JPH08219999A - Method and apparatus for calibrating surface defect-inspection optical system - Google Patents

Abstract

PURPOSE: To improve the calibration reliability and easily finely adjust a detection position to a slit light, by reducing a count of steps of a calibration work by a worker and eliminate irregularities when a relative position of an optical system is evaluated. CONSTITUTION: According to the method and apparatus for calibrating a surface defect-inspection optical system, partial roughness circular marks 36A, 36B are formed at a master roll 16, and the reflecting amount of light at the parts is measured, whereby a detection position of a line sensor to a slit light is adjusted. The master roll 16 having a plurality of reference marks 32-36A, 36B formed on the surface thereof is arranged at an inspection stage of the surface defect inspection optical system. A characteristic amount representing a relative positional relationship of the optical system is calculated from each read image of the reference marks 32-36A, 36B of the master roll.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a cylindrical inspection object (roll) such as a photosensitive drum, a fixing roll, a charging roll and a developing sleeve used in an electrophotographic image forming apparatus such as a digital copying machine and a laser printer. A method and an apparatus for calibrating a surface defect inspection optical system for automatically inspecting a surface defect of a member), particularly, a man-hour for calibration work is reduced, and highly reliable calibration is possible, and a cylindrical object to be inspected The present invention relates to a method and apparatus for calibrating a surface defect inspection optical system in which inspection accuracy is improved by facilitating fine adjustment of the light receiving position in the circumferential direction of the.

[0002]

2. Description of the Related Art Roll-shaped parts such as a photoconductor drum, a heat fixing roll, a charging roll, and a developing sleeve in an electrophotographic image forming apparatus such as a digital copying machine and a laser printer are important for controlling the quality of an output image. It is a strict component, and it is strictly required to eliminate surface defects.

The above-mentioned roll-shaped parts are manufactured by forming functional layers such as a photosensitive layer, a semiconductive layer and a protective layer on the surface of a metal roll or a rubber roll. For example, a photosensitive drum is formed by dip-coating a multi-layered photosensitive layer on a metal pipe, a heat-fixing roll is formed by electrostatically applying a release layer of fluororesin on the metal pipe, and a charging roll is formed on a polished rubber roll. The developing sleeve is manufactured by precisely polishing the surface of the metal pipe by applying a semiconductive surface layer on the surface of the metal pipe, and subjecting it to blasting or alumite treatment as the surface treatment. However, various surface defects may occur in such a manufacturing process. The types of defects include scratches that occur in the circumferential direction by using a wheel-shaped polishing device in the cutting and polishing processes of metal pipes, foreign substances that occur due to uneven coating material in the coating process, color unevenness, and coating conditions. There are more than a dozen types in total, such as dents and bulges caused by air bubbles due to non-uniformity of the above, foreign matters caused by non-uniformity of the processing material in the surface treatment step, color unevenness, dents and scratches caused by handling.
Since these defects appear as black spots or streaks on the output image surface, appearance inspection is performed to provide a high quality product free of these defects.

As a visual inspection, a visual inspection has been conventionally performed by an inspector. However, in such an appearance inspection, there are problems such as an increase in inspection cost and inspection man-hours, variation in inspection quality, and because the inspection target is a reflective surface, fatigue is severe and it is difficult to secure inspection personnel. It was pointed out.

To solve the above problem, an automatic visual inspection apparatus has been proposed by Japanese Patent Laid-Open No. 5-107197. FIG. 13 shows the configuration of the above-described surface defect automation device. This automated device for surface defects is mounted on a fluorescent lamp 1 that emits white diffused light, a slit 2 that converts the white diffused light of the fluorescent lamp 1 into band-shaped slit light, and a rotatable stage 4, and that An object to be inspected 3 such as a roll component to which slit light is irradiated, a line sensor 6 that receives slit light reflected on the surface of the object to be inspected 3 via a reduction optical system 5, and a line sensor 6 according to the amount of light received. It is configured to include a defect detection device 7 that receives a surface state signal and detects a surface defect of the inspection object 3.

The defect detection circuit 7 receives a slit light from the line sensor 6 and receives an analog level surface state signal output from the line sensor 6 when the inspection object 3 is rotated once while receiving a preset signal. A binarized signal is set according to a threshold value, and the binarized signal is stored in a frame memory to form an image in which the outer peripheral surface of the inspection object 3 is developed in a planar shape.
Then, the area value S is calculated for the defect candidate image on the surface-developed image.
If there are two or more, the device under test 3 is determined as a defective product.

In such a structure, the line sensor is located at the first detection position where the light-receiving field of view of the line sensor 6 coincides with the boundary position H1 of the slit light imaged on the surface of the object 3 to be inspected, and the center position H2 of the slit light. At the second detection position where the light-receiving visual fields of 6 coincide, the main scanning of the line sensor 6 is performed and the surface state signals thereof are output to the defect detection circuit 7 to detect the surface defect of the inspection object 3. Where the first
It is possible to detect defects such as dents and bulges on the surface of the inspection object 3 at the detection position, and defects such as color unevenness, foreign matter, dents, and scratches at the second detection position.

14 (a), (b), and (c) show the principle of detecting a gentle uneven defect on the surface at the first detection position,
If the surface of the DUT 3 is normal as in (a), or
When the inspected object 3 has a color unevenness defect 3A as shown in (c), the projected slit light is reflected in the regular reflection direction A, and the line sensor 6 has a small level at a position deviated from the regular reflection direction A. The reflected light (dotted line) B is received, but as in (b),
If there are irregularities 3B on the surface such as dents and bulges, the reflection direction of the slit light changes slightly, and the boundary portion B of the slit light B
The surface state signal representing the defect signal is output from the line sensor 6 receiving the light.

According to an experiment conducted by the inventor, a position further outside the boundary position of the slit light (hereinafter referred to as a third detection position)
Has been confirmed to be suitable for highly accurately detecting a faint color irregularity defect that is difficult to detect at the second detection position.

In the surface defect inspection method for detecting the surface defect of the object to be inspected by using the relative positional relationship between the slit light and the line sensor, the above-mentioned three kinds of detection positions (first, second, and second) are used. As in 3), the optimum light receiving position differs depending on the type of roll and the form of the defect. For this reason, a defect sample is prepared for each type of roll, and the optimum detection position is obtained in advance by experiments such as measuring scattered light by the defect sample.

As described above, in such a surface defect inspection apparatus, since the detection position of the line sensor is switched according to the type of defect, an optical system consisting of a slit light source and a line sensor is used to secure the detection power. Position calibration is important. Also, change product types, replace light source lamps, etc.
It is required to be able to quickly calibrate to the specified optical system position when various conditions such as vibration and impact near the inspection device change. The optical system is calibrated by using a reduction optical system. (1) Focus adjustment, (2) Light receiving magnification adjustment, (3) Detection position tilt adjustment with respect to slit light, (4) Object to be inspected at detection position There are five items: position adjustment in the axial direction of (1), and (5) position adjustment of the detection position in the circumferential direction of the DUT.

Now, the influence of each of the above-mentioned calibration items on the surface defect detection power will be described. First, if the focus of the reduction optical system of (1) does not match, the peripheral portion of the defective portion on the image will appear blurred, and the size of the defect will appear larger than it actually is. Further, when the light receiving magnification of (2) changes, the size on the image changes for the same defect. In the cases of (1) and (2), it becomes impossible to correctly judge the quality with respect to a predetermined area threshold. Furthermore,
(3) If the detection position is not parallel to the slit light but has an inclination, for example, the right end portion becomes the first detection position, and the left end portion becomes the third detection position, so that the entire inspection object is covered. It will not be possible to obtain uniform detection across. Furthermore,
(4) If the detection position deviates in the axial direction of the slit light, the line sensor will be displaced from the area to be inspected, resulting in inspection failure. The more important point is that (5) the detection position shifts in the circumferential direction of the slit light, and the condition where the second detection position should be the third detection position occurs depending on the type of defect. This inspection method, which takes an optimum detection position by using the above method, has a problem that sufficient detection power cannot be obtained.

Conventionally, as the position calibration of such an optical system, there is, for example, position adjustment of an optical system for reading an original of a copying machine. As shown in JP-A-5-75797, the position is adjusted to the detected position. It is known that a chart is installed, the mark shape on the adjustment chart is read, and the assembling position of an optical system for reading a document of a copying machine is adjusted.

FIG. 15 shows the above adjustment chart, which includes a triangular mark 8 for calculating the position and inclination of the optical system in the sub-scanning direction, a ladder mark 9 for calculating the focus of the optical system, and an optical system. Vertical line mark 1 for calculating the imaging magnification
0A and 10B, and a horizontal line mark 11 indicating the reference of the position of the optical system in the sub-scanning direction are attached. When this adjustment chart is read by the line sensor, a sensor signal waveform as shown in FIG. 16 is output. Here, 8'is a waveform corresponding to the triangular mark 8, 9'is a waveform corresponding to the ladder mark 9, 10A ', 10B' is a waveform corresponding to the vertical line marks 10A and 10B, and 11 'is a horizontal line mark. Waveforms corresponding to 11 are respectively shown, the width of the triangular mark 8 is calculated from the number of pixels of the waveform 8 ', and the position and the inclination of the optical system in the sub-scanning direction are adjusted based on the result, and the maximum voltage of the waveform 9'is adjusted. Adjust the focus of the optical system from the value and the minimum voltage value,
The interval of 10B is calculated from the number of pixels between the waveforms 10A 'and 10B', the imaging magnification of the optical system is adjusted based on the result, and the parallelism of the optical system is adjusted from the minimum voltage value of the waveforms 11 'on both sides. To do.

However, when it is considered that the calibration method of the optical system using the adjustment chart is applied to the slit light receiving system, the adjustment chart is a flat glass material with a mark drawn on the upper surface thereof. It is physically difficult to install the adjustment chart on the surface of the inspection object, which is the inspection position of the slit light receiving method.

Therefore, conventionally, a mark shape is drawn on the surface.
A sample roll for calibration is installed at the inspection position of the DUT.
I am doing it. Specifically, the axial direction of the sample roll
Two scratch marks are provided on the top, and this is the stage of the inspection device.
Set it on the top and read it with a line sensor, and its signal waveform
By observing the characteristics of
There is. FIG. 17 shows the signal waveform of the line sensor, and the X axis is
The position of the sample roll in the axial direction, the Y-axis is the sensor signal
Show each level. 12A and 12B are sample low
Waveform corresponding to two defective parts (scratch marks)
In this signal waveform, the peak of waveform 12A of the defective portion is
Work level V₁And normal surface level V₂From the contrast of
Focus adjustment is performed between the waveforms 12A and 12B of the defective portion.
Measuring the distance L and the position P of the waveform 12A of the defective portion
Adjust the magnification and position in the roll axis direction by
The border of the input light is received, and both
Balance so that the difference ΔV of the part corresponding to the side disappears
The tilt of the detection position is adjusted by taking
After adjustment, the average value V of the sensor waveform of the slit light ₃
Is measured to adjust the position in the roll circumferential direction. Low
Except for the adjustment of the light receiving position in the circumferential direction,
If there is no movement of the mechanical part due to relocation or relocation, readjustment is necessary.
Although it is not necessary, the detected position is
When changing the position, the light receiving position in the roll circumferential direction must be readjusted.
Need to be adjusted.

[0017]

However, according to the conventional method of calibrating the surface defect inspection optical system, the sensor waveform of the sample roll is observed with an oscilloscope or the like, and the waveform position interval or signal corresponding to the scratch mark is observed from the waveform that appears. Since the level is visually read and evaluated, the number of man-hours for the calibration work by the operator is increased, and there is a disadvantage in that the evaluation as a quantitative value varies and reliability is reduced.

The first portion for receiving the boundary portion of the slit light
In the detection at the detection position, the setting range of the light receiving position of the line sensor is narrow, and further, in the detection of the third detection position which receives the outside of the slit light, the range is a dark field portion and the received light amount of the line sensor is low. There is a problem that it is difficult to adjust the position.

Therefore, it is an object of the present invention to reduce the number of man-hours for the calibration work by an operator and to eliminate the variation when evaluating the relative position of the optical system to improve the reliability of the calibration work. Defect inspection optical system calibration method,
And to provide a device.

Another object of the present invention is to provide a method and an apparatus for calibrating a surface defect inspection optical system which can facilitate fine adjustment of a detection position for slit light.

[0021]

SUMMARY OF THE INVENTION In view of the above problems, the present invention reduces the number of man-hours for calibration work by an operator, eliminates variations when evaluating relative positions of optical systems, and improves the reliability of calibration work. In order to improve the size and make it possible to easily make fine adjustments to the detection position with respect to the slit light, a plurality of belt-shaped annular marks having the same size as the roll member and having a larger surface roughness than other regions are included. Prepare a master roll with the reference mark on the surface, place the master roll at the inspection position, irradiate the master roll with slit light, receive light from the slit light irradiation position, and the position in the vicinity thereof, and receive light level And compares the received light level with the reference received light level according to the slit light irradiation position, the boundary position of the slit light irradiation position, and the external position of the slit light irradiation position. The irradiation position, the boundary position, and there is provided a method for calibrating a surface defect inspection optical system so as to calibrate the position of the optical system, and the orientation to external positions.

Further, according to the present invention, a master roll having the same size as the roll member and having a plurality of reference marks on its surface is prepared, the master roll is arranged at an inspection position, and the master roll is irradiated with the slit light. Then, the reflected light is received to calculate the characteristic amount of the plurality of reference marks, and the characteristic amount and the reference value in a predetermined range are compared to calibrate the position and orientation of the optical system with respect to the master roll. The present invention provides a method for calibrating a surface defect inspection optical system as described above.

Further, in order to achieve the above object, the present invention provides
A master roll having the same size as the roll member and having a plurality of reference marks including a belt-shaped annular mark having a larger surface roughness than other regions on the surface, and the roll member or the master roll at the inspection position. An inspection stage to be positioned, a roll member arranged on the inspection stage, or a slit light source for irradiating the master roll with the slit light, and an irradiation position of the slit light when the master roll is irradiated with the slit light from the slit light source. , And a line sensor for detecting the light receiving level from a position in the vicinity thereof, a light receiving level, a slit light irradiation position, a boundary position of the slit light irradiation position, and a reference according to an external position of the slit light irradiation position. By comparing the received light level with the irradiation position, the boundary position, the position of the optical system with respect to the external position, and the position for calibrating the posture and There is provided a calibration apparatus of a surface defect inspection optical system having a energization adjustment means.

Furthermore, according to the present invention, a master roll having the same size as the roll member and having a plurality of reference marks on its surface, an inspection stage for positioning the roll member or the master roll at an inspection position, and an inspection stage A slit light source for irradiating the roll member or the master roll with slit light, and a line sensor for outputting image signals corresponding to a plurality of reference marks when the master roll is irradiated with slit light from the slit light source. And a signal calculation means for calculating the characteristic amount of the plurality of reference marks based on the image signal, and the position of the optical system with respect to the master roll based on the calibration amount obtained by comparing the characteristic amount with the reference value in a predetermined range. And a surface defect inspection optical system calibrating device provided with a position and attitude adjusting means for calibrating the attitude.

[0025]

Operation: It has the same size as the roll member to be inspected,
A master roll having a plurality of fiducial marks including a belt-shaped annular mark having a surface roughness larger than those of other regions is prepared, and the master roll is arranged at the inspection position of the inspection stage. The master roll is irradiated with slit light from a slit light source, and the line sensor receives light from the slit light irradiation position and a position in the vicinity thereof to detect a light reception level. The position and orientation adjusting means compares the received light level with the reference position of the slit light irradiation position, the boundary position of the slit light irradiation position, and the external position of the slit light irradiation position to determine the irradiation position and the boundary. Calibrate the position and orientation of the optical system with respect to position and external position. In addition, a master roll having the same size as the roll member to be inspected and having a plurality of reference marks on the surface is prepared,
This master roll is arranged at the inspection position of the inspection stage. The master roll is irradiated with slit light from a slit light source and the line sensor receives the reflected light, whereby the signal calculation means calculates the characteristic amounts of the plurality of reference marks. Based on the calculation result, the position and orientation adjusting means calibrates the position and orientation of the optical system with respect to the master roll by comparing the characteristic amount with a reference value in a predetermined range.

[0026]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS A method and apparatus for calibrating a surface defect inspection optical system according to the present invention will be described in detail below with reference to the accompanying drawings.

FIG. 1 shows the structure of a surface defect inspection optical system calibrating apparatus according to an embodiment of the present invention. In this surface defect inspection optical system calibration device, the slit light source 13 for emitting slit light from a straight tube fluorescent lamp and the inspected object such as the photosensitive drum, the heat fixing roll, and the charging roll are in the horizontal direction, that is, Inspection stage 14 arranged in the x-axis direction
When the optical system is calibrated, the calibration master roll 16 is arranged on the inspection stage 14 to receive the slit light 15 on its surface, and the slit light 15 reflected on the surface of the calibration master roll 16 is passed through the reduction optical system 17. Line sensor 18 that receives light by receiving light, reduction optical system 17, and line sensor 1
The adjustment mechanism 19 for adjusting the posture and the position of 8 and the surface condition signal according to the analog light receiving level output from the line sensor 18 are input to attach to the defective portion of the inspection object or the calibration master roll 16. A signal extraction circuit 20 for extracting an image signal corresponding to the formed mark, and a signal extraction circuit 20.
An image processing device 21 that stores an image signal output from the device in a frame memory, develops it in a plane, and performs a predetermined calculation at the time of inspection and calibration, and a calculation result in the image processing device 21, a defect image, and the like are displayed. The display unit 22 is provided.

The slit light source 13 is for adjusting the balance of both ends so that the slit light irradiated on the surface of the object to be inspected or the calibration master roll 16 is parallel to the axis, that is, the x axis. The adjusting mechanisms 23A and 23B are included.

The inspection stage 14 includes a pair of drive rolls 2 for holding an inspection object or a calibration master roll 16.
4A and 24B, a drive roll 24A, and by extension, a motor 25 for rotating the inspection target or the calibration master roll 16.

The calibration master roll 16 used is one in which the surface of an aluminum pipe or the like having the same diameter and axial length as the object to be inspected is mirror-finished to an average roughness of about 0.25 μm, and then black anodized. Further, on the surface, as shown in FIG. 2, a linear first reference mark 32 parallel to the axial direction,
Dot-shaped second, third, and fourth fiducial marks 33, 34,
35, strip-shaped fifth and sixth reference (annular) marks 36A and 36B extending in the circumferential direction, and a reflection mark 37 that provides a start signal when rotating.

The first fiducial mark 32 is formed by scribing and has a length of 300 mm, which is the same as that of the roll part to be inspected.
The width is about 0.4 to 1.0 mm. The scribed portion is peeled off from the alumite treatment on the surface, and when the slit light 15 is irradiated, strong scattered light is emitted as reflected light.

The second, third and fourth fiducial marks 33,
34 and 35 are formed by subjecting a center point and two points having a predetermined distance from the center point to both ends in the axial direction at a position different from the first fiducial mark to a cutting hole or the like.

Fifth and sixth fiducial marks 36A, 36B
Are the second, third and fourth fiducial marks 33, 34, 3
5, a matte reflecting surface as a partially roughened surface is continuously formed in a circumferential direction over a predetermined width. Specifically, when the entire surface is anodized, a protective seal is attached only to the fifth and sixth fiducial mark forming portions, and then a matte alumite treatment is performed. In addition, the fifth and sixth reference marks 36A, 3
In 6B, the degree of surface treatment is changed.

The adjusting mechanism section 19 is capable of adjusting its posture in three axis directions orthogonal to each other with the center of the line sensor 18 as a reference, and rotates around a direction parallel to the axis of the line sensor 18 (θ direction). A tilting stage 26 that rotates, and a tilting stage 27 that rotates around a direction parallel to the optical axis (φ direction).
And a rotary stage 28 that rotates around a direction (ψ direction) orthogonal to the axis of the line sensor 18 and the optical axis. Further, in order to adjust the central positions of the reduction optical system 17 and the line sensor 18, the front-rear moving stage 29 that moves in the optical axis direction, that is, the front and rear, and the line sensor 1
It has a lateral movement stage 30 that moves in the direction of the axis of No. 8.

The signal extraction circuit 20 receives a surface state signal corresponding to the analog light receiving level output from the line sensor 18, converts the signal into a ternary value with a predetermined threshold value, and detects the defective portion of the inspection object or the calibration master roll. The three-valued signal representing the first to fifth fiducial marks attached to 16 is taken out.

The image processing section 21 stores the ternary signal output from the signal extraction circuit 20 in a frame memory and develops it in a plane shape. At the time of inspection, the ternary signal is extracted from the image of the defective portion as a geometrical feature quantity of the defective portion. The center position, the area, the projection length in the X and Y directions are calculated, and at the time of calibration, the geometric feature amount necessary for various adjustments is calculated from each mark of the calibration master roll 16.
From this, the present value of each adjustment item is calculated as calibration data.

As shown in FIGS. 3 (a) and 3 (b), the display unit 22 displays the detected value 45 corresponding to each adjustment item in the state of calibration.
A and a predetermined reference value 45B are displayed, and the defective portion 31 is displayed in the state at the time of inspection.

The method of calibrating the surface defect inspection optical system of the present invention will be described below. First, after disposing the calibration master roll 16 between the inspection stage 14, that is, between the drive rolls 24A and 24B, the motor 25 is driven, and
Slit light is emitted from the slit light source 13.

When the motor 25 is driven, the drive roll 24
A rotates, and the calibration master roll 16 rotates in cooperation with the drive roll 24B. Further, since slit light is emitted from the slit light source 13, the reflected light corresponding to the surface state of the calibration master roll 16 reduces the reduction optical system 1.
The light is reduced by 7 and received by the line sensor 18, and the surface state signal of the calibration master roll 16 ((a) in FIG. 4) is output from the line sensor 18.

When the surface state signal is output from the line sensor 18, the signal extraction circuit 20 extracts a ternary signal (FIG. 4 (b)) obtained by ternarizing the surface state signal with a predetermined threshold value. Is output to the image processing unit 21.

When the ternary signal is input, the image processing section 21 generates a planar image of the ternary signal. In FIG. 5, the main scanning direction is the X axis and the sub scanning direction is the Y axis based on the ternary signal.
An image developed in a plane for the axis is shown. Here, 38 is the first reference mark 32 of the calibration master roll 16,
39, 40, and 41 are the second, third, and fourth reference marks 33, 34, and 35 of the calibration master roll 16,
Reference numeral 42A is a reference mark image corresponding to each of the fifth reference marks 36A of the calibration master roll 16, and the reflection mark 37 of the calibration master roll 16 determines the sensor signal acquisition start position. It is reproduced in the position of. In addition, the sixth reference mark 36
When A is provided, a sixth fiducial mark image 42B corresponding to this is formed. When the image developed in a plane shape is generated in this manner, the geometrical feature amount necessary for calibration is calculated from this, and this is calculated as the current value 45 corresponding to each adjustment item.
It is displayed on the display unit 22 as A. The operator displays the display 22
The current value 45A is the reference value 45B while checking the displayed contents
Adjusting mechanism 23 of the slit light source 13 so that
The optical system is calibrated by adjusting each stage of A, 23B and the adjusting mechanism section 19.

Next, the calculation of the geometrical characteristic amount necessary for each adjustment item and the calibration method of the inspection optical system will be described.

(1) Adjustment of parallelism of the line sensor 18 with respect to the calibration master roll 16 In this adjustment, the first reference mark image 38 corresponding to the first reference mark 32 of the above image is used, and the first reference mark is used. Projection lengths XL1 and YL1 of the image 38 in the X and Y directions
Is calculated and the ratio of the projection lengths XL1 and YL1 is calculated, that is, the image 3
An inclination θ of 8 is calculated. The parallelism is an adjustment item necessary for uniformly detecting the entire surface of the roll, and the tilt is adjusted by the tilt stage 27 of the position adjusting mechanism 19 to make the roll parallel.

(2) Adjustment of parallelism of slit light source 13 In this adjustment, the first reference mark image 38 corresponding to the first reference mark 32 is used, and the fifth reference mark 32 of the first reference mark 32 is used.
The reflected light amount level intersecting the reference mark 36A part is calculated, and the parallelism of the slit light source 13 to the calibration master roll 16 is calculated from the difference between the average light amount levels of the two. In order to detect the entire surface of the roll uniformly, it is necessary to make them parallel, and in order to make them parallel, the slit light source 1
The mounting position is adjusted by the adjusting mechanisms 23A and 23B of No. 3. (3) Adjustment of Axial Position of Line Sensor 18 In this adjustment, the second reference mark image 39 corresponding to the second reference mark 33 is used, and the position coordinate X _{1 of} the second reference mark image 39 in the X axis direction is adjusted. To calculate. In order to inspect the entire roll length without omission, it is necessary that there is no deviation between the position coordinate X ₁ and the reference position. In order to eliminate the deviation, the horizontal movement stage 30 of the position adjusting mechanism 19 causes the line sensor 18 to move.
Adjust the axial position of. (4) Adjustment of optical magnification In this adjustment, the third and fourth reference mark images 40 and 41 corresponding to the third and fourth reference marks 34 and 35 are used. X in images 40 and 41
The positions X ₂ and X ₃ in the axial direction are obtained, and from this, the distance (X ₃ −X ₂ ) between the third and fourth reference mark images 40 and 41.
To calculate. Since the distance between the third and fourth reference marks 34 and 35 on the calibration master roll 16 is known, the optical magnification can be obtained by obtaining the distance represented by the number of pixels on the surface developed image. Since the quality of the defect of the inspected object is determined by the area threshold value using the area which is one of the calculated feature amounts, the area value (the number of pixels) calculated for the same defect is made constant. It is necessary that the optical magnification does not change, and the optical path length is adjusted by the front-rear moving stage 29 of the position adjusting mechanism 19 in order to eliminate the change. (5) Focus adjustment of the reduction optical system 17 In this adjustment, the second, third and fourth reference marks 33,
Using the second, third, and fourth fiducial mark images 39, 40, and 41 corresponding to 34 and 35, the second, third, and
And area values of the fourth reference mark images 39, 40, and 41 are calculated. This is because the image of the defective portion becomes larger when the area value used for determining the quality of the defect of the inspection object is out of focus than when it is in focus.

Here, the reduction optical system will be described with reference to FIGS. 6 and 7.
The relationship between the focus position of the system 17 and the area value on the image will be described.
It In FIG. 6, the focus position of the reduction optical system 17 is at the in-focus position.
Shows the sensor signal with and without focus.
You When in the in-focus position, the sensor waveform 43 has a peak level.
High bell, signal threshold V_thBinarized signal width W cut out in
₁Is minimum and the sensor wave is out of focus.
The peak level of the shape 44 is low, and the signal threshold V_thCut out with
Binary signal width W₂Is the binarized signal width W of the in-focus position₁To
It will be larger than that. Further, FIG. 7 shows a region R of the in-focus position.₁When
Area R out of focus₂, R₃And signal peak value
(Dotted line) and the relationship between the binarized area value (solid line)
And the lens is in the focus area R₁If
High signal peak value (sensor waveform peak level)
Threshold V_thThe area value binarized by is the minimum, and the focus position
Deviated area R₂, R₃Signal peak if
Value (sensor waveform peak level) is low and signal threshold V_thso
The binarized area value is the focus range R₁If you are in
Grows bigger. Therefore, focus can be measured from the area value.
You can The area threshold is used to judge the quality of defects in the inspection object.
Since it is done by, it is necessary that there is no focus shift,
Focus is adjusted by the focusing ring of the reduction optical system 17.
To do. In addition, the central part and the periphery of the calibration master roll 16
The optimum focus position differs depending on the optical path length
Then, by trial and error, find the position where both focus points are almost in focus.
It (6) Adjustment of the detection position of the line sensor 18 In this adjustment, the fifth and sixth reference marks 36A and 36B are marked.
The corresponding fifth and sixth fiducial mark images 42A, 42B.
Then, the average pixel density of these mark images is calculated.
In the defect inspection of the object to be inspected, depending on the type of defect,
The detection position of the line sensor 18 with respect to the incident light
It is necessary that the calculated average pixel density is a predetermined first,
If the pixel density of each of the second and third detection positions matches the reference value.
The tilt stage 26 of the position adjustment mechanism
Adjust the position of the in-sensor 18 in the roll circumferential direction
It

FIG. 8 shows a reference mark detection method for adjusting the position of the line sensor 18 to the first to third detection positions which are used according to the type of defect of the inspection object. When the slit light 15 is projected on, the slit light 15 becomes blurred by the matting treatment by the fifth and sixth reference marks 36A and 36B having large surface roughness, and the width becomes wide in the circumferential direction. In particular, the sixth reference mark 36 having a surface roughness larger than that of the fifth reference mark 36A.
In A, the projection width of the slit light 15 is widest. Adjustment to each detection position of the line sensor 18 is performed by receiving the slit light 15 having a wide projection width of the fifth and sixth reference marks 36A and 36B, and the reflected light amount level of the slit light 15 from the first to third detection positions. This is performed by adjusting the tilt stage 27 of the position adjusting mechanism 19 so that the value becomes a value.

FIG. 9 shows the relationship between the detection position of the line sensor 18 and the sensor signal level (reflected light amount level).
Is the sensor signal waveform of the slit light at the fifth reference mark 36A, 47 is the sensor signal waveform of the slit light at the sixth reference mark 36B, and 48 is the sensor signal waveform of the slit light of the inspection object having no surface roughness region. Is. Here, considering the detection position of the line sensor 18 on the inspection object, the second detection position is a region R where the sensor signal level is high.
₄ , the third detection position is the region R ₅ where the sensor signal level is low, and the second detection position is the region R where the sensor signal level is low.
_The region R ₆ is slightly higher than ₅ . The detection position ranges corresponding to these areas are the second detection position range S ₁ , the third detection position range S ₂ , and the first detection position range S ₃ , respectively. As is apparent from the figure, when the sensor signal waveform 48 of the slit light of the inspection object is seen, the sensor signal level remains unchanged even if the light receiving position changes in the third detection position range S ₂ and the first detection position range S ₃ . It hardly changed, and it was difficult to adjust the light receiving position based on the sensor signal level, that is, the reflected light amount level. On the other hand, looking at the sensor signal waveform 47 of the sixth reference mark 36B having the largest surface roughness, the level difference of the sensor signal level region R ₇ in the third detection position range S ₂ is larger than that of R _5. It will be much larger. As a result, by using the sixth reference mark 36B, fine position adjustment of the line sensor 18 at the third detection position becomes easy. Further, according to the sensor signal waveform 46 of the fifth reference mark 36A with a reduced degree of matting,
In the first detection position range S ₃ , the level difference in the region R ₈ is much larger than the level difference in the sensor signal level region R ₆ in the first detection position region S ₃ of the inspection object. Therefore,
The use of the fifth reference mark 36A facilitates fine position adjustment of the line sensor 18 at the first detection position. By partially providing the matte surface on the calibration master roll 18 as described above, position adjustment for a plurality of types of detection positions can be performed with high accuracy and easily.

Next, the procedure of various adjustments during calibration will be described with reference to the flowchart of FIG.

If a plurality of adjustment items are adjusted without considering the predetermined order, it becomes difficult to match all the items with the reference value. This is because some of the adjustment items are premised on that other items match the reference value, such as adjustment of the light receiving position. Therefore, in order to perform these adjustments efficiently, the calibration work is carried out according to the steps shown in FIG.

First, the second, third, and fourth reference marks 33, 34, so that the characteristic amount of each reference mark of the calibration master roll 16 can be calculated without changing the detection position of the line sensor 18. Focus adjustment is performed by using 35 (step 1). Next, the third and fourth fiducial marks 3
The optical magnification of the reduction optical system 17 is adjusted using 4 and 35 (step 2). When the optical magnification is adjusted, the focus position is slightly shifted, so focus adjustment is performed again (step 3). Then, using the second fiducial mark 33, the line sensor 18
The position is adjusted in the x-axis direction (step 4). After that, the line sensor 18 is formed using the first reference mark 32.
The parallelism of (step 5). Furthermore, the fifth and sixth
The slit light source 1 using the reference marks 36A and 36B of
The parallelism of 3 is adjusted (step 6). Finally, the light receiving position of the line sensor 18 with respect to the position of the slit light is selected, and the line sensor 18 is adjusted from the first to third detection positions (step 7). With this calibration procedure, for example,
When there is a change in the detection position, only step 7 needs to be performed, and the slit light source and the line sensor 1 can be used.
8 is replaced or the device is relocated from step 1.

In the embodiment described above, the linear master first fiducial mark parallel to the roll axis and the dot-shaped second, third and fourth fiducial marks are used as the calibration master roll 18. The present invention is not limited to this, and it is possible to calculate the feature amount when the image is developed in a plane and to calculate the tilt and position of the line sensor 18 and the slit light source 13, and for example, in FIG. As shown, the first fiducial mark 32 may be formed by a broken line parallel to the roll axis, and the second, third, and fourth fiducial marks 33, 34, 35 are continuous in the circumferential direction. It may be composed of straight lines. Further, in the above embodiments, the fifth and sixth reference marks 36A and 36B used for adjusting the light receiving position for the slit light are formed by black matte alumite treatment on the surface of the aluminum pipe, but the present invention is not limited to this. However, the specularity of the surface should be made lower than that of the inspection roll surface by polishing the surface with a polishing roller of a lower count, and the change in the reflected light quantity distribution of the slit light irradiated on the surface should be gradual. Good.

In the above embodiment, the operator adjusts each adjusting mechanism based on the detected value 45A and the reference value 45B displayed on the display unit 22, but the driving circuit drives each adjusting mechanism. The optical system may be automatically calibrated.

FIG. 12 shows the construction of the above-described automatic calibration device for the optical system. The same parts as those in the first embodiment are designated by the same reference numerals and symbols, and thus duplicated description will be omitted.
It corresponds to the storage table 51 that stores the amount of movement according to the difference between the arbitrary position of the optical system for each adjustment item and the reference position of each adjustment item, and the detection value of each adjustment item calculated by the image processing unit 21. An adjustment amount calculation unit 51 that calculates the movement amount from the storage table 51 and a drive unit 52 that drives the position adjustment mechanism 19 based on the adjustment amount calculated by the adjustment amount calculation unit 51 are configured. In addition, the position adjustment mechanism 1
9 as well as the adjusting mechanism 23A, 2 for the slit light source 13
3B or the focus ring of the reduction optical system 17 may be automatically adjusted.

[0054]

As described above, according to the method and apparatus for calibrating the surface defect inspection optical system of the present invention, a partial roughness ring mark is formed on the master roll, and the amount of reflected light at this portion is measured to form a slit. Since the detection position of the line sensor with respect to the light is adjusted, the change in the reflected light amount distribution of the slit light irradiated on the surface can be moderated, and the delicate adjustment of the detection position of the line sensor with respect to the slit light can be easily performed, and , Can be done accurately.

Further, a master roll having a plurality of reference marks provided on its surface is arranged on the inspection stage of the surface defect inspection optical system, and the relative positional relationship of the optical system is represented from the read image of each reference mark of the master roll. Since the feature amount is calculated, it is possible to reduce the number of man-hours for the calibration work by the operator and improve the reliability of the calibration work.

[Brief description of drawings]

FIG. 1 is an explanatory diagram showing a first embodiment of the present invention.

FIG. 2 is an explanatory diagram showing a calibration master roll according to an embodiment.

FIG. 3 is an explanatory diagram showing a display state of a display unit according to an embodiment.

FIG. 4 is an explanatory diagram showing output signals of a signal extraction circuit and an image processing unit according to an embodiment.

FIG. 5 is an explanatory view showing an image in which the surface of the calibration master roll according to the embodiment is developed in a planar shape.

FIG. 6 is a graph showing a relationship between a focus degree and a sensor signal level according to an embodiment.

FIG. 7 is a graph showing a relationship between a focus degree, an area value, and a sensor signal peak value according to an embodiment.

FIG. 8 is an explanatory diagram showing a time of adjusting a light receiving position of the line sensor according to the embodiment.

FIG. 9 is a graph showing the relationship between the sensor signal level and the light receiving position of the line sensor in the roll circumferential direction according to the embodiment.

FIG. 10 is a flowchart showing a procedure of various adjustments when calibrating an optical system according to an embodiment.

FIG. 11 is an explanatory diagram showing a second embodiment of the present invention.

FIG. 12 is an explanatory diagram showing a third embodiment of the present invention.

FIG. 13 is an explanatory diagram showing a surface defect inspection apparatus.

FIG. 14 is an explanatory diagram showing slit light reflection characteristics when inspecting a surface defect of an inspection object.

FIG. 15 is an explanatory diagram showing an optical system calibration method for a conventional surface defect inspection apparatus.

FIG. 16 is an explanatory diagram showing an optical system calibration method for a conventional surface defect inspection apparatus.

FIG. 17 is an explanatory diagram showing an optical system calibration method of another conventional surface defect inspection apparatus.

[Explanation of symbols]

 1 Fluorescent lamp 2 Slit 3 Object to be inspected 4 Stage 5 Reduction optical system 6 Line sensor 7 Defect detection device 8 Triangle mark 8'Waveform corresponding to triangle mark 9 Ladder mark 9'Waveform corresponding to ladder mark 10A, 10B Vertical line mark 10A ', 10B' Waveform corresponding to vertical line mark 11 Horizontal line mark 11 'Waveform corresponding to horizontal line mark 12A, 12B Defect part 13 Slit light source 14 Inspection stage 15 Slit light 16 Calibration master roll 17 Reduction optical system 18 Line sensor 19 Position Adjusting Mechanism 20 Signal Extracting Circuit 21 Image Processing Section 22 Display Section 23A, 23B Adjusting Mechanism 24A, 24B Drive Roll 25 Motor 26 Tilt Stage 27 Tilt Stage 28 Rotating Stage 29 Front-Back Moving Stage 30 Lateral Moving Stage 31 Defect 32 First The standard of 33 33 2nd fiducial mark 34 3rd fiducial mark 35 4th fiducial mark 36A 5th fiducial mark 36B 6th fiducial mark 37 Reflection mark 38 1st fiducial mark image 39 2nd fiducial mark image 40 40th fiducial mark 3 reference mark image 41 4th reference mark image 42A 5th reference mark 42B 6th reference mark image 43, 44 Sensor waveform 45A Present value 45B Reference value 50 Storage table 51 Adjustment amount calculation unit 52 Drive unit

Claims (4)

[Claims] 1. A position of an optical system for irradiating a roll member arranged at an inspection position with slit light and receiving reflected light of the slit light from the roll member to detect a surface defect of the roll member, and In a calibration method of a surface defect inspection optical system for calibrating a posture, a plurality of reference marks including a belt-shaped annular mark having the same size as the roll member and having a larger surface roughness than other regions are provided on the surface. Prepare a master roll, arrange the master roll at the inspection position, irradiate the master roll with the slit light, detect the light receiving level by receiving light from the slit light irradiation position and a position in the vicinity thereof. A reference according to the light receiving level, the irradiation position of the slit light, the boundary position of the irradiation position of the slit light, and the external position of the irradiation position of the slit light. Compares the received light level, the irradiation position, said boundary position, and the position of the optical system with respect to the external position, and the calibration method for the surface defects optical system, characterized in that calibrating the posture. 2. A position of an optical system for irradiating a roll member arranged at an inspection position with slit light and receiving reflected light of the slit light from the roll member to detect a surface defect of the roll member, and In the calibration method of the surface defect inspection optical system for calibrating the posture, a master roll having the same size as the roll member and having a plurality of reference marks on the surface is prepared, and the master roll is arranged at the inspection position. The master roll is irradiated with the slit light and the reflected light is received to calculate the characteristic amount of the plurality of reference marks, and the optical amount is calculated by comparing the characteristic amount with a reference value in a predetermined range. A method of calibrating a surface defect optical system, characterized in that the position and orientation of the system with respect to the master roll are calibrated. 3. A position of an optical system for irradiating a roll member arranged at an inspection position with slit light, receiving reflected light of the slit light from the roll member, and detecting a surface defect of the roll member, and In a surface defect inspection optical system calibrating device for calibrating a posture, a plurality of reference marks including a belt-shaped annular mark having the same size as the roll member and having a larger surface roughness than other regions are provided on the surface. A master roll, an inspection stage that positions the roll member or the master roll at the inspection position, a roll member that is arranged on the inspection stage, or a slit light source that irradiates the master roll with the slit light. When the slit light is emitted from the slit light source to the master roll, an irradiation position of the slit light,
And a line sensor for detecting a light receiving level from a position in the vicinity thereof, the light receiving level, an irradiation position of the slit light, a boundary position of the irradiation position of the slit light, and an external position of the irradiation position of the slit light. And a position and attitude adjusting means for calibrating the position and attitude of the optical system with respect to the irradiation position, the boundary position, and the external position by comparing with a reference light receiving level according to the above. Calibration device for surface defect optical system. 4. A position of an optical system for irradiating a roll member arranged at an inspection position with slit light, receiving reflected light of the slit light from the roll member, and detecting a surface defect of the roll member, and In a calibration device for a surface defect inspection optical system for calibrating a posture, a master roll having the same size as the roll member and having a plurality of reference marks on the surface, the roll member, or the master roll at the inspection position. A slit light source for irradiating the slit light to the roll member or the master roll arranged on the inspection stage, and the slit light to the master roll from the slit light source. At this time, a line sensor that outputs an image signal corresponding to the plurality of reference marks, and the line sensor based on the image signal Signal calculating means for calculating the characteristic amount of a plurality of reference marks, and the position and orientation of the optical system with respect to the master roll based on a calibration amount obtained by comparing the characteristic amount with a reference value in a predetermined range. A calibration device for a surface defect inspection optical system, comprising a position and attitude adjusting means for calibration.