CN105893912A - Direct marking character detection method - Google Patents
Direct marking character detection method Download PDFInfo
- Publication number
- CN105893912A CN105893912A CN201410589063.5A CN201410589063A CN105893912A CN 105893912 A CN105893912 A CN 105893912A CN 201410589063 A CN201410589063 A CN 201410589063A CN 105893912 A CN105893912 A CN 105893912A
- Authority
- CN
- China
- Prior art keywords
- character
- result
- ocr
- product
- recognition device
- 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.)
- Pending
Links
Landscapes
- Character Discrimination (AREA)
Abstract
The invention discloses a direct marking character detection method. The method comprises character identification equipment and character discrimination equipment. The character identification equipment identifies character information on a product and outputs to the character discrimination equipment. The character discrimination equipment collects identification information transmitted by the character identification equipment and carries out intelligent discrimination. A qualified rate or accuracy of an identification result reaches a preset value, the product is determined to be qualified; otherwise, the product is determined to be unqualified. And an unqualified result is output to an audible and visual alarm, rejecting equipment and stop line equipment so as to complete product detection and detection result processing. By using the method of the invention, a problem that an existing OCR product can only carry out one-to-one processing on an output result and can not carry out integration statistics analysis on a dispersive result is solved.
Description
[technical field]
The present invention relates to the direct marker field on on-line checking article, especially relate to the inspection of direct tab character
Survey method.
[background technology]
At present at product packaging industry, according to the relevant laws and regulations of country, be required for aborning real-time spray printing,
Etching, burn etc. the character with the date as main contents and carry out the information such as date of manufacture of representative products.If
Mark quality does not reaches the requirement (can identification, accurateness) of relevant laws and regulations and is then considered as substandard product, no
Allow commercially to circulate.
Due to reasons such as existing production equipment, technique and way to manages, it is impossible to ensure on each product
Tab character complies fully with standard.Due to the enormous amount of product, current producer is all by manually taking out
The mode of inspection is tested, and this mode accuracy rate is low, and consumption is artificial, weak effect, it is impossible to ensure each product
Product all detect.
Direct labelling that character recognition system is mainly on on-line checking article (spray printing, is pasted, prints, is lost
Carve etc.) character qualities, character qualities comprises character position, character can identification and the correctness of character.
Character position: refer to whether character position on product meets relevant requirement
Character can identification: refer to character for human eye can identification degree and the font of character, font size
Whether meet relevant criterion
Character correctness: refer to whether the implication representated by character meets predefined content, such as day
Phase information, whether the date and time information of mark is consistent with actual date and time information.
It is raw that the product of current existing OCR and technology do not reach enterprise for the recognition effect of real-time spray printing character
The actual demand produced, can produce substantial amounts of identification mistake, cause system not come into operation.
The character qualities of the most on-the-spot labelling cannot ensure due to a variety of causes, the uniformity of quality.
Specifically, the widely different current existing OCR technique identification effect to the character of on-the-spot labelling is repeated
Fruit can not meet the demand producing reality.Mainly show as identifying mistake and recognition failures.If with single
The result of character recognition device is as final result, it will produces substantial amounts of erroneous judgement and fails to judge, it is impossible to meet raw
Produce actual demand.
[summary of the invention]
Pin disadvantage mentioned above of the present invention, uses direct tab character detection method, enters the spray printing character on product
The man-to-man real-time detection of row, in the way of substituting original artificial sampling observation.Improve product outgoing and
Detection efficiency, reduces hand labor intensity..
For reaching above-mentioned purpose, the technical solution adopted for the present invention to solve the technical problems is:
Including character recognition device and character discriminating device, it is characterised in that: described character recognition device, know
Character information on other product, and export character discriminating device, described character discriminating device is collected character and is known
The identification information of other equipment transmission, carries out intelligent distinguishing, when qualification rate or the accuracy rate of recognition result, reaches
During to numerical value set in advance, it is determined that product is qualified, otherwise, it is determined that be defective, and by described defective
Result exports to audible and visual alarm, removal equipment, stops line equipment, completes the detection to product and to detection
The process of result.
Preferably, the result that described character recognition device returns includes:
A: correctly identify all of character, and export correct character number;
B: correctly identify partial character, Unidentified character output identification symbol (?), but character: quantity
Correctly;
C: correctly identify partial character, Unidentified character output identification symbol (?), character quantity is incorrect;
D: correctly identifying partial character, but character misregistration is true, character quantity is correct;
E: correctly identifying partial character, but character misregistration is true, character quantity is incorrect;
F: mistake identify partial character, character quantity is correct;
G: mistake identify partial character, character quantity is incorrect;
H: entirely without identifying character, but output character quantity is correct.(character number can be detected);
I: entirely without identifying character, output character quantity is incorrect;
J:OCR failure, returns failure information.
Concrete, when receiving the result that described character recognition device returns, set intelligent distinguishing rule such as
Under:
In N number of result, if OCR returns this rate unsuccessfully and is more than OCR_NOK%, then differentiate coding disappearance
Or the system failure;Further,
In N number of result, if OCR return character number is more than less than the ratio of certain threshold value
OCR_Number_NOK%, then differentiate coding disappearance or the system failure, and N is the natural number more than 1;
Wherein, OCR_NOK%:OCR failure threshold value,
OCR_Number_NOK%:OCR return character amount threshold.
Concrete, when receiving the result that described character recognition device returns, set intelligent distinguishing rule such as
Under:
In N number of result, in the result that can recognize that, key character is correctly distinguished and this correct rate
Less than certain threshold value, then differentiate defective;
Key character is correctly distinguished and this rate of mistake is more than certain threshold value, then differentiate defective;
If this rate of the identification of key character is less than certain threshold value, then produce warning information.
Concrete, it is characterised in that when receiving the result that described character recognition device returns, set intelligence
Energy decision rule is as follows:
In N number of testing result, in the result of energy return character number, if returning number to be unsatisfactory for requirement
Ratio more than certain threshold value, then be judged as defective.
Preferably, described character recognition device uses the Vision Hawk product of U.S. Microscan.
Preferably, digital camera and image processing section are integrated by described character recognition device, can
With directly by serial ports, Ethernet interface output character recognition result.
Preferably, described character discriminating device is made up of a field controller able to programme and corresponding software.
Preferably, described character discriminating device, based on embedded wince system, is applied to industry spot
Small-sized industrial control equipment.
The present invention solves in existing OCR technique and product and the situation of existing direct tab character quality
Under, the spray printing character of product is carried out the problem that quality (can identification and accuracy) detects.
[accompanying drawing explanation]
The fundamental diagram of the character detection method that Fig. 1 provides for the embodiment of the present invention;
The device appearance figure of the character recognition device that Fig. 2 provides for the embodiment of the present invention;
The outside drawing of the character discriminating device that Fig. 3 provides for the embodiment of the present invention.
[detailed description of the invention]
In order to make the purpose of the present invention, technical scheme and advantage clearer, below in conjunction with embodiment,
The present invention is further elaborated.Should be appreciated that described herein is only the part of the present invention
Embodiment rather than whole embodiments.Based on the embodiment in the present invention, those of ordinary skill in the art
The every other embodiment obtained on the premise of not making creative work, broadly falls into present invention protection
Scope.
The major function of the system that the embodiment of the present invention provides is:
Coding content differentiates: the place of production, product line and date and time information are the most correct;
Coding position differentiates: coding is gone up the most in place;
Coding character disappearance detection: have NULI character to lack (imperfect character or imperfect character string);
Wrap detection in vain: with or without character on spray printing.
One of purpose of the embodiment of the present invention, is to provide direct tab character detection method, as it is shown in figure 1,
Specifically include with lower part:
Including character recognition device and character discriminating device, described character recognition device, identify the word on product
Symbol information, and export character discriminating device, described character discriminating device collects character recognition device transmission
Identification information, carries out intelligent distinguishing, when qualification rate or the accuracy rate of recognition result, reaches set in advance
During numerical value, it is determined that product is qualified, otherwise, it is determined that be defective, and described defective result is exported to sound
Light alarm, removal equipment, stop line equipment, complete the detection to product and the process to testing result,
It is characterized in that, the result that described character recognition device returns includes:
A: correctly identify all of character, and export correct character number;
B: correctly identify partial character, Unidentified character output identification symbol (?), but character: quantity
Correctly;
C: correctly identify partial character, Unidentified character output identification symbol (?), character quantity is incorrect;
D: correctly identifying partial character, but character misregistration is true, character quantity is correct;
E: correctly identifying partial character, but character misregistration is true, character quantity is incorrect;
F: mistake identify partial character, character quantity is correct;
G: mistake identify partial character, character quantity is incorrect;
H: entirely without identifying character, but output character quantity is correct.(character number can be detected);
I: separately going out character entirely without knowing, output character quantity is incorrect;
J:OCR failure, returns failure information.
Concrete, when receiving the result that described character recognition device returns, set intelligent distinguishing rule such as
Under:
In N number of result, if OCR returns this rate unsuccessfully and is more than OCR_NOK%, then differentiate coding disappearance
Or the system failure;Further,
In N number of result, if OCR return character number is more than less than the ratio of certain threshold value
OCR_Number_NOK%, then differentiate coding disappearance or the system failure, and N is the natural number more than 1;
Wherein, OCR_NOK%:OCR failure threshold value,
OCR_Number_NOK%:OCR return character amount threshold.
Concrete, when receiving the result that described character recognition device returns, set intelligent distinguishing rule such as
Under:
In N number of result, in the result that can recognize that, key character is correctly distinguished and this correct rate
Less than certain threshold value, then differentiate defective;
Key character is correctly distinguished and the ratio of mistake is more than certain threshold value, then differentiate defective;
If this rate of the identification of key character is less than certain threshold value, then produce warning information.
Concrete, it is characterised in that when receiving the result that described character recognition device returns, set intelligence
Another rule can be sentenced as follows:
In N number of testing result, in the result of energy return character number, if returning number to be unsatisfactory for requirement
This rate more than certain threshold value, then be judged as defective.
As in figure 2 it is shown, described character recognition device uses the Vision Hawk product of U.S. Microscan.
Preferably, digital camera and image processing section are integrated by described character recognition device, can
With directly by serial ports, Ethernet interface output character recognition result.
As it is shown on figure 3, described character discriminating device is by a field controller able to programme and corresponding software group
Become.
Preferably, described character discriminating device, based on embedded wince system, is applied to industry spot
Small-sized industrial control equipment.
By field test, an accuracy of direct tab character can be reached by character recognition device
60%~95%, and the main cause of recognition failures is because the OCR technique prescription to character higher than real
The prescription on border.From this point of view, it is because equipment itself and result in a lot of erroneous judgements.
In producing reality, occur that character underproof situation great majority are because equipment fault, human error
And the substantial amounts of continuous print mistake of the product caused, and the underproof feelings of individual characters occur in process of production
Condition is considerably less.
For case above, this programme by character recognition data sampling analysis, the way of intelligent distinguishing,
Solving emphatically situation high-volume mistake continuously occur, (character disappearance, position are not the reason of this phenomenon occurs
Qualified, character recognition degree is defective etc.) by manually differentiating.For the mistake of individual product, made mistakes
Time period, manual detection solve by mistake.
Compared with prior art, improve accuracy and the real-time of Product checking, improve the excellent of work efficiency
Point.The detailed description of the invention of present invention described above, is not intended that limiting the scope of the present invention.Appoint
What changes and deformation accordingly according to various other done by the technology design of the present invention, should be included in originally
In invention scope of the claims.
Claims (5)
1. direct tab character detection method, including character recognition device and character discriminating device, it is characterized in that: described character recognition device, identify the character information on product, and export character discriminating device, described character discriminating device collects the identification information of character recognition device transmission, carry out intelligent distinguishing, qualification rate or accuracy rate when recognition result, when reaching numerical value set in advance, judge that product is qualified, otherwise, it is judged to defective, and described defective result is exported to audible and visual alarm, removal equipment, stop line equipment, complete the detection to product, and the process to testing result.
2. character detection method as claimed in claim 1, it is characterised in that the result that described character recognition device returns includes:
A: correctly identify all of character, and export correct character number;
B: correctly identify partial character, Unidentified character output identification symbol, but character: quantity is correct;
C: correctly identifying partial character, Unidentified character output identification symbol, character quantity is incorrect;
D: correctly identifying partial character, but character misregistration is true, character quantity is correct;
E: correctly identifying partial character, but character misregistration is true, character quantity is incorrect;
F: mistake identify partial character, character quantity is correct;
G: mistake identify partial character, character quantity is incorrect;
H: entirely without identifying character, but output character quantity is correct;
I: entirely without identifying character, output character quantity is incorrect;
J:OCR failure, returns failure information.
3. character detection method as claimed in claim 1 or 2, it is characterised in that when receiving the result that described character recognition device returns, sets intelligent distinguishing rule as follows:
In N number of result, if OCR returns failed ratio and is more than OCR_NOK%, then differentiate coding disappearance or the system failure;Further,
In N number of result, if OCR return character number is more than OCR_Number_NOK% less than the ratio of certain threshold value, then differentiating coding disappearance or the system failure, N is the natural number more than 1;
Wherein, OCR_NOK%:OCR failure threshold value,
OCR_Number_NOK%:OCR return character amount threshold.
4. character detection method as claimed in claim 1 or 2, it is characterised in that when receiving the result that described character recognition device returns, sets intelligent distinguishing rule as follows:
In N number of result, in the result that can recognize that, key character is correctly distinguished and correct ratio is less than certain threshold value, then differentiate defective;
Key character is correctly distinguished and the ratio of mistake is more than certain threshold value, then differentiate defective;
If the identification ratio of key character is less than certain threshold value, then produce warning information.
5. character detection method as claimed in claim 1 or 2, it is characterised in that when receiving the result that described character recognition device returns, sets intelligent distinguishing rule as follows:
In N number of testing result, in the result of energy return character number, if the ratio that return number is unsatisfactory for requiring is more than certain threshold value, then it is judged as defective.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201410589063.5A CN105893912A (en) | 2014-10-29 | 2014-10-29 | Direct marking character detection method |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201410589063.5A CN105893912A (en) | 2014-10-29 | 2014-10-29 | Direct marking character detection method |
Publications (1)
Publication Number | Publication Date |
---|---|
CN105893912A true CN105893912A (en) | 2016-08-24 |
Family
ID=57001593
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201410589063.5A Pending CN105893912A (en) | 2014-10-29 | 2014-10-29 | Direct marking character detection method |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN105893912A (en) |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110039916A (en) * | 2019-04-16 | 2019-07-23 | 珠海格力电器股份有限公司 | Printing quality detection method, computer device, and computer-readable storage medium |
Citations (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101782896A (en) * | 2009-01-21 | 2010-07-21 | 汉王科技股份有限公司 | PDF character extraction method combined with OCR technology |
CN103049743A (en) * | 2012-12-27 | 2013-04-17 | 天津普达软件技术有限公司 | Detection system and detection method for characters at bottom of bowl |
-
2014
- 2014-10-29 CN CN201410589063.5A patent/CN105893912A/en active Pending
Patent Citations (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101782896A (en) * | 2009-01-21 | 2010-07-21 | 汉王科技股份有限公司 | PDF character extraction method combined with OCR technology |
CN103049743A (en) * | 2012-12-27 | 2013-04-17 | 天津普达软件技术有限公司 | Detection system and detection method for characters at bottom of bowl |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110039916A (en) * | 2019-04-16 | 2019-07-23 | 珠海格力电器股份有限公司 | Printing quality detection method, computer device, and computer-readable storage medium |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN110288032B (en) | Vehicle driving track type detection method and device | |
CN108469438A (en) | A kind of printed matter detection method, device, equipment and storage medium | |
CN206095981U (en) | Bottle lid coding quality inspection system | |
CN113744244B (en) | Online visual detection system for measuring distance from edge of lithium battery pole piece to edge of pole lug | |
CN111127517A (en) | Production line product positioning method based on monitoring video | |
CN104992449A (en) | Information identification and surface defect on-line detection method based on machine visual sense | |
CN112804336B (en) | Fault detection method, device, system and computer readable storage medium | |
CN107491752A (en) | Ship board character recognition method, device in a kind of natural scene based on deep learning | |
CN103308523B (en) | Method for detecting multi-scale bottleneck defects, and device for achieving method | |
CN105426908A (en) | Convolutional neural network based substation attribute classification method | |
CN104268134A (en) | Subjective and objective classifier building method and system | |
CN102157024A (en) | System and method for on-line secondary detection checking of checking data of large-sheet checking machine | |
CN114312004B (en) | Lottery printing quality monitoring method and system | |
CN114219799A (en) | Intelligent manufacturing defective product analysis method and system | |
CN103258218B (en) | Masking-out detects skeleton generating method, device, defect inspection method and device | |
CN102967265B (en) | Based on the automobile connecting bar splitter detection method of machine vision | |
CN203337559U (en) | Digitized full-automatic detection system for quality of presswork | |
CN106528665A (en) | AOI equipment test file searching method and system | |
CN114895634A (en) | Product production line automatic control system based on machine vision | |
CN105893912A (en) | Direct marking character detection method | |
TWI755953B (en) | Automatic detection system and operation method thereof | |
CN114323042A (en) | Map data processing method and device, nonvolatile storage medium and processor | |
CN103247105B (en) | Novel sorter waste treatment process and device | |
CN204759441U (en) | Two -dimensional code check out test set and corresponding detecting system | |
CN106096634B (en) | Fault detection method based on Adaptive windowing mental arithmetic method with interval halving algorithm |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
C06 | Publication | ||
PB01 | Publication | ||
C10 | Entry into substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
C10 | Entry into substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
DD01 | Delivery of document by public notice | ||
DD01 | Delivery of document by public notice |
Addressee: Mao Zhihong Document name: Notification of Passing Examination on Formalities |
|
RJ01 | Rejection of invention patent application after publication | ||
RJ01 | Rejection of invention patent application after publication |
Application publication date: 20160824 |