CN110188751A - A kind of M310 nuclear power unit equipment label position image-recognizing method - Google Patents
A kind of M310 nuclear power unit equipment label position image-recognizing method Download PDFInfo
- Publication number
- CN110188751A CN110188751A CN201910417232.XA CN201910417232A CN110188751A CN 110188751 A CN110188751 A CN 110188751A CN 201910417232 A CN201910417232 A CN 201910417232A CN 110188751 A CN110188751 A CN 110188751A
- Authority
- CN
- China
- Prior art keywords
- equipment
- nuclear power
- color mode
- image
- label
- 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
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/30—Information retrieval; Database structures therefor; File system structures therefor of unstructured textual data
- G06F16/33—Querying
- G06F16/3331—Query processing
- G06F16/334—Query execution
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/60—Type of objects
- G06V20/62—Text, e.g. of license plates, overlay texts or captions on TV images
- G06V20/63—Scene text, e.g. street names
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V30/00—Character recognition; Recognising digital ink; Document-oriented image-based pattern recognition
- G06V30/10—Character recognition
- G06V30/14—Image acquisition
- G06V30/148—Segmentation of character regions
- G06V30/153—Segmentation of character regions using recognition of characters or words
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V30/00—Character recognition; Recognising digital ink; Document-oriented image-based pattern recognition
- G06V30/10—Character recognition
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02D—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN INFORMATION AND COMMUNICATION TECHNOLOGIES [ICT], I.E. INFORMATION AND COMMUNICATION TECHNOLOGIES AIMING AT THE REDUCTION OF THEIR OWN ENERGY USE
- Y02D10/00—Energy efficient computing, e.g. low power processors, power management or thermal management
Abstract
The invention belongs to image identification technical fields, and in particular to a kind of M310 nuclear power unit equipment label position image-recognizing method.Include the following steps: to call open source computer vision library；Acquire the image that M310 nuclear power scene includes equipment label BGR color mode；The image of acquisition is saved as into .jpg .png format；BGR color mode is converted into HSV color mode；By reading HSV color mode picture all pixels point in HSV color space, maximum number the case where yellow, green, red, blue, white distribution, judges label shade color in tone；The picture for scanning HSV color mode in S4 again obtains coordinates regional where this kind of color shading；By the coordinates regional of acquisition, pixel interception and cutting are carried out to picture, obtain the image of the BGR color mode of equipment label；Analyze the relative position between four coordinates in the upper left corner, the upper right corner, the lower left corner, the lower right corner.The present invention can reduce the human-equation error probability of nuclear power field personnel, reduce nuclear power plant's operation cost.
Description
Technical field
The invention belongs to image identification technical fields, and in particular to a kind of M310 nuclear power unit equipment label position image knowledge
Other method.
Background technique
There is each equipment of nuclear power unit unique equipment item number to be corresponding to it.Complete M310 nuclear power unit is set
Standby position number includes that 1 unit+3 system of mark number indicates number+2 device types number of+3 equipment chains number.Such as 1CRF001PO
Indicate No. 1 unit circulation 1 pump.The maintenance at nuclear power scene, operation, management activity surround equipment item number expansion.Have
Effect identification equipment item number, can be improved nuclear power and produces movable efficiency, and the accuracy for improving equipment item number identification can effectively be kept away
Exempt from human error and causes unpredictable personal injury and equipment damage to power plant.
There are mainly three types of the methods of the equipment item number of nuclear power industry identification at present:
One, equipment item number identification is carried out by the equipment item number on personnel's visual inspection apparatus label, this kind of method relies on
The behavioural norm and the state of mind of personnel, recognition accuracy and low efficiency；
Two, equipment label position number is scanned and recognized by pasting two dimensional code or bar code on existing equipment label；
Three, by being embedded in radio frequency chip on existing equipment label, equipment label position is carried out by the matching of radiofrequency signal
Number identification.
Separate unit M310 nuclear power unit equipment label quantity passes through second and third kind of method reforming equipment position number at 10,000 or more
The higher cost of identification.At other as power grid, chemical petroleum field have similar to the side for passing through image recognition apparatus label position number
Method, but M310 nuclear power generating equipment label position number can not be identified.
Summary of the invention
The present invention proposes a kind of M310 nuclear power unit equipment label position image-recognizing method, can be to M310 unit length-width ratio
The double-colored equipment label of rectangle that example is 8:5 carries out optical image information identification, the equipment item number in extract equipment label, in data
Facility information corresponding to retrieval facility position number in library.
In order to achieve the above objectives, the technical solution used in the present invention are as follows:
A kind of M310 nuclear power unit equipment label position image-recognizing method, includes the following steps:
S1: open source computer vision library is called；
S2: acquisition M310 nuclear power scene includes the image of equipment label BGR color mode；
S3: the image acquired in S2 is saved as into .jpg .png format；
S4: the BGR color mode in S3 is converted into HSV color mode；
S5: by reading HSV color mode picture all pixels point in HSV color space, yellow in tone, green,
The case where red, blue, white are distributed maximum number, judges label shade color；
S6: by the equipment label shade color obtained in S5, the picture of HSV color mode in S4 is scanned again, is obtained
Coordinates regional where this kind of color shading；
S7: the coordinates regional that will be obtained in S6 carries out pixel interception and cutting to the picture in S2, obtains equipment label
BGR color mode image；
S8: the coordinates regional obtained in S6, analysis the upper left corner, the upper right corner, the lower left corner, the lower right corner four coordinates between
Relative position, judge pixel quantity, levelness shared by the barycentric coodinates of the coordinates regional, rectangular aspect；
S9: using in S8 barycentric coodinates, rectangular aspect, levelness parameter is to the image for obtaining BGR color mode in S7
Handled: interception label region adjusts label levelness, increases the contrast in label region, exports BGR color mode figure
Picture；
S10: optical character identification, output are carried out to BGR color mode image is exported in S9 using optical character identification library
Character string after identification；
S11: judge whether the string length identified in S10 is 9: being, into S12；It is no, return to S2；
S12: whether character string first for judging S10 identification is Arabic numerals: being, into S13；It is no, return to S2；
S13: whether the character string the second to four for judging S10 identification is capitalization English letter: being, into S14；It is no, it returns
Return S2；
S14: whether character string the 5th to seven for judging S10 identification are Arabic numerals: being, into S15；It is no, it returns
Return S2；
S15: whether character string the 8th to nine for judging S10 identification are capitalization English letter: being, into S16；It is no,
Return to S2；
S16: the string search M310 nuclear power unit equipment item number database identified using S10 judges that nine codes are
It is no to there is corresponding equipment in M310 nuclear power unit equipment item number database: to be, into S17；It is no, return to S2；
S17: nine codes of equipment identified, equipment item number corresponding to M310 nuclear power unit equipment item number database are exported
Equipment state be read and writen.
The S1: opencv open source computer vision library is called.
The S10: optical character is carried out to output BGR color mode image in S9 using ORC optical character identification library
Identification.
In the S16:M310 nuclear power unit equipment item number database comprising all M310 nuclear power unit equipment item numbers and
The description of equipment Chinese, equipment installation position, the security level of equipment, the machine group number of equipment, system where equipment, device type,
Equipment state.
It is obtained by the present invention to have the beneficial effect that
Nuclear power production activity is carried out using M310 nuclear power unit equipment label position image-recognizing method, without to existing mark
Board is transformed, and method for distinguishing is known in this kind of equipment label position number, reduces the same of the human-equation error probability of nuclear power field personnel
When, reduce nuclear power plant's operation cost.
Detailed description of the invention
Fig. 1 is M310 nuclear power unit equipment label position image-recognizing method flow chart.
Specific embodiment
The present invention is described in detail with specific embodiment below in conjunction with the accompanying drawings.
As shown in Figure 1, M310 nuclear power unit equipment label of the present invention position image-recognizing method includes the following steps:
S1: opencv (open source computer vision library) is called；
S2: acquisition M310 nuclear power scene includes the image of equipment label BGR color mode；
S3: the image acquired in S2 is saved as into .jpg .png format；
S4: the BGR color mode in S3 is converted into HSV color mode；
S5: yellow in tone (H), green by reading HSV color mode picture all pixels point in HSV color space
The case where color, red, blue, white are distributed maximum number, judges label shade color；
S6: by the equipment label shade color obtained in S5, the picture of HSV color mode in S4 is scanned again, is obtained
Coordinates regional where this kind of color shading；
S7: the coordinates regional that will be obtained in S6 carries out pixel interception and cutting to the picture in S2, obtains equipment label
BGR color mode image.
S8: the coordinates regional obtained in S6, analysis the upper left corner, the upper right corner, the lower left corner, the lower right corner four coordinates between
Relative position, judge pixel quantity, levelness shared by the barycentric coodinates of the coordinates regional, rectangular aspect.
S9: using in S8 barycentric coodinates, rectangular aspect, levelness parameter to the image for obtaining BGR color mode in S7,
Handled: interception label region adjusts label levelness, increases the contrast in label region, exports BGR color mode figure
Picture.
S10: carrying out optical character identification to output BGR color mode image in S9 using (optical character identification) library ORC,
Character string after output identification；
S11: judge whether the string length identified in S10 is 9: being, into S12；It is no, return to S2.
S12: whether character string first for judging S10 identification is Arabic numerals: being, into S13；It is no, return to S2.
S13: whether the character string the second to four for judging S10 identification is capitalization English letter: being, into S14；It is no, it returns
Return S2.
S14: whether character string the 5th to seven for judging S10 identification are Arabic numerals: being, into S15；It is no, it returns
Return S2.
S15: whether character string the 8th to nine for judging S10 identification are capitalization English letter: being, into S16；It is no,
Return to S2.
S16: the string search M310 nuclear power unit equipment item number database identified using S10.M310 nuclear power unit is set
It include all M310 nuclear power unit equipment item numbers and the description of equipment Chinese, equipment installation position, equipment in standby position number library
System, device type, equipment state where security level, the machine group number of equipment, equipment.Judge nine codes whether in M310 core
There is corresponding equipment in motor group equipment item number database: being, into S17；It is no, return to S2.
S17: nine codes of equipment identified, equipment item number corresponding to M310 nuclear power unit equipment item number database are exported
Equipment state be read and writen.
Claims (4)
1. a kind of M310 nuclear power unit equipment label position image-recognizing method, characterized by the following steps:
S1: open source computer vision library is called；
S2: acquisition M310 nuclear power scene includes the image of equipment label BGR color mode；
S3: the image acquired in S2 is saved as into .jpg .png format；
S4: the BGR color mode in S3 is converted into HSV color mode；
S5: by reading HSV color mode picture all pixels point in HSV color space, yellow in tone, green, red,
The case where blue, white are distributed maximum number, judges label shade color；
S6: by the equipment label shade color obtained in S5, the picture of HSV color mode in S4 is scanned again, obtains this kind
Coordinates regional where color shading；
S7: the coordinates regional that will be obtained in S6 carries out pixel interception and cutting to the picture in S2, obtains equipment label
The image of BGR color mode；
S8: the coordinates regional obtained in S6 analyzes the phase between four coordinates in the upper left corner, the upper right corner, the lower left corner, the lower right corner
To position, pixel quantity, levelness shared by the barycentric coodinates of the coordinates regional, rectangular aspect are judged；
S9: the image that BGR color mode is obtained in S7 is carried out using barycentric coodinates, rectangular aspect, the levelness parameter in S8
Processing: interception label region adjusts label levelness, increases the contrast in label region, exports BGR color mode image；
S10: optical character identification, output identification are carried out to BGR color mode image is exported in S9 using optical character identification library
Character string afterwards；
S11: judge whether the string length identified in S10 is 9: being, into S12；It is no, return to S2；
S12: whether character string first for judging S10 identification is Arabic numerals: being, into S13；It is no, return to S2；
S13: whether the character string the second to four for judging S10 identification is capitalization English letter: being, into S14；It is no, return to S2；
S14: whether character string the 5th to seven for judging S10 identification are Arabic numerals: being, into S15；It is no, return to S2；
S15: whether character string the 8th to nine for judging S10 identification are capitalization English letter: being, into S16；It is no, it returns
S2；
S16: using S10 identify string search M310 nuclear power unit equipment item number database, judge nine codes whether
There is corresponding equipment in M310 nuclear power unit equipment item number database: being, into S17；It is no, return to S2；
S17: exporting nine codes of equipment identified, sets to the corresponding equipment item number of M310 nuclear power unit equipment item number database
Standby state is read and writen.
2. M310 nuclear power unit equipment label according to claim 1 position image-recognizing method, it is characterised in that: described
S1: call opencv increase income computer vision library.
3. M310 nuclear power unit equipment label according to claim 1 position image-recognizing method, it is characterised in that: described
S10: using ORC optical character identification library in S9 export BGR color mode image carry out optical character identification.
4. M310 nuclear power unit equipment label according to claim 1 position image-recognizing method, it is characterised in that: described
S16:M310 nuclear power unit equipment item number database in comprising all M310 nuclear power unit equipment item numbers and equipment Chinese description,
System, device type, equipment state where equipment installation position, the security level of equipment, the machine group number of equipment, equipment.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910417232.XA CN110188751B (en) | 2019-05-20 | 2019-05-20 | M310 nuclear power unit equipment label position number image recognition method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910417232.XA CN110188751B (en) | 2019-05-20 | 2019-05-20 | M310 nuclear power unit equipment label position number image recognition method |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN110188751A true CN110188751A (en) | 2019-08-30 |
| CN110188751B CN110188751B (en) | 2023-01-03 |
Family
ID=67716954
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910417232.XA Active CN110188751B (en) | 2019-05-20 | 2019-05-20 | M310 nuclear power unit equipment label position number image recognition method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110188751B (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111738326A (en) * | 2020-06-16 | 2020-10-02 | 中国工商银行股份有限公司 | Sentence granularity marking training sample generation method and device |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2008287414A (en) * | 2007-05-16 | 2008-11-27 | B-Core Inc | Optical recognition code recognition system, method, and program |
| CN101561871A (en) * | 2009-02-17 | 2009-10-21 | 昆明理工大学 | Method for recognizing manually-set road sign in agricultural machine visual navigation |
| CN104156703A (en) * | 2014-08-01 | 2014-11-19 | 武汉工程大学 | License plate location method and system based on color texture |
| CN105005780A (en) * | 2015-06-29 | 2015-10-28 | 叶秀兰 | License plate identification method |
| CN106295643A (en) * | 2016-08-11 | 2017-01-04 | 万永秀 | Automobile license plate automatic recognition method |
| CN109492630A (en) * | 2018-10-26 | 2019-03-19 | 信雅达系统工程股份有限公司 | A method of the word area detection positioning in the financial industry image based on deep learning |
-
2019
- 2019-05-20 CN CN201910417232.XA patent/CN110188751B/en active Active
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2008287414A (en) * | 2007-05-16 | 2008-11-27 | B-Core Inc | Optical recognition code recognition system, method, and program |
| CN101561871A (en) * | 2009-02-17 | 2009-10-21 | 昆明理工大学 | Method for recognizing manually-set road sign in agricultural machine visual navigation |
| CN104156703A (en) * | 2014-08-01 | 2014-11-19 | 武汉工程大学 | License plate location method and system based on color texture |
| CN105005780A (en) * | 2015-06-29 | 2015-10-28 | 叶秀兰 | License plate identification method |
| CN106295643A (en) * | 2016-08-11 | 2017-01-04 | 万永秀 | Automobile license plate automatic recognition method |
| CN109492630A (en) * | 2018-10-26 | 2019-03-19 | 信雅达系统工程股份有限公司 | A method of the word area detection positioning in the financial industry image based on deep learning |
Non-Patent Citations (2)
| Title |
|---|
| 李谊等: "采用混合神经网络高精确度提取机票字符", 《计算机工程与应用》 * |
| 郁浩等: "基于图像处理的钢包标牌识别方法", 《电子科技》 * |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111738326A (en) * | 2020-06-16 | 2020-10-02 | 中国工商银行股份有限公司 | Sentence granularity marking training sample generation method and device |
Also Published As
| Publication number | Publication date |
|---|---|
| CN110188751B (en) | 2023-01-03 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Antonacopoulos et al. | ICDAR 2009 page segmentation competition | |
| Wang et al. | Character location in scene images from digital camera | |
| USRE44982E1 (en) | Mixed code, and method and apparatus for generating the same | |
| KR100339691B1 (en) | Apparatus for recognizing code and method therefor | |
| US7327882B2 (en) | Method and device for character location in images from digital camera | |
| US10311322B2 (en) | Character information recognition method based on image processing | |
| CN101295359B (en) | Image processing method and image processing apparatus | |
| EP2833288A1 (en) | Face calibration method and system, and computer storage medium | |
| EP1870858A2 (en) | Method of classifying colors of color based image code | |
| CN107145487A (en) | Image search method and device | |
| Zahour et al. | Text line segmentation of historical arabic documents | |
| CN105069466A (en) | Pedestrian clothing color identification method based on digital image processing | |
| US20040086153A1 (en) | Methods and systems for recognizing road signs in a digital image | |
| Zhang et al. | A novel text detection system based on character and link energies | |
| CN103593695A (en) | Method for positioning DPM two-dimension code area | |
| CN102663380A (en) | Method for identifying character in steel slab coding image | |
| KR20030029187A (en) | Face detecting method depend on image | |
| Anis et al. | Digital electric meter reading recognition based on horizontal and vertical binary pattern | |
| KR101255729B1 (en) | Method on Sub Image Retrieval Using Local Alignment | |
| CN105335744A (en) | One-dimensional code region location based on image backbone extraction strip distribution features | |
| CN112613339A (en) | Automatic identification and examination method and device for electrical drawings | |
| CN110188751A (en) | A kind of M310 nuclear power unit equipment label position image-recognizing method | |
| Damayanti et al. | Indonesian license plate recognition based on area feature extraction | |
| CN112329791B (en) | Automatic extraction method for hyperspectral image water area | |
| CN111414960A (en) | Artificial intelligence image feature extraction system and feature identification method thereof |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | 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 |