION The present invention has been made to solve such a problem, and an object of the present invention is to provide a rod surface flaw inspection apparatus capable of accurately detecting dent flaws on a rod surface and performing flaw determination. Aim.

[0008]

According to the present invention, there is provided a rod surface flaw inspection apparatus according to the present invention, wherein both ends of a rod to be inspected are fixed and rotated in a circumferential direction, and the entire surface of the rotating rod is rotated. Over time, an imaging device for imaging the surface of the rod, and based on an output signal from the imaging device, calculate the dent scratch on the surface of the cylinder / rod as an area or a volume, and correct the image data when the rod bounces. It is provided with a flaw determining means and an image display device for displaying a flaw determination result, and can detect a flaw as an area or a volume even if a rod bounces.

The flaw determining means according to the present invention comprises an A / D converter for converting an image signal of a rod surface output from an imaging device into image data, and an image storage unit for storing the A / D converted image data. A gray value setting unit that sets a gray value difference between adjacent pixels based on image data; an edge pixel detection unit that detects a pixel exceeding the set gray value difference as a flaw; An area calculation unit that groups the pixels whose intervals are within a predetermined pixel interval, and calculates the number of pixels of the grouped pixel group, and a value of the area calculation unit and a predetermined scratch area determination criterion. The image processing apparatus was provided with a flaw determining unit for performing flaw determination by comparing the values of the flaw detection unit, an image input control unit for controlling the input of the imaging device and the image, and an operating unit for controlling the start-up end and requesting the display. Scratches from the size of the area It can be detected.

In the flaw determining means according to the present invention, the number of flaw pixels detected based on the difference in gray level between the set pixels is determined as follows:
When a predetermined reference number is not reached, a gray value setting unit that changes the gray value difference to a small value is provided, so that a small flaw and a large flaw can be detected.

A flaw determining means according to the present invention converts an image signal of a rod surface output from an imaging device into image data and stores an image data, and an image storage unit for storing image data in a rod circumferential direction based on the image data. A continuous change in the gray value difference is obtained between adjacent pixels, and a pixel in which the polarity of the continuous gray value difference changes is defined as an up change point or a down change point, and the gray level from the up change point to the down change point is determined. The difference is determined as the up-down difference, the shading difference from the down-change point to the up-change point is taken as the down-height difference, and the maximum value of the down-height difference and the up-down difference is obtained as the maximum down-height difference and the maximum up-height difference, respectively. A profile detection unit that determines the section of the maximum down section and the maximum up section, a height difference setting value that determines the validity of the obtained maximum down section and the maximum up section, and the maximum down section and the maximum up section are the axes. The volume calculation unit that calculates the volume of the grouped wound pixel groups, and compares the volume calculated by the volume calculation unit with a preset scratch determination reference value to determine Since the apparatus includes the flaw determining unit for performing the determination, the volume can be calculated and the flaw can be detected based on the size of the volume.

[0012] The flaw determining means according to the present invention, when the total number of height differences detected by the continuous change of the set grayscale value difference between pixels does not reach the predetermined reference number, the height difference Is provided with a height difference setting unit for changing the value of the parameter to a small value, so that deep and shallow scratches can be detected.

A flaw determining means according to the present invention converts an image signal of a rod surface output from an image pickup device into image data and stores the image data, and an image storage unit which approaches the rod axis direction based on the image data. The brightest part of the maximum density difference in one line in the circumferential direction is sequentially obtained between pixels, and the average of this brightness and the variation in the brightness of each column are obtained. The rod splash during rotation is detected from the variation. A white line detecting section and a correcting section for correcting image data based on the output from the white line detecting section can detect a flaw even if the rod jumps.

[0014]

Embodiments of the present invention will be described below with reference to the accompanying drawings. FIG. 1 is a block diagram of a rod surface flaw inspection apparatus according to the present invention. The rod surface flaw inspection device 1 according to the present invention includes an imaging device 2, a flaw discriminating means 3, and an image display device 4.

The image pickup device 2 includes a rotation driving unit that holds the rod and rotates it around the axis, an illumination light source that can irradiate uniform illumination light in the axial direction of the rod, The signal is supplied to the flaw determining means 3 as image information 21a by a CCD camera or the like for imaging the rod surface.

When the image information 21a is supplied from the imaging device 2, the flaw determining means 3 performs A / D conversion with an A / D converter.
An area calculation unit that stores in the image storage unit and detects the size of the flaw by the area, or a volume calculation unit that detects the flaw by the depth and detects the volume, and bounces when the rod is not properly installed and there is a bounce The image display device 4 includes a correction unit that detects and corrects the image, an image input control unit that controls image input / output, an operation unit that issues an instruction such as a display request, and a display control unit that performs display control. The display signal 40a is supplied.

The image display device 4 is constituted by using a CRT display device, a liquid crystal display device, or the like. When a display signal 40a is supplied from the flaw determining means 3, an operation menu is displayed on the screen or a flaw determination standard is displayed. Flaw information can be displayed on the screen in response to various operation requests such as setting of a value and selection switching of a display image.

FIG. 2 is a block diagram of a main part of an embodiment of the rod surface flaw inspection apparatus according to the present invention. In FIG. 2, the rod surface flaw inspection device 6 includes the imaging device 2, the flaw determination means 7, and the image display device 4. Imaging device 2
Is an imaging unit 21, an illumination light source 22, and an imaging control unit 23.
And a rotation drive unit 24. The imaging unit 21 is configured using a one-dimensional CCD camera, and the rod 5 to be inspected is
Are arranged so as to face the rod 5 with the axial direction as a visual field.

The illumination light source 22 is a linear light source so that uniform illumination light can be emitted in the axial direction of the rod 5.
The rotation drive unit 24 holds the rod 5 mounted on a rod mounting unit (not shown), and rotates the rod 5 around its axis based on a rotation command signal 23a supplied from the imaging control unit 23.

When the image input command signal 37a is supplied from the flaw determining means 7, the image pickup control section 23 controls the image pickup operation of the image pickup section 21 and the rotation operation of the rotary drive section 24, and
The surface image of the entire circumference of is taken. The imaging control unit 23 supplies the imaging unit control information 23b to the imaging unit 21, thereby controlling the imaging timing of the imaging unit 21 and the output timing of the captured image signal (image signal for one line) 21a. The output timing signal 23c is supplied to the flaw determining means 7 before the output of the image signal (image signal for one line).

When the imaging of one line and the output of the image signal 21a are completed, the imaging controller 23 supplies a rotation command signal 23a to the rotation driver 24 to rotate the rod 5 by a predetermined angle. The imaging control unit 23 captures a surface image over the entire circumference of the rod 5 by imaging the rod surface and rotating the rod 5.

In the present embodiment, 4000
By using a one-dimensional CCD camera having pixels, an image in which the entire length of the rod 5 in the axial direction is 4000 pixels is obtained.
Further, the rod 5 is rotated by 0.75 degrees at a time, and one round of the rod is taken in at 480 pixels.

As a result, the entire peripheral surface of the rod 5 becomes 400
An image having 0 pixels × 480 pixels is obtained.
The number of pixels of the captured image and the moving amount of the rod 5 are as follows.
It can be set appropriately according to the size of the flaw to be detected and the detection resolution. In the present embodiment, the imaging unit 21
Although a configuration using a CCD camera having an imaging lens system is described above, a contact type image sensor may be arranged near the rod surface to obtain an image of the rod surface.

Further, in this embodiment, the configuration in which the rod 5 is rotated has been described. However, by rotating the imaging unit 21 while maintaining a predetermined distance from the rod surface, an image of the entire peripheral surface is obtained. Is also good. The rod 5 and the imaging unit 21 may be rotated in opposite directions. When the imaging unit 21 is rotated, the imaging unit 21 is rotated while the positional relationship between the imaging unit 21 and the illumination light source 22 is kept constant.

FIG. 3 is an explanatory diagram showing an example of the positional relationship between the illumination light source and the image pickup section. In the present embodiment, the rod 5
The illumination light from the illumination light source 22 is irradiated at a predetermined incident angle θ1 with respect to the normal direction of the surface of
The imaging unit 21 is arranged in a direction (for example, a direction normal to the surface of the rod 5) different from the direction in which the angle θ2 and the emission angle θ2 are equal to each other to image the rod surface.

The imaging section 21 may be arranged in the regular reflection direction. If there is a scratch on the rod surface, the reflection angle of the reflected light changes (due to the irregular reflection caused by the scratch), so that the amount of light received by the imaging unit 21 changes. As a result, a change in shading (change in brightness) occurs in the captured image, so that it is possible to detect a flaw based on the change in shading (change in brightness) in the captured image.

As shown in FIG. 2, the flaw discriminating means 7 comprises an A /
D converter 31, image storage unit (memory) 32, gray value setting unit 33, edge pixel detection unit 34, area calculation unit 3
5, a flaw determining unit 36, an image input control unit 37, an operation unit 39, and a display control unit 40. The flaw determining means 7 can be constituted by a personal computer having an image input board and the like and an image processing program.

The A / D converter 31 A / D converts the image signal 21a output from the imaging unit 21 for each pixel, and converts the image signal on the rod surface into image data 31a. In this embodiment, a monochrome CCD camera is used for the imaging unit 21 and a gray-scale image data (luminance data) 31a of 256 gradations is obtained by using an A / D converter 31 having a resolution of 8 bits. I have to.

In this embodiment, the configuration has been described in which the image signal 21a output from the imaging unit 21 is directly input to the A / D converter 31, but an analog signal processing unit is provided before the A / D converter 31. The analog signal processor amplifies the image signal 21a and corrects the DC level (average output level) so that the dynamic range of the image signal 21a (minimum value and maximum value of the voltage of the image signal) is A / D converted. It is also possible to generate an image signal within the input voltage range of the converter 31 and to input the generated image signal to the A / D converter 31.

By adopting such a configuration, a minute change in the amount of reflected light due to a flaw or the like can be detected, and the detection resolution can be improved (the flaw detection capability can be improved). The image data 31a is stored in the image storage unit 32
And is temporarily stored in the image storage unit 32.

When an inspection start request 39a is supplied from the operation unit 39, the image input control unit 37 outputs the image input command signal 3
7a is supplied to the imaging control unit 23 to start imaging the rod surface. The image input control unit 37 controls the image capture timing based on the output timing signal 23c supplied from the imaging control unit 23.

The image input control unit 37 supplies the A / D conversion timing clock signal 37b to the A / D converter 31 to A / D convert the image signal 21a for each pixel and write the image signal 21a. By sequentially supplying the address information 37c to the image storage unit 32, the image data 31a corresponding to each pixel is written at a predetermined address. Thus, the image storage unit 32 stores in the image storage unit 32 a multi-valued (2 luminance) image data (brightness data) in which the axial direction of the rod 5 is the main scanning direction and the circumferential direction of the rod is the sub-scanning direction.
(56 gradations) image is stored.

FIG. 4 is an explanatory diagram of the rod surface image stored in the image memory. As shown in FIG. 4, the surface image of the entire circumference of the rod 5 has, for example, 4000 pixels in the axial direction (X direction) and 480 pixels in the circumferential direction (Y direction) of the rod.
(80 lines), each pixel is temporarily stored in the image storage unit 32 as image information including density data (luminance data) of 256 gradations.

The gray value setting section 33 shown in FIG. 2 is constituted by a memory such as a ROM, and supplies a relatively large value (for example, 20 levels) to the edge pixel detecting section 34 as an initial value of the density difference value 33a for flaw detection. I do. The edge pixel detection unit 34 includes a line memory, a comparator, a counter, and the like.
The density data for one line in the circumferential direction (480 pixels in the Y direction in FIG. 3) is read from 2 and the read density data is temporarily stored in the density data storage unit in the edge pixel detection unit 34, and the rod circumference is read. A pixel whose gray value difference between pixels adjacent in the direction exceeds the gray value difference set by the predetermined gray value setting unit 33 is detected as a flaw. Here, the edge pixel is a pixel that has been detected as a flaw exceeding a set density difference.

Further, without providing a density data storage unit for temporarily storing density data for one line in the edge pixel detection unit 34, the edge pixel detection unit 34 reads out the density data from the image storage unit 32 sequentially, A damaged pixel may be detected.

FIG. 5 is an explanatory diagram of the flaw detection processing in the edge pixel detection section. FIGS. 5 (a) and 5 (b) are shown as graphs in order to make the change in the gray value (luminance value) easy to understand. FIGS. 5A and 5B show a part of the density data (shading values) for one line in the circumferential direction. In addition,
FIG. 5A shows a weak dent, and FIG. 5B shows a strong dent. In both figures, 11 pixels of a part for one line in the circumferential direction are displayed.

In FIG. 5A, the vertical axis represents the gray level (luminance value), and the horizontal axis represents the pixel number (corresponding to the position in the circumferential direction).
In FIG. 5, when the gray value set by the gray value setting unit 33 is set as the density difference value 33a (for example, 20 levels) for flaw detection, the gray value difference between the target pixel and the immediately preceding pixel is determined. The absolute value is compared with the flaw detection density difference value 33a to detect whether the level is equal to or higher than 20 levels. In FIG. 5A, since the density difference value for scratch detection 33a is less than 20 levels, it is not detected as a scratch.

FIG. 5B shows the target pixel (Y4 or Y9) and the immediately preceding pixel (Y3 and Y9) in Y3 and Y4 and Y8 and Y9.
Alternatively, since the absolute value of the density difference from Y8) is equal to or greater than the set density difference value 33a for scratch detection (for example, 20 levels), it can be detected as a scratch. Further, in FIG. 5B, Y0-Y3, Y
Pixels 4-Y8 and Y9-Y10 are not detected as flaws.

The first pixel (Y0) in the circumferential direction is not processed, the second pixel (Y1) is compared with the previous pixel (Y0), and the last pixel (Yn) is determined. The comparison with the first pixel (Y0) is performed. This flaw detection is performed by the edge pixel detection unit 34. The edge pixel detection unit 34 compares the absolute value of the gray value difference between the target pixel and the immediately preceding pixel without paying attention to the brightness of the pixel. The gray value set by the gray value setting unit 33 is the density difference value 3 for flaw detection.
If it has increased by 3a (for example, 20 levels) or more, the pixel of interest is determined to be an edge pixel, and the built-in counter increments the count of edge pixels by one.

This processing is performed in the circumferential direction and further in all the axial directions. If the total number of edge pixels of the counter detected after performing the entire circumference of the rod is small, the density difference value 33a (for example, 20 levels) for flaw detection is reduced (for example, 15 levels), and then again in the circumferential axial direction. Do.

By this processing, the flaw detection capability can be raised to the detection resolution of the input gray-scale image, and it becomes possible to detect shallow flaws. In the present embodiment, flaw detection is performed by paying attention only to a change in which the gray value (luminance) increases. It should be noted that the flaw may be detected based on a decreasing change in the gray value (luminance) instead of an increasing change in the gray value (luminance).

Further, both an increase and a decrease in a change in gray value (change in luminance) may be detected. Next, it is determined whether or not the detected edge pixels are continuous in the circumferential direction and the axial direction. If the detected edge pixels are continuous or within 5 pixels of the detected edge pixels, the other edge pixels have the same group number. give. The total number of edge pixels for each group is determined from the detected edge pixels having the same group number.
A group in which the number of edge pixels is larger than a specified value is determined as a flaw.

The edge pixel detecting section 34 shown in FIG. 2 supplies the counted edge pixel number to the gray value setting section 33 as an edge number signal 34a. If the total sum of the detected edge number signals 34a is 1000 or less, the gray value setting unit 33 determines that the flaw detection density difference value 33a is too high and the flaw on the rod surface cannot be accurately detected. Then, the flaw detection density difference value 33a is changed to a value lower than the initial value, and the edge pixel detection unit 34 performs the flaw detection processing again. The gray value setting unit 33 determines that the sum of the edge number signals 34a is 1000
If the number exceeds the threshold value, the grouping start information 33b is supplied to the area calculation unit 35 to start flaw pixel grouping.

The gray value setting section 33 outputs an edge number signal 34a.
Is greater than 100 and less than or equal to 1000, the flaw detection density difference value 33a is reduced by two levels. That is, the density difference value 33a for flaw detection is changed from the initial 20 level to the 18 level. When the sum of the edge number signals 34a is more than 10 and not more than 100, the gray value setting unit 33 lowers the density difference value 33a for flaw detection by three levels.

That is, the density difference value 33a for flaw detection is changed from the initial value of 20 levels to 17 levels. When the sum of the edge number signals 34a is 10 or less, the gray value setting unit 33 lowers the flaw detection density difference value 33a by 10 levels. That is, the flaw detection density difference value 33a is set to the initial value of 2
Change from level 0 to level 10.

Then, the gray value setting section 33 outputs an edge number signal 34 of a flaw pixel detected by the flaw pixel detection processing.
a density difference value 3 for flaw detection newly set based on the sum of “a”
3a is supplied to the edge pixel detection unit 34, and the edge pixel detection unit 34 performs a flaw pixel detection process based on the newly set flaw detection density difference value 33a.

If the sum of the edge number signals 34a exceeds 1000 as a result of performing the flaw pixel detection processing based on the newly set flaw detection density difference value 33a, the shading value setting section 33 starts grouping. The information 33b is supplied to the area calculation unit 35 to start flaw pixel grouping. As a result of the flaw pixel detection processing based on the newly set flaw detection density difference value 33a, the gray value setting unit 33 outputs an edge number signal 34.
If the sum of a is less than 1000, the change of the density difference value for flaw detection is repeated until the sum of the edge number signals 34a exceeds 1000.

The area calculation unit 35 includes an arithmetic circuit such as addition and subtraction,
When a grouping start information 33b is supplied from the gray value setting unit 33, the coordinate data (Xn, Yn) of the flaw pixel stored in the flaw detection result storage unit in the edge pixel detection unit 34. 34b is read in, and defective pixels are grouped.

FIG. 6 is an explanatory diagram of the grouping process in the area calculation unit. The area calculation unit 35 first performs grouping in the circumferential direction (Y direction), and then performs grouping in the axial direction (X direction). FIG. 6 shows the group range in the circumferential and axial directions. In the circumferential direction (Y direction), other flaw pixels that are within ± 5 pixels in the circumferential direction (Y direction) from the flaw pixel (pixel detected as flaw) are grouped.

In all circumferential directions (all circumferential directions), after the grouping of flawed pixels is completed, ± in the axial direction (X direction)
If another flaw pixel or another flaw pixel group exists within the range of three lines (pixels), they are grouped as the same flaw group. This grouping in the axial direction is performed over the entire axial direction.

The area calculating section 35 sequentially groups the flawed pixels within a predetermined range by repeating the grouping process shown in FIG. When there are no more flawed pixels to be grouped around the newly grouped flawed pixels, the area calculation unit 35 assigns a label name to the flawed pixel group obtained by a series of grouping and performs grouping. All the position coordinates of the damaged pixel are associated with the label name and stored in the damaged pixel group storage unit in the area calculation unit 35.

When the grouping process is completed over the entire range of the rod surface, the area calculation unit 35 sets
Determine the area of the wound.

The area calculating section 35 shown in FIG. 2 associates each area data 35a with a label name and assigns the
Supply to The flaw determination unit 36 is configured by a comparator, and performs flaw determination by comparing the flaw area extracted by the area calculation unit 35 with a preset flaw area.

The flaw determination result 36a is supplied to the display controller 40. The display control unit 40 causes the image display device 4 to display the flaw determination result 36a. The image display device 4 is configured using a CRT display device, a liquid crystal display device, or the like.

The display control unit 40 displays an operation menu on the screen of the image display device 4 so as to input a request to start image capture, to set a flaw determination reference value, and to switch selection of a display image. Various operation requests can be input.

When the setting mode of the flaw judgment reference value is designated, the display control section 40 displays a menu screen for inputting the flaw judgment reference value, and prompts the input of the flaw judgment reference value. The flaw determination reference value 39b input from the operation section 39 is stored in the flaw determination section 36, and the input flaw determination reference value 3b
9b is displayed on the screen of the image display device 4. Scratch determination unit 3
Reference numeral 6 denotes a non-volatile memory or a memory whose power supply is backed up by a battery or the like, and holds the input scratch determination reference value 39b.

When a display request 39c for the rod surface image is supplied from the operation unit 39, the display control unit 40 for displaying the image data 32 stored in the image memory 32.
Based on “a”, a grayscale image of the rod surface is generated and displayed on the screen of the image display device 4. The display control unit 40 scrolls and displays the grayscale image of the rod surface based on the display request 39c from the operation unit 39. The display control unit 40 may display a pseudo color image instead of the grayscale image.

When the display control unit 40 is supplied with the display request 39c of the defect pixel detection result from the operation unit 39, the coordinate data of the defect pixel stored in the defect detection result storage unit in the edge pixel detection unit 34 ( Xn, Yn) and a pseudo color display image of the position of the flaw pixel and the number of gray values based on the gray value number 34c, and the detection result of the flaw pixel is displayed on the screen of the image display device 4 in pseudo color.

When the display control section 40 is supplied with the display request 39c of the result of the processing for grouping the defective pixels from the operation section 39,
Based on the information 35b stored in the flaw pixel group storage unit in the area calculation unit 35, a flaw pixel grouped image is generated and displayed on the screen of the image display device 4. In addition,
The display control unit 40 receives the defect determination result from the defect determination unit 36 and the label name of the defective pixel group determined to be defective, and provides a grouped image of the defective pixel determined to be defective, position information, area information, and the like. The information may be displayed on the screen of the image display device 4 together with the defect determination result.

FIG. 7 is a block diagram of another rod surface flaw inspection apparatus. In FIG. 7, the rod surface flaw inspection device 8 includes the imaging device 2, the flaw determination means 9, and the image display device 4. The imaging device 2 supplies the image signal 21 a of the imaged rod to the flaw determining means 9.

The structure of the flaw determining means 9 is the same as that of the A / D converter 3.
1, image storage unit 32, flaw determination unit 36, image input control unit 3
7, operation unit 39, display control unit 40, height difference setting unit 43,
It comprises a profile detection unit 44 and a volume calculation unit 45. In the configuration of the flaw determining means 9, the A / D converter 3
1, image storage unit 32, image input control unit 37, operation unit 3
9 and the display control unit 40 are the same as the flaw determining means 7 of FIG.

The flaw determining means 9 of FIG. 7 and the flaw determining means 7 of FIG.
A description will be given of different parts from the above. The height difference setting unit 43 is configured by a memory such as a ROM, and sets a height difference value for detecting the depth of a flaw, and sets a relatively large value (for example, 50 levels) as an initial value of the height difference value 43a. This is supplied to the profile detection unit 44.

The profile detecting section 44 is composed of an arithmetic circuit and a storage circuit, reads density data of one line in the circumferential direction (480 pixels in the Y direction in FIG. 4) from the image storage section 32, and reads the read density data. To the profile detection unit 44
And a pixel whose height difference calculated based on the density difference between the pixels adjacent in the rod circumferential direction exceeds the height difference value 43a set in advance by the height difference setting unit 43 while being temporarily stored in the density data storage unit. Is detected as a profile flaw.
Note that the profile detection unit 44 does not provide a density data storage unit for temporarily storing density data for one line in the profile detection unit 44, and the profile detection unit 44 detects the profile flaw pixels while sequentially reading the density data from the image storage unit 32. You may make it.

FIG. 8 is an explanatory diagram of the profile flaw detection processing in the profile detection section. FIG. 8A shows a graph for easy understanding of a change in gray value (luminance value) and a part of density data (density value) for one line in the circumferential direction. FIG. 8B shows an example of distribution of detected profile flaw pixels.

In FIG. 8A, the vertical axis represents the gray scale value (luminance value), and the horizontal axis represents the pixel number (corresponding to the position in the circumferential direction).
When detecting a profile flaw pixel in the circumferential direction, the profile detection unit 44 compares the gray level of the pixel of interest with the gray level of the immediately preceding pixel without paying attention to the brightness of the pixel. To determine whether it is increasing or decreasing.

That is, when the gray value of the pixel of interest is larger than the gray value of the immediately preceding pixel, it is up, and otherwise, it is down. In FIG. 8A, Y2-Y6, Y
10-Y12, Y13-Y16 are descending, and Y6-Y
10, Y12-Y13 are upstream. When the pixel of interest is up and the previous one is down, the previous one is a valley bottom. Y6 and Y12 in FIG. 8A are valley bottoms. Similarly, when the target pixel is down and the previous one is up, the previous one is a peak. Y10 and Y13 in FIG. 8A are the peaks.

When the valley bottom is located immediately before and the peak has already been detected, the difference between the gray value of the peak and the gray value of the valley bottom is calculated, and the difference is determined as the descending height difference. If this is greater than the maximum down-difference, the value is taken as the maximum down-difference. Also, the section from the peak to the bottom of the valley is the maximum descending section. If the valley bottom has already been detected before the peak, the difference between the gray value at the peak and the gray value at the valley is calculated, and the difference is calculated as the ascending elevation difference. If this is greater than the maximum up-to-date difference, this value is used as the maximum up-down difference. Also, the section from the bottom of the valley to the top of the mountain is defined as a maximum ascending section. This process is performed for one line in the circumferential direction. In FIG. 8A, Y2-Y6 is the maximum descending section, and the maximum descending height difference is 110-65 = 45. Further, Y6-Y10 is the maximum ascending section, and the maximum ascending height difference is 115-65 = 50.

Next, the validity and invalidity of the detected maximum up section and maximum down section are determined. That is, a maximum down section having a maximum down height difference that does not reach a certain reference value (for example, 25 levels) is invalidated and there is no maximum down section. Similarly, a maximum down section having a maximum up / down difference that does not reach a reference value (for example, 25 levels) is invalid, and there is no maximum up section. This processing is performed for all the axial directions.

Here, with respect to a certain X position, a line segment in the Y direction formed from a valid maximum down section and a valid maximum up section, only a valid down section, or only a valid up section is defined as a profile flaw. Pixels. If this profile flaw pixel is continuous with the adjacent row of profile flaw pixels, or other profile flaw pixels within 5 pixels, give the same group number.

The volume calculation unit 45 includes an arithmetic circuit for addition and subtraction,
When the grouping start information 43b is supplied from the height difference setting unit 43, the coordinate data (Xn, Yn) of the profile flaw pixel stored in the flaw detection result storage unit in the profile detection unit 44. 44b is read in, and profile flaw pixels are grouped.

When there is no flaw pixel to be grouped around the newly grouped profile flaw pixels, the volume calculation unit 45 assigns a label name to the flaw pixel group obtained by a series of grouping, and All the position coordinates of the grouped flaw pixels are associated with the label name and stored in the flaw pixel group storage unit in the volume calculation unit 45.

When the group processing is completed over the entire range of the rod surface, the volume calculation section 45 calculates the volume of the wound for each wound group. The volume is obtained by adding the height difference from the peak for each profile flaw pixel and performing the same processing for all the profile flaw pixels in the same group. In FIG. 8A, since the profile flaw pixel is composed of a maximum descending section and a maximum ascending section, for the maximum descending section, the summit is Y = 2, the gray level = 110,
The valley bottom is Y = 6, the gray value = 65, and the sum of the height difference from the peak is (110−105) + (110 + 98) +
(110-68) + (110-65) = 104.
For the maximum uphill section, the summit is Y = 10 and the gray value = 1.
15, the bottom of the valley is Y = 6, the gray level = 65, and the sum of the height differences from the peak is (115-65) + (115-67).
+ (115−85) + (115−100) = 143. This is sequentially performed for the same group as shown in FIG. 8B, thereby obtaining the volume of the flaw of the group.

By this processing, even if the image is clear (the difference in height from the previous one is large) or slightly blurred (the difference in height from the previous one is small), it is detected as long as the up or down continues. Therefore, the detection ability is improved. Further, the flaw detection capability can be increased to the detection resolution of the input gray-scale image by this processing, and a shallow flaw can be detected.

The volume calculation unit 45 shown in FIG. 7 associates each volume data 45a with a label name and assigns the
Supply to The flaw determination unit 36 is configured by a comparator, and performs flaw determination by comparing the flaw volume extracted by the volume calculation unit 45 with a preset flaw volume. In addition,
Although the same scratch determination unit 36 is used in FIGS. 2 and 7, only the numerical value of the area or volume is used for comparison, so that it can be used in common.

Further, if dust or the like is attached to either the roller or the rod by the mechanism for rotating the rod, the rod may jump slightly when the roller and the rod come around the portion. At this time, the image captured by the camera has a white line at the same position in the axial direction.

FIG. 9 is a block diagram showing a main part of an embodiment of a rod surface flaw inspection apparatus for detecting rod bounce according to the present invention. In FIG. 9, the rod surface flaw inspection device 10
It comprises an imaging device 2, a flaw determining means 11 and an image display device 4. The imaging device 2 generates an image signal 21 of the imaged rod.
a is supplied to the flaw determining means 11.

The structure of the flaw determining means 11 is the same as that of the A / D converter 3.
1, image storage unit 32, flaw determination unit 36, image input control unit 3
7, operation unit 39, display control unit 40, height difference setting unit 43,
It comprises a profile detection unit 44, a volume calculation unit 45, a white line detection unit 53, and a correction unit 54. A / D converter 31, image storage unit 32, image input control unit 37, operation unit 39, display control unit 40, height difference setting unit 43, profile detection unit 44, volume calculation Part 4
5 is the same as the flaw determining means 9 in FIG.

The different parts of the configuration of the flaw determining means 11 of FIG. 9 and the flaw determining means 9 of FIG. 7 will be described. FIG. 9 is a block diagram showing the configuration of the flaw discriminating means 11 which detects the bounce of the rod and corrects the image to detect the volume of the flaw.

The white line detecting section 53 shown in FIG.
Is added to FIG. 2 or FIG. The white line detection unit 53
It is composed of a memory and a calculation unit, reads out density data for one line in the circumferential direction (480 pixels in the Y direction in FIG. 4) from the image storage unit 32, and reads the read density data for the white line detection unit 53.
In this case, the density data is temporarily stored in the density data storage unit, and further, the density data is added, and averaging is performed to divide the total data, thereby removing noise.

Next, the top Y of the maximum gray level difference in the circumferential direction
TOP is calculated, and the calculation of the top of this climb is performed in the axial direction. Averaging is performed by adding all the peaks in the axial direction and dividing by the total number to obtain an average value YAV of the peaks in the axial direction. The variation V at the top in the circumferential direction is obtained by calculating the sum of (YAV-YTOP) ² and dividing the sum by the number of pixels in the axial direction.

The variation V in the top in the circumferential direction is compared with a reference value, and if it is smaller than the reference value, it is determined that there is a white line due to rod bouncing. In other words, it indicates that the peaks of the maximum gray level difference in the circumferential direction are linearly distributed in the axial direction.

FIG. 10A is an explanatory diagram of white line detection in the white line detection section. YTP at the top of the maximum density difference in the circumferential direction
O is distributed linearly in the axial direction. FIG. 10 (b) is shown as a graph for easy understanding of the change in the gray value (luminance value). The portion projecting more than the average value in FIG. 10B is a white line.

When the white line is detected, the white line detection unit 5
The image data temporarily stored in the density data storage unit 3 is corrected by the correction unit 54 and output to the edge pixel detection unit 34 in FIG. 2 or the profile detection unit 44 in FIG.

The correction section 54 is composed of an arithmetic circuit, a memory, and the like, calculates the average of the gray values in the circumferential direction, and further calculates the average of the gray values in the axial direction. The average value Ynav of the density in the axial direction is obtained by an average value obtained by dividing the sum of the density values in the axial direction by the total number of lines 4000. Average value Anav of shading in the circumferential axis direction
Is obtained by performing the average value Ynav of the shading in the axial direction for all circumferential directions and dividing the total sum of Ynav by the total number of lines 480 in the circumferential direction.

The correction data Za of the image is obtained by Expression 1.

[0086]

## EQU1 ## Za = Zb + (Anav-Ynav)

Here, Za is the corrected image data, Zb is the original image data, Anav is the average value of the density in the circumferential direction,
Ynav is an average value of shading in the circumferential direction.

When a white line is detected by the white line detection unit 53, this correction is performed by the correction unit 54. Also, the white line detection unit 5
If no white line is detected in step 3, the image data is supplied to the edge pixel detection unit or the profile detection unit as it is.

By performing the white line detection, the image data can be corrected even when the rod bounces, and the area or volume of the flaw can be accurately detected.

In this embodiment, the flaw determining means 9 shown in FIG.
The functions of the white line detection unit 53 and the correction unit 54 are added to the flaw determination unit 11, but the flaw determination unit 7 of FIG.
3. The function of the correction unit 54 may be added so that the rod bounce is corrected first to detect an area scratch.

[0091]

As described above, in the rod damage inspection apparatus according to the present invention, the both ends of the rod to be inspected are fixed and rotated in the circumferential direction, and the surface of the rod is rotated over the entire peripheral surface of the rotating rod. An image pickup device for picking up an image, and a flaw discriminating means for calculating a dent flaw and a surface flaw on the surface of the cylinder rod as an area or a volume based on an output signal from the image pickup apparatus and correcting the image data when the rod bounces. And, with an image display device that displays the determination result of the flaw, even if there is a rod bounce, the flaw can be detected as an area or a volume,
Accurate scratch detection can be performed.

The rod surface flaw inspection apparatus according to the present invention stores an A / D converter for converting an image signal of the rod surface output from the imaging device into image data, and stores the A / D converted image data. An image storage unit, a gray value setting unit that sets a gray value difference between adjacent pixels based on image data, and an edge pixel detection unit that detects a pixel that exceeds the set gray value difference as a flaw. An area calculating unit that calculates the number of pixels of the grouped pixel group by grouping pixels in which the interval between pixels detected as flaws is within a predetermined pixel interval, and setting the value of the area calculating unit and a preset value A scratch determination unit that performs a scratch determination by comparing the values of the wound area determination criteria;
Since the image input device and the image input control unit that controls the input of the image, and the operation unit that performs control of the end of startup and a display request is provided,
Flaws can be detected from the size of the grouped area, and dent-like flaws on the rod surface can be accurately detected to make a high-quality judgment.

The flaw determining means according to the present invention is arranged such that the number of flaw pixels detected based on the difference in gray level between the set pixels is:
When the number of reference values does not reach the preset reference number, a small and large flaw can be detected because of the provision of the gray value setting section for changing the difference between the gray values to a small value.

A rod surface flaw inspection apparatus according to the present invention converts an image signal of a rod surface output from an imaging device into image data, and stores an image storage section for storing image data. A continuous change in the gray value difference is obtained between pixels adjacent in the circumferential direction, and a pixel in which the polarity of the continuous gray value difference changes is defined as an up change point or a down change point, and a down change from the up change point is performed. The shade difference between the points is the up-and-down difference, the shade difference from the down-change point to the up-change point is the down-height difference, and the maximum value of the down-height difference and the up-height difference is the maximum down-height difference and the maximum up-height difference. A profile detection unit that calculates and determines each section as a maximum down section and a maximum up section, a height difference setting value that determines the validity of the obtained maximum down section and the maximum up section, A volume calculation unit that calculates the volume of the group of flawed pixels, and a volume determination unit that calculates the volume of the grouped flaw pixel groups, and a flaw determination reference value that sets the volume calculated by the volume calculation unit in advance. And a flaw determining unit that performs flaw determination by comparing the flaws, so that the volume can be calculated and the flaw can be detected based on the size of the volume. Scratches and dents can be distinguished, and dents can be detected accurately.

The flaw determining means according to the present invention, when the total number of height differences detected by the continuous change of the gray level difference between the set pixels does not reach the preset reference number, sets the height difference. Is provided with a height difference setting unit for changing the value of the parameter to a small value, so that deep and shallow scratches can be detected.

Further, the rod surface flaw inspection apparatus according to the present invention comprises: an image storage unit for converting an image signal of the rod surface output from the imaging device into image data and storing the image data; and a rod shaft based on the image data. Between the pixels adjacent to each other in the direction, the brightest part of the maximum density difference in one line in the circumferential direction is sequentially obtained, and the average of this brightness and the variation in the brightness of each column are obtained. Since a white line detecting unit for detecting a splash and a blur correcting unit for correcting image data based on an output from the white line detecting unit are provided, even if the rod has jumped, the rod can be corrected and a flaw can be accurately detected.

[Brief description of the drawings]

FIG. 1 is a block diagram of a rod surface flaw inspection apparatus according to the present invention.

FIG. 2 is a block diagram of a main part of a rod surface flaw inspection apparatus according to an embodiment of the present invention;

FIG. 3 is an explanatory diagram illustrating an example of a positional relationship between a light source for illumination and an imaging unit.

FIG. 4 is an explanatory diagram of a rod surface image stored in an image memory.

FIG. 5 is an explanatory diagram of a flaw detection process in the edge pixel detection unit.

FIG. 6 is an explanatory diagram of a grouping process in an area calculation unit.

FIG. 7 is a block diagram of a main part of an embodiment of another rod surface flaw inspection apparatus according to the present invention.

FIG. 8 is an explanatory diagram of a profile flaw detection process in the profile detection unit.

FIG. 9 is a block diagram of a main part of an embodiment of a rod surface flaw inspection apparatus for detecting rod bouncing according to the present invention.

FIG. 10 is an explanatory diagram of white line detection in a white line detection unit.

[Explanation of symbols]

1,6,8,10 ... Rod surface flaw inspection device, 2 ... Imaging device, 3,7,9,11 ... Flaw determination means, 4 ... Image display device, 5 ... Rod (test object), 21 ... Imaging unit (One-dimensional CCD camera), 22: illumination light source, 24: rotating mechanism unit, 31: A / D converter, 32: image storage unit, 33: gray value setting unit, 34: edge pixel detection unit, 35 ... Area calculation unit, 36: flaw determination unit, 37: image input control unit, 39: operation unit, 40: display control unit, 43: height difference detection unit, 44 ...
Profile detection unit, 45: volume calculation unit, 53: white line detection unit, 54: correction unit.

 ──────────────────────────────────────────────────続 き Continued on the front page F term (reference) 2F065 AA25 AA49 AA59 BB06 FF01 FF04 GG16 HH12 JJ02 JJ08 JJ09 JJ25 QQ03 QQ24 QQ31 RR06 2G051 AA90 AB02 CA03 CD07 DA08 EA08 EA11 EA14 EA01 EA01 EA01 EB01

Claims (6)

[Claims] 1. A rod surface damage inspection device for inspecting dent scratches and surface scratches on the surface of a cylinder rod used for a shock absorber of an automobile, wherein both ends of the rod to be inspected are fixed and rotated in the circumferential direction.
An imaging device for imaging the surface of the rod over the entire peripheral surface of the rotating rod, and based on an output signal from the imaging device, dent scratches and surface scratches on the surface of the cylinder / rod are calculated as an area or a volume. A flaw discriminating means for correcting image data when a rod bounces, and an image display device for displaying a flaw judgment result. 2. An A / D converter for converting an image signal of a rod surface output from the imaging device into image data, and an image storage unit for storing the A / D converted image data. And a gray value setting unit that sets a gray value difference between adjacent pixels based on the image data, an edge pixel detection unit that detects a pixel exceeding the set gray value difference as a flaw, An area calculation unit for grouping pixels whose detected pixel intervals are within a predetermined pixel interval and calculating the number of pixels of the grouped pixel group, and a value of the area calculation unit and a predetermined scratch area determination A rod dent scar which includes a scratch determining unit for comparing the reference values and performing a scratch determination, an image input control unit for controlling an imaging device and image input, and an operating unit for controlling start-up and requesting display. Is calculated by the area The rod surface flaw inspection device according to claim 1. 3. When the number of pixels of a flaw detected due to a set difference in gray level between pixels does not reach a preset reference number, the gray level difference is changed to a small value. The rod surface flaw inspection apparatus according to claim 1, further comprising a setting unit. 4. The flaw determining means converts an image signal of the rod surface output from the imaging device into image data,
The image storage unit for storing image data, and the continuous change of the gray value difference between pixels adjacent in the rod circumferential direction based on the image data, and the polarity of the continuous gray value difference changes. A pixel is defined as an up-change point or a down-change point, and a shade difference from an up-change point to a down-change point is defined as an up-height difference, and a shade difference from a down-change point to an up-change point is a down-height difference. A profile detector for calculating the maximum value of each difference as a maximum down-difference and a maximum ascending-difference, and determining the respective sections as a maximum descent section and a maximum ascending section, and the validity of the determined maximum descent section and the maximum ascending section. A height difference setting value, and a volume calculation unit for grouping those in which the maximum descending section and the maximum ascending section are continuous in the axial direction, and calculating the volume of the grouped wound pixel group; and A wound determination unit which compares with a previously wound determination reference value set and the calculated in output unit volume performs flaw determination,
The rod surface flaw inspection device according to claim 1, wherein the rod dent flaw is calculated by volume. 5. The method of changing the height difference to a small value when the total number of height differences detected by the continuous change of the gray value difference between the set pixels does not reach a preset reference number. 2. The apparatus according to claim 1, further comprising a difference setting unit.
5. The rod surface flaw inspection device according to claim 4. 6. The flaw determining means converts an image signal of the rod surface output from the imaging device into image data,
An image storage unit for storing image data and a pixel in the circumferential direction sequentially between pixels adjacent in the rod axis direction based on the image data.
Find the brightest part of the maximum density difference in the line, find the average of this brightness and the variation in the brightness of each column, and use this variation to detect the rod splash during rotation, and this white line detection The rod surface flaw inspection apparatus according to claim 1, further comprising: a correction unit configured to correct image data based on an output from the unit.