JPS6356761A - Shape inspecting instrument - Google Patents

Abstract

PURPOSE:To shorten an arithmetic processing time necessary to detect an additional defect and a lacking defect by executing the logical operation of a brightness inverted image and an image to be inspected after the necessary quantity of the image of a normal object to be inspected is expanded or reduced. CONSTITUTION:First, after the necessary quantity of the image of a normal object to be inspected is expanded, the AND of a first comparing image 4 obtained by brightness-inverting and an image to be inspected 1 is obtained by an AND circuit 6 and an additional defect 2a is inspected by a bright pattern in an AND image 9. After the necessary quantity of the image of the normal object to be inspected is reduced, the OR of a second comparing image 5 obtained by the brightness-inverting and the image to be inspected image 1 is obtained by an OR computing element 7 and a lacking defect 3a is detected by the bright pattern in an OR image 11. The exclusive OR of the AND image 9 and the OR image 11 is obtained by an exclusive OR computing element 8, and then, both an additional defect 2b and the lacking defect 3a are displayed by a dark pattern on one picture 12.

Description

[Detailed Description of the Invention] [Object of the Invention] (Industrial Application Field) The present invention relates to a shape inspection device for inspecting the quality of object shapes such as printed circuits, IC mask patterns, printed characters and marks, etc. .

(Prior Art) A conventional shape inspection device for inspecting the quality of an object shape is generally configured as shown in FIG.

In FIG. 2, 21 is an object to be inspected (object to be inspected) consisting of, for example, the printed character "T", and on the illustrated printed character rTJ,
Compared to the normal rTJ character pattern, additional printing defects 22 and missing printing defects 23 occur.

The shape inspection device detects this additional and missing print 2
2.23 on the image storage to inspect the quality of the shape of the object to be inspected 21, and is equipped with a hi B device 24 such as an ITV camera and an inspection section 25, as well as a diagram for displaying images. It is equipped with an optional display.

The inspection section 25 is composed of an image calculator 26 and the like, and the inspection section 25 also includes a printed character "T" which is the object 21 to be inspected.
A first comparison image @ 27 in which the image of a normal pattern of `` is expanded by a required amount of 8'' and a second comparison image 28 in which an image of a similar normal pattern is expanded by a required amount are prepared in advance.

rfy in the first comparison image 27; the contour line of the bulge pattern is the detection limit line for detecting the additional printing defect 22, and the contour line of the contraction pattern in the second comparison image 28 is the detection limit line for detecting the additional printing defect 22; This is a detection limit line for detecting printing defects 23.

Then, the object to be inspected 21 is transported by the image head 24, and the inspected image and the first and second comparison images 1127.28 are calculated and compared in the inspection section 25, and the first comparison image is The quality of the shape of the object to be inspected 21 is inspected by detecting an additional defect protruding from the expansion pattern in 27 or a missing defect in the contraction pattern in the second comparison image 28.

FIG. 3 shows a first conventional example showing the shape inspection device in more detail (Japanese Patent Application Laid-Open No. 57-120807).

Note that the warm image device and the like are not shown in FIG. 3 J3 and FIG. 4 in the dredging description.

In FIG. 3, reference numeral 31 is an image to be inspected obtained by imaging an object to be inspected similar to the printed character "T" in FIG. (bright), and the background portion 31b is composed of black pixels (dark). Reference numerals 32 and 33 indicate additional defects and missing defects that appear in the image to be inspected 31, respectively.

In addition, each of the first and second comparison images 27.2 prepared in advance
8, similarly to the image to be inspected 31, the expansion pattern 27a and the contraction pattern 28a are made up of white pixels, and the background parts 2ab and 28b are made up of black pixels.

34 and 35 are the first and second OR operators, respectively;
6.37 are the first and second exclusive OR operators, respectively, and 38 is the third OR operator.

Then, the first comparison image 27 and the image to be inspected 31 are ORed in the first OR operator 34 to obtain the OR pattern 39a, and the first exclusive logic oil (one unit 36 , both images 39 of the logical sum pattern 39a and the first comparison image 27 are added,
Defect 32a is detected as a bright pattern.

On the other hand, the second comparison image 28 and the image to be inspected 31 are subjected to a logical OR operation in the second logical sum operator 35, and the logical sum pattern 41a is incremented by 1''1, and the second 11j!I!!
Exclusive -I! of both images 41 of the disjunction pattern 41a and the image to be inspected 31 using the disjunctive operation oil 7! I! A sum operation is performed and the missing defect 33a is detected as a bright pattern.

In the third logical sum operator 38, a logical sum operation is performed on both images 42 of the additional defect 32a and an image @43 of the missing defect 33a, and the additional defect 32a and the missing defect 33a are displayed on one screen 44. The quality of the shape of the object to be inspected is inspected using the additional defects 32a and the missing defects 33a displayed on one side 44.

By the way, each operation such as the above-mentioned logical sum operation and exclusive logical sum operation between images is executed by calculating between each white pixel and black pixel forming each image.

Therefore, when one image is composed of, for example, 512×512 pixels, approximately 256,000 pixel operations are required for one operation between the images, and the amount of X kl becomes a considerable amount.

However, in the first conventional example, the first and second logical summation units 34, 35, the first and second 1) altruistic logic function units 36, 37, and the third logical summation unit 34, 35, 38, and a total of five calculations are required between the images during inspection, which increases the length of time required for calculation processing during inspection, resulting in a low inspection efficiency. .

Next, FIG. 4 shows a second conventional example of a shape inspection device (MARK CHECKERMODEL2).
01. In FIG. 4, parts that are the same or equivalent to those shown in FIG. 3 are designated by the same reference numerals, and redundant explanation will be omitted.

In this conventional example, as the first comparison image 45 prepared in advance, the expansion pattern 45a is composed of black pixels,
The background portion 45b made up of white pixels is used.

46 is an inverter, and 47 and 48 are first and second AND operators, respectively.

Then, the first comparison image 45 and the image to be inspected @31 are subjected to a logical product operation in the first logical product operator 47, and the additional defect 32 is
a is detected as a bright pattern.

On the other hand, the inverter 46 inverts the brightness and darkness of the image 31 to be inspected, and the second AND operator 48 performs an AND operation on the four inverted images and the second comparison side @28 to remove the missing defect 33a. Detected as a bright pattern.

The logical sum operator 38 further calculates the image @42 of the additional defect 32a.
A logical sum operation is performed on the image 43 of the missing defect 33a and the additional defect 32a and the missing defect 33a are displayed on one double-sided screen 44, and the additional defect 32a displayed on this one screen 44 is
The quality of the shape of the object to be inspected is inspected using the missing defect 33a.

However, in the second conventional example, the inverter 46
, first and second AND operations 347 and 48, and an OR operation unit 38, and the number of operations required for image processing during inspection is four times in total. For this reason, this second conventional example also has the problem that the time required for arithmetic processing during inspection is long and the inspection efficiency is low.

<Problems to be Solved by the Invention) The first conventional example requires 5 times to process images during inspection, and the second conventional example requires 4 times. There were problems in that the time required was long and the inspection efficiency was low.

This invention was made based on the above-mentioned circumstances, and it is an object of the present invention to provide a shape inspection device that requires less time for arithmetic processing during inspection and has high inspection efficiency.

[Structure of the Invention] (Means for Solving the Problems) In order to solve the above problems, the present invention detects additional defects and missing defects that appear in an image to be inspected obtained by imaging an object to be inspected. An apparatus for shape inspection, which detects additional defects by performing an AND operation between a first comparison image obtained by expanding an image of a normal object to be inspected by a required amount and inverting the brightness and darkness of the above-mentioned image to be inspected. a defect detection means, and a missing defect detection means for detecting a missing defect by performing an OR operation on the above-mentioned image to be inspected and a second comparison image obtained by shrinking the image of a normal object to be inspected by a required zero and avoiding contrast inversion. The gist is to have the following.

(Function) The first and second comparison images are images of a normal object to be inspected, which are expanded or contracted by the required amount, and then both brightness and darkness are reversed. Additional defects are detected by performing a logical product operation on the first comparison image and the image to be inspected.

Furthermore, the missing defect detection means performs an AND operation on the second comparison image and the image to be inspected to detect missing defects.

Additional defects and missing defects are detected through two image calculation processes, and the quality of the shape of the object to be inspected is inspected.

(Example) Hereinafter, an embodiment of the present invention will be described based on FIG.

In FIG. 1, 1 is an image to be inspected obtained by YQI'lx an object to be inspected similar to the printed characters rTJ in FIG. The portion 1b is made up of white pixels (bright), and the background portion 1b is made up of black pixels (dark). 2.3 is the image to be inspected 1, respectively.
These are addition defects and omission defects that appeared in .

4 is a first comparison image having an expansion pattern 4a;
Reference numeral 5 indicates a second comparison image to the right of the contraction pattern 5a, and both comparison images 4.5 are prepared in advance prior to the inspection.

The expansion pattern 4a is created by expanding both peaks of the object to be inspected which has a normal shape by a required amount (7), and the contraction pattern 5a is created by contracting the image of the object to be inspected which has a normal shape by the required amount. 17.

Here, regarding the contrast relationship between the pattern part and the background part, as in the above-mentioned image to be inspected 1, the character pattern 1a part is made up of white pixels (pixel), and the background part 1b is made up of black pixels (pixel part).
Assuming that the brightness relationship in a normal image is that the expansion pattern 4a and the contraction pattern 5a are composed of black pixels (dark), the first and second comparison images 4.5 both have the expansion pattern 4a and the contraction pattern 5a composed of black pixels (dark). and each back inkstone part 4b, 5b is a white pixel (
The image is composed of bright images (bright), which is an inversion of the brightness relationship of the normal image.

6 is an AND operator, and this AND operator 6 allows the first
An additional defect detector stage is configured to perform an AND operation on the comparison image 4 and the image to be inspected 1 to detect additional defects.

7 is an OR operator, and this OR operator 7 allows the second
Missing defect means is configured to perform a logical sum operation on the comparison image 5 and the image to be inspected 1 to detect missing defects.

8 (It is an exclusive OR operator.

Next, the effect will be explained.

At the time of inspection, the first comparison image 4 and the image to be inspected 1 are ANDed by the AND operation unit 6, and the additional defect 2a is detected as a bright pattern in the AND image 9.

On the other hand, the second comparison image 5 and the image to be inspected 1g with - working oil 0 device 7! A logical sum operation of 1 is performed, and the missing defect 3a is detected as a bright pattern in the logical sum image 11.

Therefore, by checking the number of white pixels included in the logical sum image 9, it is possible to detect the presence or absence of the additional defect 2a and the degree of the defect, and by checking the number of black pixels contained in the logical sum image 11. The presence or absence of the missing defect 3a and the extent of the defect can be detected.

The quality of the shape of the object to be inspected is inspected based on the presence/absence a3 of the additional defect 2a and the missing defect 3a and their degree.

In this way, by the logical product image 9 and the logical sum image 11,
The additional defect 2a and the missing defect 3a are detected, but when the exclusive OR operator 8 performs the exclusive OR of the logical 10 image 9a3 and the logical sum image 11, the additional defect 2b and the missing defect are detected. The defect 3a is displayed as a dark pattern on both sides 12, and the quality of the shape of the object to be inspected can be inspected on both sides 12.

As described above, in this embodiment, the additional defect 2a and the missing defect 3 ah<
Detected and performed? The time required for 1e processing can be shortened.

[Effects of the Invention] As explained above, according to the configuration of the present invention, the first and second comparison images are both bright and bright after the images of the normal subject to be examined are dilated or contracted as required. It is assumed that the image is dark-inverted, and during inspection, additional defects are detected by performing an AND operation between the first comparison image and the image to be inspected. Further, missing defects are detected by calculating the logical product Q between the second comparison image and the image to be inspected. Therefore, during inspection, 11 defects and missing defects are detected with a small number of image processing times, and the quality of the shape of the Pi inspection object is inspected, so the time required for calculation processing during inspection is reduced. This has the advantage of shortening the time and increasing inspection efficiency.

[Brief explanation of drawings]

Fig. 1 is a configuration diagram showing an example of the shape inspection device according to the present invention together with each processed image, etc., Fig. 2 is a configuration diagram conceptually showing a conventional shape inspection device, and Fig. 3 is a configuration diagram showing an example of the shape inspection device according to the present invention. The first conventional example of inspection equipment! 2!11′! l′! FIG. 4 is a block diagram showing the second conventional example together with each processed image. 1: image to be inspected, 2: additional defect, 3: missing defect, 4: first comparison image, 4a: expansion pattern, 52 second comparison image, 5a: contraction pattern,
6: AND operator, 7: OR operator, 8: EXCLUSIVE OR operator.

Claims (1)

[Claims] An apparatus for performing shape inspection by detecting additional defects and missing defects that appear in an image to be inspected obtained by imaging an object to be inspected, the device expanding the image of a normal object to be inspected by a required amount. additional defect detection means for detecting additional defects by performing an AND operation between the first comparative image whose brightness and darkness have been inverted and the above-mentioned image to be inspected; A shape inspection apparatus comprising a missing defect detection means for detecting a missing defect by performing a logical sum operation on the second comparison image and the image to be inspected.