CA2033359C - Method of matching patterns and apparatus therefor - Google Patents

Method of matching patterns and apparatus therefor

Info

Publication number
CA2033359C
CA2033359C CA002033359A CA2033359A CA2033359C CA 2033359 C CA2033359 C CA 2033359C CA 002033359 A CA002033359 A CA 002033359A CA 2033359 A CA2033359 A CA 2033359A CA 2033359 C CA2033359 C CA 2033359C
Authority
CA
Canada
Prior art keywords
outline
recognized
data
master
image
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Expired - Fee Related
Application number
CA002033359A
Other languages
French (fr)
Other versions
CA2033359A1 (en
Inventor
Masami Nishio
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Futec Inc
Original Assignee
Futec Inc
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Futec Inc filed Critical Futec Inc
Publication of CA2033359A1 publication Critical patent/CA2033359A1/en
Application granted granted Critical
Publication of CA2033359C publication Critical patent/CA2033359C/en
Anticipated expiration legal-status Critical
Expired - Fee Related legal-status Critical Current

Links

Classifications

    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06TIMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00Image analysis
    • G06T7/0002Inspection of images, e.g. flaw detection
    • G06T7/0004Industrial image inspection
    • G06T7/001Industrial image inspection using an image reference approach
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06VIMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V10/00Arrangements for image or video recognition or understanding
    • G06V10/40Extraction of image or video features
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06VIMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V10/00Arrangements for image or video recognition or understanding
    • G06V10/70Arrangements for image or video recognition or understanding using pattern recognition or machine learning
    • G06V10/74Image or video pattern matching; Proximity measures in feature spaces
    • G06V10/75Organisation of the matching processes, e.g. simultaneous or sequential comparisons of image or video features; Coarse-fine approaches, e.g. multi-scale approaches; using context analysis; Selection of dictionaries
    • G06V10/751Comparing pixel values or logical combinations thereof, or feature values having positional relevance, e.g. template matching
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06TIMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2207/00Indexing scheme for image analysis or image enhancement
    • G06T2207/30Subject of image; Context of image processing
    • G06T2207/30108Industrial image inspection
    • G06T2207/30164Workpiece; Machine component

Abstract

A method matching patterns includes the steps of optically scanning a master image and a to-be-recognized image and outputting master image data and to-be-recognized image data, extracting master outline data representing an outline of the master image from the master image data, extracting to-be-recognized outline data representing an outline of a to-be-recognized image from the to-be-recognized image data, performing an enlargement process for the to-be-recognized outline data to enlarge the to-be-recognized outline, thereby forming to-be-recognized outline data, and collating the master outline data with the to-be-recognized outline data, and if a portion of the master outline projects from the to-be-recognized outline, determining that the to-be-recognized image has a short-defect indicating an omission of the image.

Description

- 1 - 20~3359 The present invention relates to a method of matching patterns and an apparatus therefor used in a checking system for checking the quality of offset sheet-fed printing or screen process printing performed on a molded product.
A technique of checking the quality of a printed pattern or the like performed on a product by means of pattern matching is conventionally known, and this tech-nique is performed as follows.
First, a product determined to be a good product by eye inspection or the like is prepared, and a to-be-checked object such as a printed pattern formed on the product is imaged by a television camera. An image signal from the camera is signal-processed, and outline data of the to-be-checked object is extracted. Enlarged outline data obtained by performing an enlargement pro-cess for the outline in the vertical and horizontal directions is stored as master pattern data in a memory.
A slight error based on variations in to-be-checked objects or conveyance precision of products can be allowed by this enlargement process. Thereafter, a to-be-checked object of a to-be-checked product is imaged by the television camera. Outline data is extracted from an image signal corresponding to the to-be-checked object by an outline extraction circuit and compared with the master pattern data stored in the memory, thereby performing pattern matching. By such ~L
2(~333~9 a conventional method of matching patterns, information for determining the quality of a to-be-checked object such as a printed pattern is obtained.
Defects encountered in a printed pattern are roughly classified into a short-defect and a splashing defect. In the short-defect, a portion of a printed pattern is omitted or becomes faint. In the splashing defect, splashed ink, dirt, or foreign matters are adhered on a portion other than a predetermined printed portion.
The above conventional technique is suitably used in detection of splashing defect information. Since, however, this conventional technique is based on the master pattern obtained by performing an enlargement process for the outline and an omitted portion of a printed pattern may be included in the master pattern, the technique is unsuitable to recognize short-defect information.
It is an object of the present invention to provide a method of matching patterns, which allows recognition of both a short-defect and a splashing defect.
According to the method of matching patterns of the present invention, image data of a prepared master is fetched in an outline extraction circuit, and outline data of the image is extracted. Master outline data obtained by the extraction is stored in a memory, and outline data of an image of a to-be-recognized object is - 3 - 2~3359 extracted by the outline extraction circuit. A pattern enlargement process is performed for the extracted outline data to form a to-be-recognized pattern. This to-be-recognized pattern data and the master outline data are collated with each other.
According to the present invention, in order to detect both a short-defect and a splashing defect, image data of a prepared master is fetched in the outline extraction circuit, and outline data of the image is extracted. Master outline data obtained by the extrac-tion is stored in the memory, and outline data of an image of a to-be-recognized object is extracted by the outline extraction circuit. This extracted outline data is enlargement-processed to form a first to-be-recognized pattern. Enlarged master pattern dataobtained by performing an enlargement process for the master outline data of the master image extracted from the outline extraction circuit is stored in a second memory, and the outline data of the image of the to-be-recognized object is extracted by the outline extractioncircuit to form a second to-be-recognized pattern. A
first pattern matching process for collating the first to-be-recognized pattern with the master outline data and a second pattern matching process for collating the second to-be-recognized pattern data with the enlarged master pattern data are performed parallel to each other.

_ 4 - 2033359 An apparatus of matching patterns according to the present invention comprises an outline extraction cir-cuit for independently fetching image data of a prepared master and image data of a to-be-recognized object and extracting outline data of the images, a pattern enlargement circuit, connected to the output terminal of the outline extraction circuit, for performing an enlargement process for the extracted outline data, a first memory, connected to the output terminal of the outline extraction circuit, for storing the outline data extracted for the master image as master outline data, a first pattern matching circuit, connected to the output terminals of the first memory and the pattern enlarge-ment circuit, for collating the master outline data with first to-be-recognized pattern data of the to-be-recognized object obtained by the pattern enlargement circuit, a second memory, connected to the output ter-minal of the pattern enlargement circuit, for storing enlarged master pattern data of the master image obtained by the pattern enlargement circuit, and a second pattern matching circuit, connected to the output terminals of the second memory and the outline extraction circuit, for collating the enlarged master pattern data of the master image with second to-be-recognized pattern data of the to-be-recognized object obtained by the outline extraction circuit.
In the present invention, master outline data Z0?3359 serving as a comparison reference is outline data extracted from prepared master image data, and this master outline data is stored in the memory without being enlargement-processed. Subsequently, to-be-recognized pattern data to be compared with the masteroutline data is formed by performing an enlargement process for outline data of a to-be-recognized object.
The two pattern data are collated with each other, and a portion of the master outline projecting from the to-be-recognized pattern in this collation is recognized as short-defect information.
In the present invention, when pattern matching is executed, splashing defect information is recognized parallel to recognition of short-defect information. In the recognition of splashing defect information, master enlarged pattern data formed by performing an enlarge-ment process for the master outline data extracted from the master image data and stored in the memory is collated with outline (to-be-recognized pattern) data of the to-be-recognized object. In this collation, a portion of the to-be-recognized pattern projecting from the master enlarged pattern is recognized as splashing defect information.
In the present invention, after the outline extrac-5 tion circuit extracts the outline data of the prepared~, master image, the pattern enlargement circuit pattern-enlarges the outline data to form enlarged master - 6 - Z0~3~59 pattern data. This enlarged pattern data is stored in the second memory. Since the outline data of the master image is supplied to the first memory without being enlargement-processed parallel to the storage of the pattern data, the first memory stores the outline data of the master image as master outline data. After this storage, the outline extraction circuit extracts outline data of an image of a to-be-recognized object.
This outline data is supplied to the second pattern matching circuit without being enlarged by the pattern enlargement circuit. This matching circuit collates the to-be-recognized outline data with the enlarged master pattern data read out from the second memory, thereby recognizing the presence/absence of splashing defect information. Since the outline data of the image of the to-be-recognized object is enlargement-processed by the pattern enlargement circuit and supplied to the first pattern matching circuit parallel to the collation, this matching circuit collates the to-be-recognized outline data with the outline data of the master image read out from the first memory, thereby recognizing the presence/absence of short-defect information.
This invention can be more fully understood from the following detailed description when taken in con-junction-with the accompanying drawings, in which:
Fig. l is a block diagram showing an apparatus for carrying out a method of matching patterns according to _ 7 _ X0~33~9 an embodiment of the present invention;
Fig. 2 is a block diagram showing a relationship between an arrangement of a short-defect recognizing unit for recognizing short-defect information and various types of patterns;
Fig. 3 is a block diagram showing a relationship between an arrangement of a splashing defect recognizing unit for recognizing splashing defect information and various types of patterns; and Fig. 4 is a block diagram of a pattern matching apparatus according to another embodiment of the present invention.
Referring to Fig. 1, a conveyor 1 for a to-be-recognized object conveys a to-be-recognized object 2 having a printed pattern formed by, e.g., screen process printing on its surface in a direction indicated by an arrow. A television camera 3 is arranged to oppose a portion of the conveyor 1, and a field of view of the camera 3 is illuminated by a lighting device 4 as needed.
An electronic camera incorporating a linear array image sensor is used as the camera 3 and scans the printed pattern on the object 2 along a direction per-pendicular to a conveyance direction of the conveyor 1.
Picture (image) information output from the camera 3 is processed by a defect detecting device 5 shown in Fig. 1, and the quality of the printed pattern on the object 2 is checked accordingly.

- 8 - 2~3359 The defect detecting device 5 comprises an AGC cir-cuit 6, an outline extraction circuit 7, a short-defect recognizing unit 8, a short-defect determining circuit 9, a splashing defect recognizing unit 10, a splashing defect determining circuit 11, and a control circuit 12.
The AGC circuit 6 is connected to the output ter-minal of the television camera 3 and used to maintain a transmission output constant with respect to a gain variation in the entire defect detecting device 5. The outline extraction circuit 7 is connected to the output terminal of the AGC circuit 6. The outline extraction circuit 7 constituted by a differentiation circuit fetches all of signals output from the camera 3 and emphasizes changes in signals, thereby extracting outline data of the image.
As shown in Figs. 1 and 2, the short-defect recognizing unit 8 is constituted by a first pattern enlargement circuit 15, a first master pattern memory 16, and a first pattern matching circuit 17. The first pattern enlargement circuit 15 is connected to the out-put terminal of the outline extraction circuit 7. The pattern enlargement circuit 15 performs an enlargement process for data concerning an outline of an image in the vertical and horizontal directions in order to allow an error based on a variation in printed pattern of the to-be-recognized object 2 and precision of the conveyor 1, and its enlargement size can allow a slight error 9 20~3359 based on an initial variation or conveyance precision of products. This enlargement size is designated upon ini-tialization. The pattern enlargement circuit 15 per-forms an enlargement process of the outline pattern to form a to-be-recognized pattern. Note that the outline extraction circuit, the pattern enlargement circuit, and the like are described in U.S.P. Application No.
07/208,097 (patented).
The first master pattern memory 16 is connected to the output terminal of the outline extraction circuit 7 and can be subjected to both write and read operations.
The pattern memory 16 stores master outline data extracted by the outline extraction circuit 7 from a product determined as a good product by eye inspection or another suitable inspection. While to-be-recognized objects 2 of the same type are inspected, the memory 16 does not change but keeps the stored outline data.
The first pattern matching circuit 17 is connected to the output terminals of the first pattern enlargement circuit 15 and the first master pattern memory 16. The matching circuit 17 collates the master outline data read out from the first master pattern memory 16 with the to-be-recognized pattern data obtained by the first pattern enlargement circuit 15 and checks whether the two patterns coincide with each other.
The short-}defect determining circuit 9 is connected to the output terminal of the first pattern matching - 10 - ZQ:~33S9 circuit 17. The short-defect determining circuit 9 checks whether the size of defect information recognized by the short-defect recognizing unit 8 indicates a defect. Note that the size of a defect is set in the determining circuit 9 upon initialization.
As shown in Figs. 1 and 3, the splashing defect recognizing unit 10 is constituted by a second pattern enlargement circuit 21, a second master pattern memory 22, and a second pattern matching circuit 23.
The second pattern enlargement circuit 21 is connected to the output terminal of the outline extrac-tion circuit 7. The second pattern enlargement circuit 21 has the same arrangement as that of the first pattern enlargement circuit 15 and performs an enlargement pro-cess for outline data to enlarge an outline, thereby forming enlarged master pattern data.
The second master pattern memory 22 is connected to the output terminal of the second pattern enlargement circuit 21. This memory 22 can also be subjected to both write and read operations. The memory 22 stores enlarged master pattern data extracted from a product determined as a good product by eye inspection or another suitable inspection and processed by the circuit 21. While to-be-recognized objects 2 of the same type are inspected, the enlarged master pattern data stored in the memory 16 is kept unchanged.
The second pattern matching circuit 23 is connected - 11 - Z~33~9 to the output terminals of the outline extraction cir-cuit 7 and the second master pattern memory 22. The matching circuit 23 collates the enlarged master pattern data read out from the second master pattern memory 22 with the outline data of the to-be-recognized object 2 obtained by the outline extraction circuit 7 and checks whether the two pattern data coincide with each other.
The splashing defect determining circuit 11 is connected to the output terminal of the second pattern matching circuit 23. The second pattern matching circuit 23 checks whether the size of defect information recognized by the splashing defect recognizing unit 10 indicates a defect. Note that the size of a defect is set in the determining circuit ll upon initialization.
The control circuit 12 controls operations of the entire apparatus, i.e., the AGC circuit 6, the outline extraction circuit 7, the short-defect recognizing unit 8, the short-defect determining circuit 9, the splashing defect recognizing unit 10, and the splashing defect determining circuit 11.
An operation of detecting a defect by using the apparatus having the above arrangement will be described below.
After a necessary initialization operation e.g.
setting of the defect size and so on is performed, a printed pattern of a product determined as a good product by eye inspection or another suitable inspection - 12 - 20~359 is imaged by the television camera 3, and outline data is extracted from image data of the good product (master) by the outline extraction circuit 7. The extracted outline data is stored in the short-defect recognizing unit 8 and the splashing defect recognizing unit 10, as shown in Figs. 2 and 3.
More specifically, in the short-defect recognizing unit 8, master image data output from the outline extraction circuit 7 and indicated by reference symbol a in Fig. 2, e.g., outline (b) data of a character A
is stored as master outline data in the first master pattern memory 16. In the splashing defect recognizing unit 10, the outline (b) data of the master image A out-put from the outline extraction circuit 7 and indicated by a in Fig. 3 is enlargement-processed by the second pattern enlargement circuit 21, and enlarged master pattern (c) data obtained by this enlargement process is stored in the second master pattern memory 22.
After pattern setting according to the master image is performed, to-be-recognized objects 2 are sequen-tially conveyed to the field of view of the television camera 3 by the conveyor 1 and imaged. when defect image (d) data output from the camera 3 is input to the outline extraction circuit 7 via the AGC circuit 6, outline (e) data is extracted from the image (d) data by the outline extraction circuit 7 and supplied to the short-defect recognizing unit 8 and the splashing defect 20333~9 recognizing unit 10.
In the short-defect recognizing unit 8, therefore, the outline (e) data according to the to-be-recognized object 2 output from the outline extraction circuit 7 as shown in Fig. 2 is enlargement-processed by the first pattern enlargement circuit 15. To-be-recognized pat-tern (f) data obtained by the enlargement process is supplied to the first pattern matching circuit 17. The to-be-recognized pattern (f) data is input to the cir-cuit 17 in synchronism with a read operation of themaster outline (b) data corresponding to the master image A from the first master pattern memory 16 to the first pattern matching circuit 17, and the two patterns are collated with each other.
As a result, if the image d has a short-defect X
(this defect X is larger than an enlargement size set in the enlargement circuit 15) as shown in Fig. 2, a por-tion of the outline b corresponding to the master image projects from the enlarged to-be-recognized pattern _.
The first pattern matching circuit 17 recognizes this projecting portion as short-defect information.
This defect information is supplied to the short-defect determining circuit 9, and the circuit 9 checks whether the defect information is a true defect. This determination is performed on the basis of whether the input defect information has a size larger than .
a predetermined size.

- 14 - 2~359 The outline (e) data of the image (d) corresponding to the to-be-recognized object 2 supplied to the splashing defect recognizing unit 10 is directly supplied to the second pattern matching circuit 23 without being enlargement-processed. Outline (el) data corresponding to the object 2 is input to the circuit 23 in synchronism with a read operation of the enlarged master pattern (c) data corresponding to the master image A from the second master pattern memory 22 to the second pattern matching circuit 23, and the two patterns are collated with each other.
If the image d has a splashing defect Y as shown in Fig. 3, therefore, a portion of an outline e' corre-sponding to the to-be-recognized object 2 projects from the enlarged master pattern c. The second pattern - matching circuit 23 recognizes this projecting portion as splashing defect information. This defect infor-mation is supplied to the splashing defect determining circuit 11, and the circuit 11 checks whether the defect information is a true defect. Similar to the deter-mination of short-defect information, the determination of splashing defect information is performed on the basis of whether the input defect information has a size larger than a predetermined size.
Although the pattern enlargement circuit is pro-vided for each recognizing unit corresponding to the type of defect in the above embodiment, a single pattern - 15 - 2033~9 enlargement circuit 15 may be commonly used for the two recognizing units 8 and 10 as shown in Fig. 4.
According to this embodiment, a single pattern enlarge-ment circuit 15 is connected to a selector switch 25.
5 The contacts a and b of the selector switch 25 are con-nected to the first pattern matching circuit 17 of the short-defect recognizing unit 8 and the second master pattern memory 22 of the splashing defect recognizing unit 10, respectively.
In this pattern matching apparatus, at first the selector switch 25 iS switched to the contact b, and the outline (b) data of the master image is enlargement-processed by the pattern enlargement circuit 15. The enlarged master pattern (c) output from the pattern enlargement circuit 15 is stored in the second pattern memory 22 through the selector switch 25. The stored master pattern (c) data is used for collating with each of the to-be-recognized pattern (e') data sequentially input from the outline extraction circuit 7.
After the enlarged master pattern (c) data is stored in the second master pattern memory 22, the selector switch 25 is switched to the contact a. Then, the pattern enlargement circuit 15 enlargement-processes the to-be-recognized pattern (e) data sequentially input from the outline extraction circuit 7. The enlarged to-be-recognized pattern (f) data is input to the first pattern matching circuit 17 through the switch 25 and - 16 - 2Q~33S9 collated with the first master pattern (b) data stored in the first pattern memory 16.
According to the embodiment of Fig. 4, the single pattern enlargement circuit 15 enlargement-processes at once the master outline (b) data for storing in the second master pattern memory 22, and then is used for enlargement-processing the outline (e) data of the to-be-recognized image data sequentially produced from the outline extraction circuit 7.
According to the pattern matching method of the present invention described above, an outline pattern extracted from a prepared master image is stored as a master outline, and this master outline is collated with a to-be-recognized pattern formed by performing an enlargement process for an outline pattern obtained for a to-be-recognized object. In this collation, since a portion of the master outline projecting from the to-be-recognized pattern is recognized as defect information, a short-defect can be recognized.
An outline extracted from a prepared master image is stored as a master outline and collated with a to-be-recognized pattern formed by enlarging the outline of a to-be-recognized object, and a portion of the master outline projecting from the to-be-recognized pattern is recognized as defect information in this collation. In addition, parallel to this process, an enlarged master pattern formed by enlarging the outline extracted from - 17 - 2~333~

the master image is stored, and the outline of the to-be-recognized object is directly collated with the stored enlarged master pattern. In this collation, a portion of the to-be-recognized pattern projecting from the enlarged master pattern can be recognized as defect information. Therefore, both the short-defect and splashing defect information can be recognized parallel to each other.
Since the pattern matching apparatus the present invention comprises the outline extraction circuit, the pattern enlargement circuit, the first and second memories, and the first and second pattern matching circuits, it is capable of recognizing both of short-defect and splashing defect information parallel to each other.

Claims (16)

1. A method of matching patterns, comprising the steps of:
optically scanning a reference image and a to-be-recognized image and outputting master image data and to-be-recognized image data;
extracting master outline data representing an outline of the reference image from the master image data;
extracting to-be-recognized outline data repre-senting an outline of the to-be-recognized image from the to-be-recognized image data;
performing an enlargement process for the to-be-recognized outline data to enlarge the to-be-recognized outline, thereby forming the to-be-recognized outline data; and collating the master outline data with the to-be-recognized outline data, to determine that the to-be-recognized image data is a defect by recognizing that a portion of the master outline projects from the to-be-recognized outline and output defect information.
2. A method according to claim 1, wherein said collating step includes the step of checking whether the defect information is true detect information.
3. A method according to claim 1, wherein said collating step comprises the step of fetching the master outline data and the to-be-recognized outline data in synchronism with each other and comparing the two data.
4. A method of matching patterns, comprising the steps of:
optically scanning a reference image and a to-be-recognized image and outputting master image data and to-be-recognized image data;
extracting master outline data representing an outline of the reference image from the master image data to form master outline data;
extracting to-be-recognized outline data repre-senting an outline of the to-be-recognized image from the to-be-recognized image data;
performing an enlargement process for the master outline data and the to-be-recognized outline data to enlarge the master outline and the to-be-recognized outline and form enlarged master outline data and enlarged to-be-recognized outline data;
collating the master outline data with the enlarged to-be-recognized outline data, to determine that the to-be-recognized image has a short-defect indicating an omission of an image by recognizing that a portion of the master outline projects from the to-be-recognized outline and output first defect information; and collating the enlarged master outline data with the to-be-recognized image outline data, to determine that the to-be-recognized image is splashing defect including an image of a different type by recognizing that a portion of the to-be-recognized image projects from the enlarged and output second defect information.
5. A method according to claim 1, wherein said first and second determining steps respectively include the steps of checking whether the first and second defect information are true defect information.
6. A method according to claim 1, wherein said first determining step comprises the step of fetching the master outline data and the enlarged to-be-recognized outline data in synchronism with each other and comparing the two data, and said second determining step comprises the step of fetching the enlarged master outline data and the to-be-recognized outline data in synchronism with each other and comparing the two data.
7. An apparatus for matching patterns, comprising:
image data output means for optically scanning a reference image and a to-be-recognized image and out-putting master image data and to-be-recognized image data;
outline data output means for extracting and out-putting master outline data representing an outline of the reference image from the master image data and extracting and outputting to-be-recognized outline data representing an outline of the to-be-recognized image from the to-be-recognized image data;
enlarged outline data output means for performing an enlargement process for the to-be-recognized outline data to enlarge the to-be-recognized outline and output enlarged to-be-recognized outline data; and image determining means for collating the master outline data with the enlarged to-be-recognized outline data, determining that the to-be-recognized outline by recognizing that a portion of the master outline pro-jects from the to-be-recognized outline and outputting defect information.
8. An apparatus according to claim 7, wherein said image determining means includes a determining circuit for checking whether the defect information is true defect information.
9. An apparatus according to claim 7, wherein said image determining means comprises a pattern matching circuit for fetching the master outline data and the to-be-recognized outline data in synchronism with each other and comparing the two data.
10. An apparatus according to claim 7, wherein said outline data output means has memory means for storing master outline data, and said image determining means comprises a pattern matching circuit for comparing the master outline data read out from said memory means with the enlarged to-be-recognized outline data input from said enlarged outline data output means in synchronism with the read operation of the master outline data.
11. An apparatus according to claim 7, wherein said image data output means comprises image sensor means for optically scanning a plurality of to-be-checked objects each having a predetermined printed pattern correspond-ing to the to-be-recognized object.
12. An apparatus for matching patterns, comprising:
image data output means for optically scanning a reference image and a to-be-recognized image and out-putting master image data and to-be-recognized image data;
outline data output means for extracting and out-putting master outline data representing an outline of the reference image from the master image data and extracting and outputting to-be-recognized outline data representing an outline of the to-be-recognized image from the to-be-recognized image data;
enlarged data output means for performing an enlargement process for the master outline data and the to-be-recognized outline data to enlarge the master outline and the to-be-recognized outline and output enlarged master outline data and enlarged to-be-recognized outline data;
first determining means for collating the master outline data with the enlarged to-be-recognized outline data, determining that the to-be-recognized object has a short-defect indicating an omission of the image by recognizing that a portion of the master outline projects from the to-be-recognized outline, and out-putting first defect information; and second determining means for collating the enlarged master outline data with the to-be-recognized image outline data, determining that the to-be-recognized image is a splashing defect including an image of a different type by recognizing that a portion of the to-be-recognized image projects from the enlarged master outline and outputting second defect information.
13. An apparatus according to claim 12, wherein said first and second determining means respectively include first and second determining circuits for checking whether the first and second defect information are true defect information.
14. An apparatus according to claim 12, wherein said first determining means comprises a first pattern matching circuit for fetching the master outline data and the enlarged to-be-recognized outline data in synchronism with each other and comparing the two data, and said second determining means comprises a second pattern matching circuit for fetching the enlarged master outline data and the to-be-recognized outline data in synchronism with each other and comparing the two data.
15. A method of matching patterns, comprising:

a first pattern matching process for fetching an image of a prepared master into outline extracting means to extract an outline of the image, storing a master outline obtained by the extraction into memory means, extracting an outline of an image of a to-be-recognized object by said outline extracting means, performing a pattern enlargement process for the extracted outline to form a first to-be-recognized pattern, and collating the to-be-recognized pattern with the master outline; and a second pattern matching process for storing an enlarged master pattern obtained by performing the pattern enlargement process for the master outline of the master image extracted by said outline extracting means into second memory means, extracting the outline of the image of the to-be-recognized object by said outline extracting means to form a second to-be-recognized pattern, and collating the second to-be-recognized pattern with the enlarged master pattern, wherein said first and second pattern matching pro-cesses are performed parallel to each other.
16. An apparatus for matching patterns, comprising:
outline extracting means for independently fetching an image of a prepared master and an image of a to-be-recognized object and extracting outlines of the images;
pattern enlargement processing means, connected to said outline extracting means, for performing an enlargement process for the extracted outlines;

first memory means, connected to said outline extracting means, for storing the outline extracted for the master image as a master outline;
first pattern matching means, connected to output terminals of said first memory means and said pattern enlargement processing means, for collating the master outline with a first to-be-recognized pattern for the to-be-recognized object obtained by said pattern enlargement processing means;
second memory means, connected to the output ter-minal of said pattern enlargement processing means, for storing an enlarged master pattern for the master image obtained by said pattern enlargement processing means;
and second pattern matching means, connected to output terminals of said second memory means and said outline extracting means, for collating the enlarged master pattern for the master image with a second to-be-recognized pattern for the to-be-recognized object obtained by said outline extracting means.
CA002033359A 1990-01-12 1990-12-28 Method of matching patterns and apparatus therefor Expired - Fee Related CA2033359C (en)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
JP2-5632 1990-01-12
JP2005632A JPH03210679A (en) 1990-01-12 1990-01-12 Method and device for pattern matching

Publications (2)

Publication Number Publication Date
CA2033359A1 CA2033359A1 (en) 1991-07-13
CA2033359C true CA2033359C (en) 1995-12-12

Family

ID=11616527

Family Applications (1)

Application Number Title Priority Date Filing Date
CA002033359A Expired - Fee Related CA2033359C (en) 1990-01-12 1990-12-28 Method of matching patterns and apparatus therefor

Country Status (7)

Country Link
US (1) US5253306A (en)
EP (1) EP0437273B1 (en)
JP (1) JPH03210679A (en)
KR (1) KR930009741B1 (en)
CA (1) CA2033359C (en)
DE (1) DE69127835T2 (en)
FI (1) FI910097A (en)

Families Citing this family (62)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US5600734A (en) * 1991-10-04 1997-02-04 Fujitsu Limited Electron beam tester
US6067379A (en) * 1988-12-09 2000-05-23 Cognex Corporation Method and apparatus for locating patterns in an optical image
DE4041484A1 (en) * 1990-12-22 1992-06-25 Thomson Brandt Gmbh DEMAGNETIZING CIRCUIT FOR THE PIPES IN A TELEVISION RECEIVER
JP3062338B2 (en) * 1991-03-01 2000-07-10 キヤノン株式会社 Image processing device
JP3237928B2 (en) * 1992-12-04 2001-12-10 株式会社東芝 Pattern inspection method and apparatus
JPH07117498B2 (en) * 1991-12-11 1995-12-18 インターナショナル・ビジネス・マシーンズ・コーポレイション Inspection system
IL102659A (en) * 1992-07-27 1997-07-13 Orbot Instr Ltd Apparatus and method for comparing and aligning two digital representations of an image
US5859923A (en) * 1992-12-29 1999-01-12 Cognex Corporation Mark quality inspection apparatus and method
JP3132794B2 (en) * 1993-06-21 2001-02-05 シオノギクオリカプス株式会社 Inspection device for printed characters of solid preparations
US5452368A (en) * 1993-08-02 1995-09-19 Motorola, Inc. Method of detecting defects in semiconductor package leads
US5479683A (en) * 1993-12-29 1996-01-02 Bausch & Lomb Incorporated Three-dimensional eyewinder apparatus
US6026176A (en) 1995-07-25 2000-02-15 Cognex Corporation Machine vision methods and articles of manufacture for ball grid array inspection
US5872870A (en) * 1996-02-16 1999-02-16 Cognex Corporation Machine vision methods for identifying extrema of objects in rotated reference frames
US5909504A (en) * 1996-03-15 1999-06-01 Cognex Corporation Method of testing a machine vision inspection system
US6259827B1 (en) 1996-03-21 2001-07-10 Cognex Corporation Machine vision methods for enhancing the contrast between an object and its background using multiple on-axis images
US6298149B1 (en) 1996-03-21 2001-10-02 Cognex Corporation Semiconductor device image inspection with contrast enhancement
US5978502A (en) * 1996-04-01 1999-11-02 Cognex Corporation Machine vision methods for determining characteristics of three-dimensional objects
US5835634A (en) * 1996-05-31 1998-11-10 Adobe Systems Incorporated Bitmap comparison apparatus and method using an outline mask and differently weighted bits
US6137893A (en) * 1996-10-07 2000-10-24 Cognex Corporation Machine vision calibration targets and methods of determining their location and orientation in an image
US5960125A (en) 1996-11-21 1999-09-28 Cognex Corporation Nonfeedback-based machine vision method for determining a calibration relationship between a camera and a moveable object
US5953130A (en) * 1997-01-06 1999-09-14 Cognex Corporation Machine vision methods and apparatus for machine vision illumination of an object
US6075881A (en) 1997-03-18 2000-06-13 Cognex Corporation Machine vision methods for identifying collinear sets of points from an image
US5974169A (en) * 1997-03-20 1999-10-26 Cognex Corporation Machine vision methods for determining characteristics of an object using boundary points and bounding regions
US6141033A (en) * 1997-05-15 2000-10-31 Cognex Corporation Bandwidth reduction of multichannel images for machine vision
US6608647B1 (en) 1997-06-24 2003-08-19 Cognex Corporation Methods and apparatus for charge coupled device image acquisition with independent integration and readout
US5978080A (en) * 1997-09-25 1999-11-02 Cognex Corporation Machine vision methods using feedback to determine an orientation, pixel width and pixel height of a field of view
WO1999017250A1 (en) * 1997-10-01 1999-04-08 Island Graphics Corporation Image comparing system
US6025854A (en) * 1997-12-31 2000-02-15 Cognex Corporation Method and apparatus for high speed image acquisition
US6282328B1 (en) 1998-01-28 2001-08-28 Cognex Corporation Machine vision systems and methods for morphological transformation of an image with non-uniform offsets
US6236769B1 (en) 1998-01-28 2001-05-22 Cognex Corporation Machine vision systems and methods for morphological transformation of an image with zero or other uniform offsets
US6215915B1 (en) 1998-02-20 2001-04-10 Cognex Corporation Image processing methods and apparatus for separable, general affine transformation of an image
US6381375B1 (en) 1998-02-20 2002-04-30 Cognex Corporation Methods and apparatus for generating a projection of an image
US6075886A (en) * 1998-03-09 2000-06-13 Xerox Corporation Method and apparatus for reducing the complexity of color correction using subsampling
US6173213B1 (en) 1998-05-11 2001-01-09 Ellison Machinery Company Motorized inbound laser orientation and wheel recognition station
JP2000067247A (en) * 1998-06-08 2000-03-03 Toshiba Corp Image recognizing device
US6154567A (en) * 1998-07-01 2000-11-28 Cognex Corporation Pattern similarity metric for image search, registration, and comparison
US7016539B1 (en) 1998-07-13 2006-03-21 Cognex Corporation Method for fast, robust, multi-dimensional pattern recognition
US6381366B1 (en) 1998-12-18 2002-04-30 Cognex Corporation Machine vision methods and system for boundary point-based comparison of patterns and images
US6687402B1 (en) 1998-12-18 2004-02-03 Cognex Corporation Machine vision methods and systems for boundary feature comparison of patterns and images
JP4519323B2 (en) * 1999-02-11 2010-08-04 ブリティッシュ・テレコミュニケーションズ・パブリック・リミテッド・カンパニー Video signal quality analysis
KR100326563B1 (en) * 1999-03-30 2002-03-12 곽정소 Degaussing Circuit and Method of Driving The Same
US6735327B1 (en) 1999-09-16 2004-05-11 Shofner Engineering Associates, Inc. Color and trash measurements by image analysis
US6684402B1 (en) 1999-12-01 2004-01-27 Cognex Technology And Investment Corporation Control methods and apparatus for coupling multiple image acquisition devices to a digital data processor
JP3604993B2 (en) * 2000-03-16 2004-12-22 シャープ株式会社 Image encoding device, image encoding method, image decoding device, and image decoding method
US6748104B1 (en) 2000-03-24 2004-06-08 Cognex Corporation Methods and apparatus for machine vision inspection using single and multiple templates or patterns
DE10103958C1 (en) * 2001-01-30 2002-05-29 Infineon Technologies Ag Defect detection method for semiconductor exposure mask uses comparison of actual mask image with reference mask image
US6959112B1 (en) 2001-06-29 2005-10-25 Cognex Technology And Investment Corporation Method for finding a pattern which may fall partially outside an image
US7613316B2 (en) * 2003-07-22 2009-11-03 L-3 Communications Security and Detection Systems Inc. Methods and apparatus for detecting objects in baggage
US7190834B2 (en) 2003-07-22 2007-03-13 Cognex Technology And Investment Corporation Methods for finding and characterizing a deformed pattern in an image
US8081820B2 (en) * 2003-07-22 2011-12-20 Cognex Technology And Investment Corporation Method for partitioning a pattern into optimized sub-patterns
US7423280B2 (en) * 2004-08-09 2008-09-09 Quad/Tech, Inc. Web inspection module including contact image sensors
US9098932B2 (en) * 2004-08-11 2015-08-04 Ati Technologies Ulc Graphics processing logic with variable arithmetic logic unit control and method therefor
JP4154374B2 (en) * 2004-08-25 2008-09-24 株式会社日立ハイテクノロジーズ Pattern matching device and scanning electron microscope using the same
US8437502B1 (en) 2004-09-25 2013-05-07 Cognex Technology And Investment Corporation General pose refinement and tracking tool
US20060147707A1 (en) * 2004-12-30 2006-07-06 Jian Meng Compacted, chopped fiber glass strands
DE102005026630A1 (en) * 2005-06-03 2006-12-07 Würth Elektronik GmbH & Co. KG Identical objects, e.g. electronic printed circuit boards, inspecting method, involves recording, storing and displaying image of printed circuit board to be inspected after inspection of another board to be inspected
US8111904B2 (en) 2005-10-07 2012-02-07 Cognex Technology And Investment Corp. Methods and apparatus for practical 3D vision system
US8162584B2 (en) 2006-08-23 2012-04-24 Cognex Corporation Method and apparatus for semiconductor wafer alignment
US8103085B1 (en) 2007-09-25 2012-01-24 Cognex Corporation System and method for detecting flaws in objects using machine vision
WO2010041447A1 (en) * 2008-10-09 2010-04-15 日本電気株式会社 Failure detection system, failure detection method, and medium containing failure detection program
JP5719967B2 (en) * 2012-03-13 2015-05-20 富士フイルム株式会社 Imaging device with projector and control method thereof
US9679224B2 (en) 2013-06-28 2017-06-13 Cognex Corporation Semi-supervised method for training multiple pattern recognition and registration tool models

Family Cites Families (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
GB2129546B (en) * 1982-11-02 1985-09-25 Cambridge Instr Ltd Image comparison
US4589140A (en) * 1983-03-21 1986-05-13 Beltronics, Inc. Method of and apparatus for real-time high-speed inspection of objects for identifying or recognizing known and unknown portions thereof, including defects and the like
DE3347645C1 (en) * 1983-12-30 1985-10-10 Dr.-Ing. Ludwig Pietzsch Gmbh & Co, 7505 Ettlingen Method and device for opto-electronic testing of a surface pattern on an object
JPS60230281A (en) * 1984-04-27 1985-11-15 Fuji Electric Co Ltd Character discriminator
US4953224A (en) * 1984-09-27 1990-08-28 Hitachi, Ltd. Pattern defects detection method and apparatus
US4648053A (en) * 1984-10-30 1987-03-03 Kollmorgen Technologies, Corp. High speed optical inspection system
EP0195161B1 (en) * 1985-03-14 1993-09-15 Nikon Corporation Apparatus for automatically inspecting objects and identifying or recognizing known and unknown portions thereof, including defects and the like and method

Also Published As

Publication number Publication date
FI910097A0 (en) 1991-01-08
KR930009741B1 (en) 1993-10-09
JPH03210679A (en) 1991-09-13
EP0437273A3 (en) 1992-09-02
EP0437273A2 (en) 1991-07-17
US5253306A (en) 1993-10-12
KR910014841A (en) 1991-08-31
DE69127835D1 (en) 1997-11-13
EP0437273B1 (en) 1997-10-08
FI910097A (en) 1991-07-13
CA2033359A1 (en) 1991-07-13
DE69127835T2 (en) 1998-04-30

Similar Documents

Publication Publication Date Title
CA2033359C (en) Method of matching patterns and apparatus therefor
JPH10506733A (en) Method and apparatus for separating foreground from background in text-containing images
EP0536570A2 (en) Method for displaying defect and apparatus therefor
JP4593729B2 (en) A method of automatically recognizing text on a structured background by combining a background model and a character model
US6636632B2 (en) Image processor and image processing method
EP0443062A1 (en) Device for inspecting quality of printed matter and method thereof
KR20030013520A (en) Image processing system
JP2003086919A (en) Pattern inspection device
JPH07101048A (en) Printing inspecting method and apparatus
JPH0560537A (en) Through-hole inspection device
JP3233205B2 (en) Circuit inspection method and device
JP3919207B2 (en) Appearance inspection device
JP3107264B2 (en) Printed board inspection method and apparatus and printed board intensive inspection system
JP2002314982A (en) Method for detecting defect
JP2676990B2 (en) Wiring pattern inspection equipment
JP2745763B2 (en) Wiring pattern inspection equipment
JP2003004659A (en) Method and apparatus for inspection of multiple pattern work
JP3557814B2 (en) Image inspection apparatus and method
JPH11271232A (en) Method and device for detecting defect in wiring on printed board
JPH0760459B2 (en) Corner detector
JPH04161841A (en) Optical printed board inspecting apparatus
JP3267372B2 (en) Method and apparatus for generating matching pattern signal in pattern matching inspection
JPH0238980B2 (en)
JPH0894335A (en) Image processing detector
JP2003279331A (en) Appearance examining apparatus

Legal Events

Date Code Title Description
EEER Examination request
MKLA Lapsed