JPH0247491Y2 - - Google Patents

Description

[Detailed Description of the Invention] An object of the present invention is to provide a surface defect inspection device that can easily and reliably detect defects such as dents and scratches on the surface of rods and the like.

Conventionally, to detect surface defects such as rods,
After wiping the surface of the rod with a cloth or the like to completely remove dust from the surface, the surface is illuminated with a light source and the reflected light is visually observed by the inspector. If there are no defects on the rod surface, the reflected light looks uniformly dark, and if there is a defect, the larger the area and depth, the more diffusely the light is reflected and locally shines brightly, making it possible to identify the defect.

However, according to the above-mentioned conventional method, the magnitude of the reflected light intensity is determined simply by the inspector's intuition.
It requires skill, the test results may vary depending on the inspector, and the test results are unstable.Also, it is practically difficult to completely remove dust from the rod surface, and it is difficult to remove dust from the surface. If dust remains, there is a drawback that light may be diffusely reflected by the dust and may be mistaken for a defect. (This defect can be prevented to some extent by wiping the surface with a cloth or changing the angle of the reflected light, but it still requires experience and skill, and the test results are unstable.) This invention calculates the ratio of the number of gratings of two types of brightness in the measurement area of the inspected member under two different reflected light conditions, and when both ratios are larger than a certain value, By determining that there is a defect,
The purpose of the present invention is to provide a surface defect inspection device that eliminates the instability of inspection and eliminates the above-mentioned conventional drawbacks, and its configuration includes an illumination means for illuminating the surface of a member to be inspected, and an illumination means for illuminating the surface of the inspected member. an imaging means for capturing an image of the surface upon incident light; a comparison means for comparing and identifying a plurality of divisions defining a measurement area of the surface according to at least two types of brightness; and divisions of the two types of brightness. The two ratios calculated by the calculation means are both larger than a certain value in two cases in which the conditions of the reflected light from the same measurement area to the imaging means are different from the calculation means that calculates the number of objects and calculates the ratio. and determining means for determining that there is a defect on the surface of the inspected member when the inspection target member has a defect.

Next, each example will be explained.

1 and 2 are a side view and a plan view, respectively, of an embodiment of the surface defect inspection apparatus according to the present invention.
In the figure, the surface defect inspection device 1 has an actuator 2 on the floor, and the actuator 2 has arrows A and B.
As will be described later, a rod 4 as a member to be inspected is placed in a concave portion on the upper surface of a rod receiver 3 fixed to the upper end of a column 2a which is slidable in any direction. As shown in FIG. 3, the rod 4 has a shaft portion 4b at each end of a round bar-shaped portion to be inspected 4a.

Actuators 5, 6, 7, and 8 are similarly arranged at the four corners of the floor surface, and have pillars 5a, 6a, 7a, and 8a that are slidable in the directions of arrows A and B, respectively.

9 and 10 are a pair of lower rails, the lower rail 9 is connected to the upper ends of the columns 5a and 7a, the lower rail 10 is connected to the upper ends of the columns 6a and 8a, and both rails 9 and 1 are connected to the upper ends of the columns 5a and 7a.
0 is parallel.

11, 12, 13, and 14 are actuators arranged at predetermined height positions, respectively.
3 has a column 13a that is slidable in the directions of arrows C and D, and the actuator 14 is slidable in the directions of arrows E and F.
A column 14a that is slidable in any direction is fixed to the actuator 13.

15 and 16 are a pair of upper rails, the upper rail 15
is attached to the actuator 11, and the upper rail 16 is similarly attached to the actuator 12,
They can be slid in the directions of arrows C and D, respectively. still,
Upper rails 15, 16 (cam part 15 at the upper right end of each
a, 16a) are spaced apart from each other above the lower rails 9, 10, and are arranged parallel to each other.

21 is an ITV camera, and 22 is a pair of illumination devices integrally arranged on both sides of the camera 21, which are fixed to the tip of the support column 13a (the mode is shown in FIG. 3) and can be slid in the directions of arrows C and D. It is.

A limit switch 23 is fixed to the upright rod 24 and is switched by a cam portion 15a of the upper rail 15, as will be described later.

25, 26, and 27 are belt conveyors, respectively. The conveyor 25 is for transporting rods before inspection, and is controlled by the computer 24 to run intermittently. Conveyors 26 and 27 are for conveying good and defective rods after inspection, respectively, and always travel in the direction of the arrow.

28 is a computer, actuator 5~
8, 11 to 14, input signal processing for the limit switch 23, and defect determination processing for the rod 4.

Next, the operation of the surface defect inspection apparatus 1 will be explained. In advance, in FIG. 1, the actuator 11,
12, the upper rails 15 and 16 are in a position where they have been retracted and slid by a predetermined distance in the direction of arrow C from the position shown in the figure, and the actuators 5, 6 and 7, 8
As a result of the respective operations, the lower rails 9 and 10 are in a state where the right side in FIG. The pillar 2 is moved by the operation of the actuator 2.
a is substantially fully extended in the direction of arrow A, and the rod receiver 3 is at its upper limit of movement.

Here, the belt conveyor 25 moves the rod 4
is transported and dropped, and both end shaft portions 4b are placed in contact with the lower rails 9 and 10, respectively. Lower rail 9, 10
Since it is tilted as shown above, rod 4 is shown in Figure 1.
It rolls to the left in FIG. 2, fits into the recess on the upper surface of the rod receiver 3, and comes to rest.

Next, the actuators 11 and 12 operate, and the upper rails 15 and 16 protrude and slide in the direction of arrow D, reaching the positions shown in FIGS. 1 and 2, and the cam portion 15a
contacts the limit switch 23 and operates it.
As a result, the operation of the actuators 11 and 12 is stopped, and the sliding of the upper rails 15 and 16 is also stopped.
At this time, the ITV camera 21 is located directly above the center of the rod 4 of the rod receiver 3. Subsequently, the actuators 5 to 8 operate, and the columns 5a to 8a slide upward in the direction of arrow A, raising the lower rails 9 and 10 upward while keeping them in a horizontal state, and lowering the lower rail 9.
Both end shaft portions 4b of the rod 4 are pressed and held between the upper rail 15 and between the lower rail 10 and the upper rail 16, respectively. Thereafter, the actuator 2 operates and the rod receiver 3 is moved downward and separated to a position where rotation of the rod 4 is not hindered.

Next, the actuators 11 and 12 operate to move the upper rails 15 and 16 synchronously in the direction of arrow C by the dimension x.
minute (x is the dimension required for the cam portion 15a to operate the limit switch 23).During this period, the rod 4 rotates counterclockwise in FIG. 1, causing rolling friction with the upper rails 15 and 16. The center of rod 4 moves in the same direction by a distance of x/2.
(Generally, when a rotating body is held between two parallel surfaces under pressure and the rotating body is moved by rolling and friction by sliding one side in the direction in which the surfaces extend, the center of the rotating body will move when the one surface slides by a dimension x (The distance is x/2) From this state, the defect inspection on the surface of the rod 4 is started. , it is assumed that a measurement area 30 ₁ (also having a width m) is particularly set in the observation area 29 (having a width l in the four-axis direction of the rod) by the ITV camera 21.
This measurement area 30 ₁ is divided into 64 grids, for example, by dividing it vertically and horizontally into 8 equal parts. Therefore, the measurement area 30 ₁ is illuminated by the illumination device 22, the ITV camera 21 images the measurement area 30 ₁ , and sends the image signal to the image memory 31 as shown in FIG.

The image memory 31 compares the brightness of the 64 grids individually, classifies them into groups of four levels, for example, the brightest part, the brightest part, the darkest part, and the darkest part, and outputs them as digital signals as shown in FIG. The data is sent to the data processing determination device 32. The data processing/judgment device 32 specifically counts the number of brightest parts and the number of bright parts of each group, respectively, and calculates the ratio α (=number of brightest parts/number of bright parts). (The more brightest parts there are, the higher the presence of defects such as dents and scratches, and the higher the degree of dust, so the value of α is larger.) Next, the actuators 11 and 12 operate again and the upper The rails 15 and 16 slide in the direction of the arrow C, and the rod 4 rolls in the same direction and moves by y/ ₂ . During this time, the ITV camera 21 also moves in the same direction by y/ ₂ due to the operation of the actuator 13. The state shown in FIG. 5A changes to the state shown in FIG. 5B.
At this time, as shown in FIG. 4, the measurement area 30 ₁ in the observation area 29 of the ITV camera 21 has moved to the position indicated by the two-dot chain line, and from the measurement area 30 ₁ to the ITV camera 2
The conditions for the reflected light incident on 1 are different from the previous one. Here again, the number of brightest parts and the number of bright parts among the four-stage groups of the 64 gratings in the measurement area 30 ₁ are counted, respectively, and the ratio β is calculated. here,
When the values of α and β are both larger than a certain value, the computer 28 determines that there is a defect in the measurement area 30 ₁ as will be described later. The reason for taking two values α and β is that, for example, if α is only larger than a certain value, it is not a defect but dust (noise: in this case, noise is a small scratch that humans cannot recognize). be)
may only exist across each grid, but if the rod 4 is rotated by a certain angle to change the reflection conditions, the diffuse reflection will weaken in the case of dust and β will become smaller than a certain value. This makes it possible to selectively determine only the presence of defects such as dents and scratches while ignoring the presence of dust.

Here, in FIG. 5B, the ITV camera 21 simultaneously images the front measurement area 30 ₂ and its α value is determined, and then the rod 4 is rotated and the measurement area 30 2 is imaged in the same manner as in the above case. The β value of ₂ is determined, and the values of α and β of each measurement area 30 ₃ , 30 ₄ , .

FIG. 6 shows the measurement areas 30 ₁ , 30 ₂ , 30 ₃ ,
The number of brightest parts (A ₁ , A ₂ ) and the number of bright parts (B ₁ ,
The results of measuring B ₂ ) are shown.

Next, when the measurement of the width m of the rod 4 for one rotation is completed as described above, the actuator 14 is operated and the support column 14a is retracted and slid in the direction of the arrow E. As a result, the ITV camera 21 slides in the same direction, that is, in the direction of the four axes of the rod, by a distance m, and corresponds to an area having a width m adjacent to the measurement area for one revolution completed above.

Next, the actuators 11 and 12 operate,
This time, the upper rails 15 and 16 successively protrude and slide in the direction of arrow D, and the rod 4 sequentially moves clockwise in FIG. While moving in the direction, the measurement of the adjacent measurement area is completed.

Subsequently, in the same manner, the rod 4 and the ITV camera 21 repeatedly move back and forth in the directions of arrows C and D, and the camera 21 sequentially moves repeatedly in the four axis directions of the rod to measure the entire circumference of the member to be inspected 4a of the rod 4. Complete. The number of reciprocating movements in the directions of arrows C and D and the number of movements of the rod in the four-axis directions are input in advance into the computer 28 according to the outer diameter and length of the rod 4, and are controlled by the computer 28.

When the measurement of the entire surface of the rod 4 is completed, the computer 28 determines the presence or absence of defects. If there is no defect, the actuators 7 and 8 operate to move the columns 7a and 8a downward, and the lower rails 9 and 10 are tilted so that the left side in FIG. 1 is lowered. Therefore, Rod 4
are rotated by their own weight and placed on the belt conveyor 26 for transporting non-defective products, where they are sorted into non-defective groups.

In addition, if there is a defect, the actuator 5,
6 is operated, the columns 5a and 6a move downward, and the right side portions of the lower rails 9 and 10 in FIG. 1 are tilted downward. Therefore, the rods 4 are rotated by their own weight and placed on the belt conveyor 27 for transporting defective products, where they are sorted into groups of defective products.

After the above operation is completed, all actuators 5
8, 11 to 14 return to their initial state, the belt conveyor 25 travels only the distance necessary to supply the next rod 4, and the new uninspected rod 4 is received in the same way as in the above case. 3, and defect inspection is performed in the same manner.

According to the defect inspection method of Rod 4 above, the ratio of the number of gratings of two types of brightness (brightest part, bright part) in the measurement area is determined in two ways with different reflected light conditions.
α and β are determined by the ITV camera 21 and attached circuit, and when both α and β are larger than a certain value, the computer 28 determines that there is a defect, so everything is done digitally and automatically without manual effort. It can be determined and inspected visually, and the inspection results are extremely good and stable. Furthermore, since the measurement is carried out under two different reflected light conditions, there is no risk of misidentifying dust as a defect, and the inspection accuracy can be further improved.

Further, according to the device 1, both shaft portions 4 of the rod 4
Since the defect inspection is carried out by holding and rotating the parts b between the upper and lower rails 15, 9; 16, 10, respectively, the parts to be tested 4a The entire surface can be easily inspected, and there is no risk of scratches occurring on the inspected portion 4a during chucking.

Furthermore, by moving the ITV camera 21 at half the sliding speed of the upper rails 15 and 16, a good inspection can be easily performed with the camera 21 always aligned with the central axis of the rod 4.

Also, after the inspection of the rod 4 is completed, by simply tilting the lower rails 9 and 10 in either direction,
It is convenient to group 4 rods into good and defective products.

FIG. 7 and FIG. 8 are an overall plan view and a partially cutaway side view of the main parts of another embodiment of the above-mentioned surface defect inspection apparatus, respectively. In Fig. 7, surface defect inspection device 4
1, rods 44 are sequentially supplied from a rod supply section 45 to rod storage bags 43 provided at equal pitches on the outer surface of a belt conveyor 42 that runs endlessly, travel in the direction of the arrow, and are inspected for defects in an inspection unit 46. After that, it is stored in the rod storage section 47. The rod 44 is for a shock absorber, for example, and has a threaded portion 44b at the top of a round bar-shaped portion to be inspected 44a in FIG. 8, and conical convex portions 44c for centering at the upper and lower ends, respectively.

The inspection unit 46 has an actuator 48 in a fixed position as shown in FIG.
The conical hole 48c of the hard resin non-magnetic material 48b at the tip of the piston rod 48a, which can freely slide in the directions of arrows A and B, is connected to the conical convex portion 44c at the lower end of the rod 44.
It fits concentrically to the top and pushes upward.

49 is a rotor that also serves as an electromagnet, is wound with a coil 50, and has a central hole 49a and a conical hole 4 at its bottom.
9b, the coil 50 has needle electrodes 51a and 51b connected to both ends, respectively, and a gear 52 screwed into the upper end shaft portion 49c. 53 is a pulse motor, a gear 54 coaxial with this gear meshes with the gear 52, and a rotary encoder 5 connected to the pulse motor 53.
5 detects the rotation speed of the motor 53 and inputs it to the computer 61.

62 is a current supply unit that supplies a driving current to the rotor 49 which also serves as an electromagnet, and as shown in FIGS. 8 and 9, ring-shaped conductors 64 and 65 ( The conductors 64 and 65 are respectively connected to the anode and cathode of a DC power source 66. Also, the center hole 3 of the disc 63
6a is loosely fitted into the shaft portion 49c of the rotor 49, and a support column 67a of the fixed actuator 67 is fitted and fixed into the hole 63c of the projecting plate portion 63b of the disc 63. Therefore, as the support column 67a slides in the directions of arrows A and B, the unit 62 also slides back and forth in the same direction, and the pair of electrodes 51a and 51b are inserted into the grooves 64, respectively.
a, 65a, and when inserted, the groove 64
The rotor 49 is rotating in contact with the inside of a and 65a.
The coil 50 of is continuously excited.

Reference numeral 68 indicates an ITV camera, and reference numeral 69 indicates a pair of illumination devices, which are fixed to the tip of a support 70a of an actuator 70 that is slidable in the directions of arrows A and B.

71 is a marker, marker actuator 7
2, it can be slid freely in the directions of arrows C and D.

Next, the operation of the device 41 will be explained. First, the rod 44 stored in the bag 43 of the belt conveyor 42 is connected to the piston rod 4 of the actuator 48.
When reaching the axis of 8a, the conveyor 42 stops. Then, the actuator 48 operates, and the piston rod 48a moves upward in the direction of arrow A, lifting the rod 44 upward. Thus, the threaded portion 44b of the rod 44 is inserted into the hole 49a of the rotor 49, the upper end surface of the portion to be inspected 44c abuts the lower end surface of the rotor 49, and the upper and lower conical convex portions 44c are inserted into the conical hole 49b of the rotor 49 and the piston rod 48a, respectively.
is inserted into the conical hole 48c of the non-magnetic part 48b,
The rotor 49 and the piston rod 48a are held exactly concentrically, resulting in the state shown in FIG.

Next, the actuator 67 operates, and the column 67
The current supply unit 62 moves downward in the direction of the arrow B through the arrow A, and the needle-shaped electrodes 51a and 51a of the rotor 49 are relatively connected to each other.
51b are grooves 64a and 65a of conductors 64 and 65, respectively.
8, and comes into contact with it, resulting in the state shown in FIG. Therefore, from the anode of the DC power supply 66 to the conductor 64 and the electrode 5
1a, the coil 50, the electrode 51b, and the conductor 65, a circuit returning to the cathode of the DC power supply 66 is closed, and current flows through the coil 50. Therefore, rotor 4
9 becomes an electromagnet and firmly attracts the rod 44, allowing it to rotate stably as a unit, after which the DC power supply 6
6 is continuously powered on. At this time
The ITV camera 68 is the inspected part 44a of the rod 44.
In FIG. 8, for example, the lower end width dimension l is observed as shown in FIG. , find the number of bright parts and find its α value.

Next, the pulse motor 53 operates and the gears 54,
52, the integral rotor 49, rod 4
4 by a predetermined angle to change the reflected light conditions and measure the measurement area again to determine the β value. Thus, it is determined that there is a defect on the surface of the rod 44 only when α and β are both greater than a certain value. Similarly, one round of the width m of the rod 44 is measured.

Next, the ITV camera 68 slides in the direction of the arrow A by a distance m by the operation of the actuator 70, and then the rod 44 rotates in the same manner to measure the next round. 4
4a are all measured and tested.

Thus, the computer 61 determines whether or not the rod 44 is defective, and if determined to be defective, the marker actuator 72 is operated to slide the marker 71 in the direction of arrow C to contact the rod 44. Then mark the defective product with red paint or the like, and after marking, the marker 71 is retracted and slid.

Next, the current supply unit 62 is moved upward in the direction of arrow A by the operation of the actuator 67, and the electrodes 51a and 51b are relatively moved into the grooves 64a and 65, respectively.
a, the excitation of the coil 50 of the rotor 49 is released, and the attraction of the rod 44 is released. If an alternating current is applied to the coil 50 at this time, the residual magnetism of the rotor 49 disappears quickly, and the attraction of the rod 44 can be quickly released.

Next, the actuator 48 is operated to move the piston rod 48a downward, and the rod 44 is moved downward by its own weight and returned to the storage bag 43. Furthermore, another electromagnet may be provided below the rod 44, and this electromagnet may be used to suck and slide the rod 44 downward after the inspection is completed, and return it to the storage bag 43.
This method allows the rod 44 to be placed horizontally for inspection.

Here, the ITV camera 68 and the lighting device 69 are also operated by the actuator 7 according to instructions from the computer 61.
0 returns to the initial setting position.

According to the device 41, the rod 44 is connected to the rotor 4.
Since the rod 44 is attracted and fixed by electromagnet 9 and rotated, it is easier to work without the need to change the rod 44 compared to the conventional method in which the rod was changed and chucked, and it also avoids scratches when chucking. There is no fear, and since the rod 44 is rotated at the same position, the space is small, and after the inspection is completed, the rod 44 can be moved down by its own weight by simply moving the piston rod 48a down, and the storage bag 43 can be moved down.
It can be stored inside the machine, reducing the number of work hours.

In the above embodiments, the configurations of the current supply unit 62 and the needle electrodes 51a, 51b are not limited to these.
If space is available, a normal slip-ring type may be used.

In each of the above embodiments, the members to be inspected are all round rods, but they are not limited to this, and may be flat plates.
In this case, both values of α and β may be determined by sequentially moving the flat plate in the direction in which the flat plate extends.

As described above, according to the surface defect inspection apparatus of the present invention, the comparison means compares and identifies a plurality of gratings in the measurement area of the inspected member into gratings of at least two types of brightness, and the calculation means identifies the two types of brightness. The ratio of the number of brightness lattices is calculated under two different reflected light conditions as α and β, and the determination means determines that there is a defect when both α and β are larger than a certain value. All work such as determining the thickness, calculating α and β, and determining defects can be performed digitally and automatically, and stable inspection results can be obtained regardless of the inspector's skill level and inspection accuracy can be improved. In addition, since it is determined that there is a defect when the two values of α and β are both greater than a certain value, only true defects such as dents and scratches can be correctly detected without fear of mistaking dust etc. as defects. It has features such as being able to further improve inspection accuracy.

[Brief explanation of drawings]

1 and 2 are a side view and a plan view, respectively, of an embodiment of the surface defect inspection device according to the present invention, FIG. 3 is a front view of the main parts of the device, and FIG. Figures 5A and 5B are diagrams showing the inspection status by the above apparatus with changes in the position of the inspected member, and Figure 6 is a diagram showing the measurement area of the inspected member. FIGS. 7 and 8 are diagrams showing the number of brightest areas and changes in the number of bright areas for each area under different reflection conditions, respectively, and FIGS. FIG. 9 is a partially cutaway side view and a bottom view of the current supply unit of the above device. 1,41...Surface defect inspection device, 2,5-8,
11-14, 48, 67, 70, 72... Actuator, 4, 44... Rod, 9, 10... Lower rail, 15, 16... Upper rail, 21, 68...
...ITV camera, 22,69...Lighting device, 25
~27,42... Conveyor, 28,61... Computer, 29... Observation area, 30 ₁ , 30 ₂ ...
Measurement area, 31... Image memory, 32... Data processing/judgment device, 43... Rod storage bag, 46...
Inspection unit, 49... Rotor, 50... Coil, 51a, 51b, 64, 65... Electrode, 53
...Pulse motor, 62...Current supply unit,
63...Insulating disc, 66...DC power supply.

Claims (1)

[Claims for Utility Model Registration] 1. Illumination means for illuminating the surface of a member to be inspected, imaging means for capturing an image of the surface by receiving reflected light from the surface, and defining a measurement area on the surface. a comparison means for separately comparing and identifying a plurality of sections with at least two types of brightness, a calculation means for calculating the number of sections each having the two types of brightness and calculating a ratio thereof, and an imaging means from the same measurement area. and determining means for determining that there is a defect on the surface of the inspected member when the two ratios calculated by the calculating means are both greater than a certain value in two modes in which the conditions of the reflected light are different. A surface defect inspection device. 2. The illumination means and the imaging means are integrally configured with each other, and by displacing and driving at least one of the illumination means, the imaging means, and the member to be inspected relative to the other, two different modes of the reflected light can be obtained. Claim No. 1 for Utility Model Registration consists of a structure further provided with a driving means for causing the
The surface defect inspection device described in Section 1.