CN106530289A - Code spraying definition machine vision detecting method based on lattices - Google Patents
Code spraying definition machine vision detecting method based on lattices Download PDFInfo
- Publication number
- CN106530289A CN106530289A CN201610954433.XA CN201610954433A CN106530289A CN 106530289 A CN106530289 A CN 106530289A CN 201610954433 A CN201610954433 A CN 201610954433A CN 106530289 A CN106530289 A CN 106530289A
- Authority
- CN
- China
- Prior art keywords
- character
- dot matrix
- pixel
- color
- threshold value
- 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.)
- Granted
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/0002—Inspection of images, e.g. flaw detection
- G06T7/0004—Industrial image inspection
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/30—Subject of image; Context of image processing
- G06T2207/30108—Industrial image inspection
- G06T2207/30144—Printing quality
Landscapes
- Engineering & Computer Science (AREA)
- Quality & Reliability (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Image Processing (AREA)
- Image Analysis (AREA)
Abstract
The invention relates to a code spraying definition machine vision detecting method based on lattices. The method is technically characterized in that a human-computer interaction device sets and adjusts the measurement area of a character string image, and a setting result is transmitted to an intelligent camera; the intelligent camera acquires the character string image of code spraying characters of a package; the intelligent camera extracts the designated color gray value of pixels in all lattices of all characters in the measurement area; the intelligent camera calculates the color area of all lattices of all characters in the measurement area; the intelligent camera determines the definition; the intelligent camera issues a control command to a removing mechanism; and the removing mechanism removes products with unclear code spraying characters. According to the invention, the design is reasonable; the code spraying characters of the package can be accurately and reliably identified and judged; products with unclear code spraying characters are removed; and the method has the characteristics of accurate judgment, high work efficiency, easy use and the like.
Description
Technical field
The invention belongs to Visual identification technology field, especially a kind of coding definition Machine Vision Detection based on dot matrix
Method.
Background technology
At present, in the product of the industries such as medicine, food, it is common to use laser or ink jet type code-spraying equipment are in product packaging
Upper spray printing batch number, date of manufacture, shelf-life, effect duration etc. identify.Code-spraying equipment is set as a kind of contactless mark
The standby spray printing that can provide character,, during coding, it sometimes appear that the unclear problem of coding character, coding character is or not which
Prestige and the use of commodity can be clearly affected, therefore, manufacturer will be detected to the coding in product packaging before dispatching from the factory.
Traditional detection method is to adopt manual detection, but missing inspection often occurs using manual detection method meeting Jing.Meanwhile, with production
The raising of speed, human eye have not caught up with the speed of service of product.In recent years, coding character picture automatic identification technology is obtained
Certain application, but, due to the font of spray printing character, color, background, font gradient and it is quick it is mobile in other
The impact of interference factor, often causes the situation for recognizing mistake or None- identified to occur.It is emphasized that enabling identification
Go out character, it is clearly not represent character;Simultaneously as in jet printing process character deformation, such as incline, stretch, cause
Cannot automatic identification, such case can not conclude that character is unintelligible again.
The content of the invention
It is an object of the invention to overcome the deficiencies in the prior art, there is provided a kind of reasonable in design, high precision and can improve
The coding definition machine vision detection method based on dot matrix of work efficiency.
The present invention solves its technical problem and takes technical scheme below to realize:
A kind of coding definition machine vision detection method based on dot matrix, comprises the following steps:
Step 1, the measured zone that character string picture is set, adjusted by human-computer interaction device, and will setting result transmission
To smart camera;
Step 2, smart camera obtain the character string picture of package coding character;
Step 3, smart camera extract the pixel designated color ash in measured zone in all dot matrix of all characters respectively
Angle value, and make the following judgment:
If the grey scale pixel value is less than the white threshold value of color gray scale, the pixel is labeled as into white pixel accumulative dot matrix
The quantity of interior white pixel;
If the grey scale pixel value is more than the black threshold value of color gray scale, the pixel is labeled as into black pixel accumulative dot matrix
The quantity of interior black pixel;
If the grey scale pixel value is other values, the pixel is labeled as into vague image in fuzzy pixel accumulative dot matrix
The quantity of element;
The color area of all dot matrix of all characters in step 4, smart camera difference computation and measurement region, and carry out as
Lower judgement:
If black elemental area is less than color area blank threshold plus fuzzy elemental area, the point is labeled as into 0;
If black elemental area is more than color area solid black threshold value, the point is labeled as into 1;
If other values, then the point is labeled as obscuring;
Step 5, smart camera enter line definition judgement as follows:
If any one point is noted as obscuring in the dot matrix of measured zone, judge the dot character as unintelligible;
If not being noted as fuzzy point, the dot matrix for being labeled as 0 and 1 is compared with standard character library dot matrix, if
The character for fitting like a glove can be matched in standard character library, is then judged that the dot matrix is clear, is otherwise judged to unintelligible;
Step 6, smart camera send control command and by rejecting mechanism by coding character unsharp product to rejecting mechanism
Product are rejected.
The concrete grammar of the step 1 is comprised the following steps:
(1) standard characters image is input into and is shown;
(2) character pattern number, character font and character color are input into;
(3) dot matrix measured zone is automatically generated by dot matrix number, the measured zone is divided into the quadrilateral area at interval;
(4) dot matrix measured zone is adjusted using flexible, inclined method, be allowed to coincide with actual characters image lattice;
(5) set the white threshold value of color gray scale of pixel, the black threshold value of color gray scale, set dot matrix color area blank threshold,
Color area solid black threshold value;
(6) read LED matrix individual characters or image are pressed, for character or pattern in non-standard character repertoire, individually with dot matrix side
Formula sets up model;
(7) setting result is sent to smart camera.
The character string picture includes Chinese character, letter, numeral, fixed character and blank character.
The method that (3) step generates dot matrix measured zone is:A character rectangular lattice is firstly generated, then
Four angles of dilatory rectangle, the character is covered, until all characters in character string are covered, so as to obtain the dot matrix specified
Measured zone, the dot matrix measured zone is by being about to character string automatic separation, and each character is individually divided into by coding character library
Magnetic dot arrays.
The character pattern is the rectangular lattice between 5 × 5 to 32 × 20.
The white threshold value of the color gray scale, the black threshold value span of color gray scale are 0~255, the white threshold value of color gray scale and
The black threshold value of color gray scale is set to 25 and 230;The color area blank threshold and color area solid black threshold value are set to 10%
With 90%.
Advantages of the present invention and good effect are:
The present invention automatically generates dot matrix measured zone can according to character string picture by dot matrix number by human-computer interaction device
Adjustment measured zone, smart camera according to the pixel designated color gray value in all dot matrix of all characters in measured zone and
The color area of dot matrix carries out definition judgment, can accurately and reliably complete the identification to package coding character and judge, and
By coding character unsharp product reject, with accuracy of judgement, high working efficiency, it is easy to use the features such as.
Description of the drawings
Fig. 1 is the system connection figure of the present invention;
Fig. 2 is character pattern structure (the 7 row X5 row) schematic diagram of the present invention.
Specific embodiment
The embodiment of the present invention is further described below in conjunction with accompanying drawing:
A kind of coding definition machine vision detection method based on dot matrix, is real in identifying system as shown in Figure 1
Existing, the identifying system includes smart camera 1, human-computer interaction device 2 and rejecting mechanism 3, smart camera and human-computer interaction device
And rejecting mechanism is connected.Smart camera is used to obtain coding character picture, carry out the judgement of coding character sharpness and to rejecting
Mechanism sends rejecting control command;Human-computer interaction device is shown for character string picture, measured zone sets and is adjusted and to intelligence
Can camera transmission;Rejecting mechanism receives the control command of smart camera and rejects coding character unsharp product.
The present invention is comprised the following steps:
Step 1, the measured zone that character string picture is set, adjusted by human-computer interaction device, and will setting result transmission
To smart camera.Concrete grammar includes following processing procedure:
(1) character string picture is input into and shows
In this step, the clear character string picture for detecting coding character is needed to be input to human-computer interaction device this batch
In and shown.Character string picture input can carry out image by smart camera and obtain and send to human-computer interaction device,
And shown by human-computer interaction device.
Character string includes:Chinese character, letter, numeral, fixed character and blank character.
(2) character pattern number, character font and character color are input into
In this step, by each character setting be 5 × 5 to 32 × 20 between rectangular lattice.The present embodiment is with 7X5 points
Illustrate as a example by battle array, the rectangular lattice shown on the left of such as Fig. 2.
(3) dot matrix measured zone is automatically generated by dot matrix number:It is divided into the quadrilateral area at interval inside measured zone.
In this step, the concrete generation method of dot matrix measured zone is:Firstly generate the character rectangle of a 7X5=35
Dot matrix, then draws four angles of rectangle, and the character is covered, until all characters in character string are covered (including empty word
Symbol), obtain the dot matrix measured zone specified.In fact, dot matrix measured zone is by being about to character string automatic separation, and by each word
Symbol is individually divided into magnetic dot arrays by coding character library.Dot chart is displayed in measured zone, and dot chart can also be moved, every to adapt to
The position of individual character.
(4) dot matrix measured zone adjustment:Stretch, incline the measured zone of single character, be allowed to and actual characters picture point
Battle array is coincide.
Four points in measured zone, arbitrarily can draw, and the dot matrix of the inside is with dilatory auto Deformation.In this step
In, the size, angle and deformation according to character picture adjusts width, length and the angle of inclination of measured zone, so as to incite somebody to action
Character is covered.As coding character is possible to as tilted character, therefore, for tilted character, rectangle can be stretched
Contracting, inclination, obtain inclined rectangular lattice as shown in the right side of fig 2.
When region adjustment is measured, each row can individually set angle of inclination, and in row, each interval can
Length is adjusted individually.The gradient of each character is the same in principle.
(5) by read LED matrix individual characters or image:For character or pattern in non-standard character repertoire, can individually with point
Battle array mode sets up model.
(6) the white threshold value of color gray scale, the black threshold value of color gray scale of pixel are set;The color area blank threshold of setting dot matrix
Value, color area solid black threshold value.
The white threshold value of color gray scale, the black threshold value of color gray scale can be 0~255, in the present embodiment, the white threshold value of color gray scale
25 and 230 are set to the black threshold value of color gray scale.Color area blank threshold, color area solid black threshold value are set to 10% and 90%,
When setting dot matrix value as 1, gray scale accounts for dot matrix area ratio for 90%.
(7) send the measured zone and measurement parameter after setting to smart camera.
Step 2, smart camera obtain the character string picture of package coding character;
Step 3, smart camera extract the pixel designated color ash in measured zone in all dot matrix of all characters respectively
Angle value, and make the following judgment:
(1) if the grey scale pixel value is less than the white threshold value of color gray scale, the pixel is labeled as into white pixel;Accumulative point
The quantity of white pixel in battle array;
(2) if the grey scale pixel value is more than the black threshold value of color gray scale, the pixel is labeled as into black pixel;Accumulative point
The quantity of black pixel in battle array;
(3) if the grey scale pixel value is other values, the pixel is labeled as into fuzzy pixel, adds up to obscure in dot matrix
The quantity of pixel.
The color area of all dot matrix of all characters in step 4, smart camera difference computation and measurement region, and carry out as
Lower judgement:
(1) if black elemental area+fuzzy elemental area is less than color area blank threshold, the point is labeled as into 0;
(2) if black elemental area is more than color area solid black threshold value, the point is labeled as into 1;
(3) if other values, then the point is labeled as obscuring.
Step 5, smart camera enter line definition judgement
(1) in the dot matrix of measured zone, any one point is noted as obscuring, then judge the dot character as unintelligible;
(2) if not being noted as fuzzy point, the dot matrix of 0 and 1 will be labeled as, be compared with standard character library dot matrix, such as
Fruit can match the character for fitting like a glove in character library, then judge that the dot matrix is clear, be otherwise judged to unintelligible.
It is emphasized that embodiment of the present invention is illustrative, rather than it is determinate, therefore present invention bag
The embodiment for being not limited to described in specific embodiment is included, it is every by those skilled in the art's technology according to the present invention scheme
The other embodiment for drawing, also belongs to the scope of protection of the invention.
Claims (6)
1. a kind of coding definition machine vision detection method based on dot matrix, it is characterised in that comprise the following steps:
Step 1, the measured zone that character string picture is set, adjusted by human-computer interaction device, and send setting result to intelligence
Can camera;
Step 2, smart camera obtain the character string picture of package coding character;
Step 3, smart camera extract the pixel designated color gray value in measured zone in all dot matrix of all characters respectively,
And make the following judgment:
If the grey scale pixel value is less than the white threshold value of color gray scale, the pixel is labeled as into white pixel accumulative dot matrix words spoken by an actor from offstage
The quantity of pixel;
If the grey scale pixel value is more than the black threshold value of color gray scale, the pixel is labeled as black in black pixel accumulative dot matrix
The quantity of pixel;
If the grey scale pixel value is other values, the pixel is labeled as obscuring pixel in fuzzy pixel accumulative dot matrix
Quantity;
The color area of all dot matrix of all characters in step 4, smart camera difference computation and measurement region, and sentenced as follows
It is disconnected:
If black elemental area is less than color area blank threshold plus fuzzy elemental area, the point is labeled as into 0;
If black elemental area is more than color area solid black threshold value, the point is labeled as into 1;
If other values, then the point is labeled as obscuring;
Step 5, smart camera enter line definition judgement as follows:
If any one point is noted as obscuring in the dot matrix of measured zone, judge the dot character as unintelligible;
If not being noted as fuzzy point, the dot matrix for being labeled as 0 and 1 is compared with standard character library dot matrix, if can be
The character for fitting like a glove is matched in standard character library, is then judged that the dot matrix is clear, is otherwise judged to unintelligible;
Step 6, smart camera send control command to rejecting mechanism and pick coding character unsharp product by rejecting mechanism
Remove.
2. a kind of coding definition machine vision detection method based on dot matrix according to claim 1, it is characterised in that:
The concrete grammar of the step 1 is comprised the following steps:
(1) clear character string picture is input into and is shown;
(2) character pattern number, character font and character color are input into;
(3) dot matrix measured zone is automatically generated by dot matrix number, the measured zone is divided into the quadrilateral area at interval;
(4) dot matrix measured zone is adjusted using flexible, inclined method, be allowed to coincide with actual characters image lattice;
(5) the white threshold value of color gray scale, the black threshold value of color gray scale of pixel are set, color area blank threshold, the color of dot matrix is set
Area solid black threshold value;
(6) read LED matrix individual characters or image are pressed, for character or pattern in non-standard character repertoire, is individually built with dot matrix way
Formwork erection type;
(7) setting result is sent to smart camera.
3. a kind of coding definition machine vision detection method based on dot matrix according to claim 1, it is characterised in that:
The character string picture includes Chinese character, letter, numeral, fixed character and blank character.
4. a kind of coding definition machine vision detection method based on dot matrix according to claim 2, it is characterised in that:
The method that (3) step generates dot matrix measured zone is:A character rectangular lattice is firstly generated, rectangle is then drawn
Four angles, the character is covered, until all characters in character string are covered, so as to the dot matrix measurement zone for obtaining specifying
Domain, the dot matrix measured zone is by being about to character string automatic separation, and each character is divided into magnetic dot arrays by coding character library individually.
5. a kind of coding definition machine vision detection method based on dot matrix according to claim 2, it is characterised in that:
The character pattern is the rectangular lattice between 5 × 5 to 32 × 20.
6. a kind of coding definition machine vision detection method based on dot matrix according to claim 1, it is characterised in that:
The white threshold value of the color gray scale, the black threshold value span of color gray scale are 0~255, the white threshold value of color gray scale and color gray scale
Black threshold value is set to 25 and 230;The color area blank threshold and color area solid black threshold value are set to 10% and 90%.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201610954433.XA CN106530289B (en) | 2016-11-03 | 2016-11-03 | A kind of coding clarity machine vision detection method based on dot matrix |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201610954433.XA CN106530289B (en) | 2016-11-03 | 2016-11-03 | A kind of coding clarity machine vision detection method based on dot matrix |
Publications (2)
Publication Number | Publication Date |
---|---|
CN106530289A true CN106530289A (en) | 2017-03-22 |
CN106530289B CN106530289B (en) | 2019-04-30 |
Family
ID=58325444
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201610954433.XA Expired - Fee Related CN106530289B (en) | 2016-11-03 | 2016-11-03 | A kind of coding clarity machine vision detection method based on dot matrix |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN106530289B (en) |
Citations (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JPH0683943A (en) * | 1992-08-31 | 1994-03-25 | Taiyo Yuden Co Ltd | Print inspecting device |
JPH08320931A (en) * | 1995-05-25 | 1996-12-03 | Taiyo Yuden Co Ltd | Method and device for inspecting character |
JP2007326327A (en) * | 2006-06-09 | 2007-12-20 | Kyodo Printing Co Ltd | Inspection method and device of printed matter |
CN104239909A (en) * | 2014-08-11 | 2014-12-24 | 北京捷通华声语音技术有限公司 | Method and device for recognizing images |
CN104268538A (en) * | 2014-10-13 | 2015-01-07 | 江南大学 | Online visual inspection method for dot matrix sprayed code characters of beverage cans |
CN104614384A (en) * | 2015-02-06 | 2015-05-13 | 北京印刷学院 | Quality detecting method of printing product character |
CN105741330A (en) * | 2014-12-10 | 2016-07-06 | 上海新宇箴诚电控科技有限公司 | Code spraying method and code spraying device based on image dot matrix conversion |
-
2016
- 2016-11-03 CN CN201610954433.XA patent/CN106530289B/en not_active Expired - Fee Related
Patent Citations (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JPH0683943A (en) * | 1992-08-31 | 1994-03-25 | Taiyo Yuden Co Ltd | Print inspecting device |
JPH08320931A (en) * | 1995-05-25 | 1996-12-03 | Taiyo Yuden Co Ltd | Method and device for inspecting character |
JP2007326327A (en) * | 2006-06-09 | 2007-12-20 | Kyodo Printing Co Ltd | Inspection method and device of printed matter |
CN104239909A (en) * | 2014-08-11 | 2014-12-24 | 北京捷通华声语音技术有限公司 | Method and device for recognizing images |
CN104268538A (en) * | 2014-10-13 | 2015-01-07 | 江南大学 | Online visual inspection method for dot matrix sprayed code characters of beverage cans |
CN105741330A (en) * | 2014-12-10 | 2016-07-06 | 上海新宇箴诚电控科技有限公司 | Code spraying method and code spraying device based on image dot matrix conversion |
CN104614384A (en) * | 2015-02-06 | 2015-05-13 | 北京印刷学院 | Quality detecting method of printing product character |
Non-Patent Citations (1)
Title |
---|
毛洪辉, 鲍芳: "一种卡片喷码实时检测***", 《自动化技术与应用》 * |
Also Published As
Publication number | Publication date |
---|---|
CN106530289B (en) | 2019-04-30 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN104992449B (en) | Information identification and surface defect online test method based on machine vision | |
CN201159903Y (en) | License plate recognition device | |
CN104318573B (en) | Label defect inspection method | |
CN109308700A (en) | A kind of visual identity defect inspection method based on printed matter character | |
CN101477631B (en) | Method, equipment for extracting target from image and human-machine interaction system | |
WO2018086233A1 (en) | Character segmentation method and device, and element detection method and device | |
CN112964724B (en) | Multi-target multi-region visual detection method and detection system | |
CN105424722A (en) | Bottle cap unqualified product marking method based on machine vision | |
CN104463124A (en) | Milk box spray-printed character recognition method | |
CN106052793A (en) | Machine vision based liquid level sub-quality product marking method | |
CN104881665A (en) | Chip character identification and verification method and apparatus | |
CN104794480A (en) | Method for eliminating milk boxes with defective printed characters | |
CN111591715A (en) | Belt longitudinal tearing detection method and device | |
EP3051461B1 (en) | A computer implemented system and method for extracting and recognizing alphanumeric characters from traffic signs | |
CN104700423A (en) | Method and device for detecting bottle cap | |
CN109693140A (en) | A kind of intelligent flexible production line and its working method | |
CN111652541B (en) | Industrial production monitoring method, system and computer readable storage medium | |
CN114511704B (en) | Spray printing code identification and detection method based on high-speed production line | |
CN103699876A (en) | Method and device for identifying vehicle number based on linear array CCD (Charge Coupled Device) images | |
CN114266751A (en) | AI technology-based product packaging bag coding defect detection method and system | |
CN106530289A (en) | Code spraying definition machine vision detecting method based on lattices | |
CN105791815B (en) | A kind of TV line automatic judging methods | |
CN106442536A (en) | Machine-vision-based IC card automatic thread take-up device and method | |
CN108827974B (en) | Ceramic tile defect detection method and system | |
CN113408388B (en) | Detection method and device based on dot-like code spraying |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
C06 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
GR01 | Patent grant | ||
GR01 | Patent grant | ||
CF01 | Termination of patent right due to non-payment of annual fee | ||
CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20190430 Termination date: 20211103 |