JPH04169807A - Inspecting apparatus of surface flaw - Google Patents

Abstract

PURPOSE:To improve the precision in determination of good and bad articles by providing a signal level threshold and an area threshold corresponding to the kind of a surface flaw of a test specimen and by determining the surface flaw on the basis of the two thresholds. CONSTITUTION:When it is assumed that there are a dint, a short transverse line flaw and a long transverse line flaw in the surface of a test specimen, a comparator 13a compares output signals 10 thereof with a threshold 11 and detects all the flaws, and a binary-coded image memory 14a stores binary-coded signals thereof. Meanwhile, a comparator 13b compares them with a threshold 12 and detects only a bruise flaw having a high peak value, and a memory 14b stores a binary-coded signal thereof. Area computing elements 15a and 15b compute the area of the flaw from these binary- coded signals and comparators 17a and 17b compare it with area thresholds 16a and 16b. For the threshold 16a, herein, a value smaller than an area S3 and larger than S4 is selected, while a value smaller than an area S2 is selected for the threshold 16b. As the result, the comparators 17a and 17b judge only the long transverse line flaw and the bruise flaw as flaws affecting the quality of a picture respectively, while they do not judge the short traverse line flaw as such a flaw as the above. Accordingly, the possibility of determining a good article as a bad one is reduced.

Description

Detailed Description of the Invention (Industrial Application Field) The present invention relates to a surface flaw inspection device, and in particular to a surface flaw inspection device that can accurately determine whether an object to be inspected is a good product or a defective product. Regarding equipment.

(Prior Art) Scratches formed on the surface of an object to be inspected, such as a cylindrical drum of a copying machine, pose a serious problem. Therefore, apparatuses for inspecting such surface flaws on objects to be inspected have been developed.

Conventional surface flaw inspection equipment projects a slit light from, for example, a halogen light source onto the surface of an object to be inspected, uses a line sensor to receive scattered light from flaws on the surface of the object to be inspected, and uses a signal processing circuit to detect the scattered light. It is processed to determine whether there are any scratches or not.

The signal processing circuit generally uses binarization processing, which allows easy determination, so that it can be used for high-speed, continuous inspection processes.

An example of a conventional signal processing circuit will be explained with reference to FIG.

In the figure, the line sensor 29 receives scattered light from scratches on the surface of the object to be inspected, as described above. For example, the line sensor 29 outputs a signal a having a waveform as shown in FIG. 6(1). In the figure, the reason why the signal output is small at the periphery and large at the center is due to the aberration of the lens that focuses the scattered light onto the Linesen-Fj-29, and the urgent pulses indicate scratches.

Signal a is converted by an A/D converter 30 into, for example, an 8-bit digital signal. The waveform diagram of the output signal of the A/D converter 30 is as shown in FIG. 6(2) (b). In FIG. 5, the signal output for each pixel expressed in 8 bits, that is, 256 gradations is expressed as (Do, Di, D
2, □-==, Dn). 32 is a threshold value corresponding to each pixel of the signal output. This threshold (TO
, Tl, T2, . . . , Tn) are determined for each pixel according to the level of signal output due to the aberration of the lens.

The signal output 31 and the threshold value 32 are compared by a comparator 33 and binarized. The binarized signal C has a waveform C as shown in FIG. 6, and a city with a large amount of scattered light is detected.

34 is a binarized image memory, in which the comparison results by the comparator 33 are sequentially stored. This binarized image memory 34 stores the comparison result between the data 31 obtained by converting information of a large number of lines on the surface of the object to be inspected read by the line sensor 29 into multivalued data and the threshold value 32. Ru. For this reason,
The binarized image memory 34 stores the binarized signal of a predetermined area of the surface of the object to be inspected.

Reference numeral 35 denotes an area calculating section, which calculates the area of each scratch from the signal representing the scratches stored in the binarized image memory 34. The area of the determined scratch is compared with a predetermined area threshold 36 by a comparator 37, and scratches exceeding this area threshold are displayed on the display section 38, while scratches not exceeding the area threshold are ignored.

As described above, an object to be inspected with a flaw larger than a certain area can be determined to be a defective product, and an object to be inspected with a small flaw can be determined to be a good product.

For example, in the case of a photoconductor drum for a copying machine, the determination as to whether it is a defective product is based on whether or not there is a problem with the quality of the copied image, and the problem with the image quality is almost determined by the area of the scratch. Become. The conventional device described above determines that the product is good if the scratch area is smaller than a predetermined threshold.
By selecting this threshold value to be the minimum value of the area that is considered to affect the image quality, it can be used to inspect for scratches on the photoreceptor drum of a copying machine.

(Problem to be Solved by the Invention) However, when the inventor of the present invention used the flaw inspection device to determine whether the photoreceptor drums of a copying machine were good or defective, some of the products determined to be defective were: It was found that there was a mix of non-defective products that did not affect the quality of the copied images.

Therefore, when conventional flaw detection devices are used, good products are mistakenly judged as defective, so in order to improve the yield rate, it is necessary to further visually inspect the inspected objects that have been judged as defective to select good products. There was a problem in that the film had to be stained and a re-inspection process was required.

SUMMARY OF THE INVENTION An object of the present invention is to provide a surface flaw inspection device that eliminates the problems of the conventional device and improves the accuracy of determining good and defective products.

(Means and Effects for Solving the Problems) In order to achieve the above object, the present invention provides a plurality of signal level thresholds set depending on the type of surface flaw, and a method for detecting scattered light from the surface of an object to be inspected. A plurality of two-channel sensors that compare the line sensor output to be photoelectrically converted with each of the signal level thresholds and output a binary signal.
a digitization means, a plurality of area calculation units that calculate an area of a surface flaw from the binarized signal, and a plurality of area calculation units according to the type of the surface flaw.

a plurality of area thresholds provided in the area calculation section, a plurality of areas calculated by the area calculation section and each of the area thresholds, and a determination means for determining whether the product is good or defective;
It is characterized by the fact that

According to the present invention, a signal level threshold and an area threshold are provided according to the types of surface flaws on the object to be inspected, and surface flaws can be determined using both thresholds. It is possible to determine whether the inspected object is good or defective.

Therefore, it is possible to improve the accuracy of determining whether the inspected object is a good product or a defective product.

(Example) The present invention will be described in detail below with reference to the drawings.

First, the present invention will be explained in detail.

Scratches that occur on the surface of an object to be inspected, such as a photoreceptor drum of a copying machine, can be broadly classified into dent scratches and horizontal line scratches. FIG. 7(a) shows a signal output waveform when a line sensor detects a dent, and FIG. 7(b) shows a signal output waveform when a horizontal line flaw is detected.

In addition, 51 in the figure (a) indicates the area of the cross section of the dent at the signal output level a1, and S2 indicates the area of the cross section of the dent at the signal output level a2. Further, 53 in the diagram (b) indicates the cross-sectional area of the horizontal line flaw at the signal output level a1.

As is clear from FIG. 7(a), the characteristics of the dents are that the peak p1 of the signal output is high, but the area of the scratches is small. For example, the peak p1 of the signal output is 200 or more for 256 gradations, and the area s1 is approximately 6 pixels on one side of one pixel.
Assuming 0 μm, there are approximately 50 pixels.

On the other hand, the characteristic of surface scratches is that the peak p2 of the signal output is low, but the area of the scratches is large, as is clear from FIG. 2(b). For example, the peak of signal output p2 is 2
It is about 40 to 50 for 56 gradations, and the area S3 is 1
00 pixels or more.

As a result of the experiments conducted by the inventor Mori, the things that affect the image quality of the copied image are scratches such as dents, which have a high signal output peak p1 and have a certain area S2 at the signal output level a2. It was also found that the scratches had a large area s3 (s3 > sl) at the signal output level a1, like long horizontal line scratches. Furthermore, it was found that even if there is a horizontal line scratch on the surface Rs1 or more, if the scratch area is less than 53, it does not significantly affect the image quality and can be determined to be a good product.

The present invention has been made based on the above experimental results, and one embodiment will be described in detail below.

FIG. 2 is a conceptual diagram of the entire apparatus of this embodiment, in which 1 is a support section for the object to be inspected, 2 is a motor that rotates the support section 1, and 3 is a conceptual diagram of the entire device.
4 is a light source device that generates slit light; 4 is a line sensor;
5 is a condensing lens that focuses the scattered light from the object to be inspected onto the line sensor 4, and 6 is an A/D converter that converts the output signal of the line sensor 4 into, for example, an 8-bit digital signal.

The output of the A/D converter 6 is led to a surface flaw determination circuit that determines whether a flaw on the surface of the object to be inspected is a flaw that affects the quality of the copied image.

FIG. 1 is a circuit diagram of one embodiment of this surface flaw determination circuit.

In the figure, 10 is a signal output of scattered light converted by the A/D converter 6 into, for example, 8 bits and 256 gradations. Further, 11 is the same level as the threshold value 32 in FIG. 5, and the end values TI, T2, . . .
・Tn is about 20, and the value at the center is about 40, that is, about 20 to 40. This level is determined by measuring the scattered light of a photoreceptor drum with no scratches on its surface, called the master drum, and multiplying the measured value by n times (for example, 2
~3 times) as the threshold value. Note that this n changes depending on the type of flaw to be detected.

On the other hand, 12 is a threshold value higher than the threshold value 11. The threshold value 12 is set, for example, to about twice the threshold value 11.

FIG. 3 is a diagram showing the concept of the threshold value 11.12. In this figure, al indicates the threshold value 11, and a2 indicates the threshold value 12.
It shows.

As shown in the figure, there are dents and scratches on the surface of the object to be inspected.
If there are short horizontal line scratches or long horizontal line scratches, the first comparator 13a detects all of these scratches, and the binarized signals thereof are stored in the first binarized image memory 14a. On the other hand, the second comparator 13b detects only the dents with high peak values, and the binarized signal thereof is stored in the first binarized image memory 14b. These binarized image memories 14a and 14b store the scattered light of each line obtained by irradiating the object to be inspected with the slit light.
Since the digitized signals are sequentially accumulated, data of flaws on the entire surface of the object to be inspected is accumulated in the binary image memories 14a and 14b. Note that since the total surface area of the object to be inspected is large, the data of scratches on the entire surface is stored in the binarized image memory 14.
If it does not fall within the ranges a and 14b, the test may be conducted in multiple batches.

The binarized signals stored in the first and second binarized memories 14a and 14b are stored in the first and second area calculation units 15a and 1.
5b, the area of the scratch is calculated. Then, the comparators 17a and 17b compare the first area threshold 16a1 with the second area threshold 16b.

Here, the first area threshold 16a is smaller than the area S3,
A value larger than s4 is selected. In addition, the second area threshold 1
As 6b, a value smaller than the area S2 by [2] is selected.

As a result, according to the first comparison 517a, it is the long horizontal line flaw in FIG. 3 that is determined to be a flaw that affects image quality, and the dent flaw and short horizontal line flaw are no longer detected. On the other hand, the second comparator 17b determines that the scratches that affect image quality are the dents shown in FIG. 3, and long horizontal line scratches and short horizontal line scratches are not detected.

The signal outputs of the first and second comparators 17a and 17b are sent to a display section 19 via an OR circuit 18 and displayed.

As is clear from the above description, according to this embodiment, short horizontal scratches that do not affect image quality are not judged as scratches, so it is possible to reduce the possibility that a non-defective product will be judged as a defective product. can.

Next, a second embodiment of the present invention will be described.

In this embodiment, it is possible to detect a peeling of the sensitive material, which is a different type of flaw from the above. A photoreceptor dropout is a spot of the photoreceptor coated on the photoreceptor drum surface, and the projected light is regularly reflected by the mirror-finished drum base material (for example, an aluminum vibrator). , almost no scattered light is generated.

Figure 4 shows the situation.1. This is a diagram. In the figure, 4
1.41a shows the signal output of scattered light generated by a scratch, and 42 shows the signal output of a surface without scratches. Also, 43.
43a is the iN output of the portion where the photosensitive material is removed; 44 is a threshold for detecting the removal of the photosensitive material at a level slightly lower than the signal output 42;
5 indicates a threshold for surface flaw detection similar to the threshold 11 of the first embodiment.

In this embodiment, threshold values 11 and 1. of FIG. The threshold value 45 is used as the threshold value 12, the threshold value 44 is used as the threshold value 12, and the comparator 13
Reverse the tens and - in b. Also, the area threshold 16b of the circuit of the series of thresholds 44,! :, l sets the minimum value of the area where the image quality is affected by the removal of the sensitive material.

By doing so, the circuit having the threshold value 11 can detect surface flaws, and the circuit having the threshold value 12 can detect the removal of the sensitive material.

Incidentally, since even a small omission of the sensitive material causes a deterioration in image quality, if the comparator 13b determines that there is a omission of the sensitive material, it may be determined that the product is defective without comparing with the area threshold value.

Furthermore, a circuit for detecting the omission of the sensitive material may be added to the circuit shown in FIG.

(Effects of the Invention) As is clear from the above description, according to the present invention, small scratches that are not a problem in actual use can be detected and eliminated, so that the false alarm rate of erroneously determining good products as defective products can be reduced. can.

In addition, when the object to be inspected is a photoreceptor drum of a copying machine, it is possible to detect the dropout of the photoreceptor, eliminating the possibility of incorrectly determining a defective product as a good product, and improving the process for inspecting the photoreceptor dropout. can be reduced.

[Brief explanation of drawings]

Fig. 1 is a block diagram of an embodiment of the present invention, Fig. 2 is an external perspective view of the device of this embodiment, and Fig. 3 shows the signal output level and cross-sectional area according to the type of surface flaw on the object to be inspected. 4 is a signal output waveform diagram of a photoreceptor drum having a photosensitive material dropout, FIG. 5 is a block diagram of a conventional device, FIG. 6 is a signal waveform diagram of the main part of FIG. 5, and FIG. The figure is a diagram illustrating the signal output level and cross-sectional area depending on the type of surface flaw on the object to be inspected. 10...Signal output from the inspected object, 11.12...
Threshold value, 13a, 13b--Comparator, 14a114b-
...Binarized image memory, 15a, 15b...Area calculation section, 16a, 16b...Area threshold, 17a, 17b.
... Comparator, 19... Display section. Agent: Patent attorney Hiraki and 1 other person Figure 2 Figure 3

Claims (1)

[Claims] (1) A surface flaw inspection device that projects light onto the surface of an object to be inspected, collects scattered light from the surface, performs photoelectric conversion using a line sensor, and then tests the presence or absence of surface flaws on the surface. a plurality of signal level thresholds provided according to the type of scratch; a plurality of binarization means that compares the line sensor output with each of the signal level thresholds and outputs a binarized signal; and the binary signal. a plurality of area calculation units that calculate the area of a surface flaw from a surface flaw signal, a plurality of area thresholds provided according to the type of the surface flaw, and a plurality of areas calculated by the area calculation unit and the area threshold value. 1. A surface flaw inspection device characterized by comprising a determination means for comparing each product and determining whether the product is good or defective.