CN108009547A - A kind of nameplate recognition methods of substation equipment and device - Google Patents
A kind of nameplate recognition methods of substation equipment and device Download PDFInfo
- Publication number
- CN108009547A CN108009547A CN201711436199.2A CN201711436199A CN108009547A CN 108009547 A CN108009547 A CN 108009547A CN 201711436199 A CN201711436199 A CN 201711436199A CN 108009547 A CN108009547 A CN 108009547A
- Authority
- CN
- China
- Prior art keywords
- measured
- nameplate
- image
- contour
- substation equipment
- 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
Classifications
-
- 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
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/214—Generating training patterns; Bootstrap methods, e.g. bagging or boosting
-
- 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
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- Multimedia (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Evolutionary Computation (AREA)
- Evolutionary Biology (AREA)
- General Engineering & Computer Science (AREA)
- Bioinformatics & Computational Biology (AREA)
- Artificial Intelligence (AREA)
- Life Sciences & Earth Sciences (AREA)
- Image Analysis (AREA)
- Image Processing (AREA)
Abstract
Nameplate recognition methods and device the invention discloses a kind of substation equipment.The described method includes：Obtain the scene image to be measured containing substation equipment nameplate to be measured；Using default Boundary extracting algorithm, edge extracting processing is carried out to scene image to be measured, obtains corresponding contour images to be measured, and position the nameplate contour area in contour images to be measured；Nameplate contour area to be measured is cut and correction process, obtain nameplate region image to be measured；Using default convolutional neural networks, the character in nameplate region image to be measured is identified.The present invention can identify the name plate information of different power equipments in the non-contact case by the nameplate recognition methods of offer, the significant increase efficiency and quality of inspection, and the identification to name plate information has very strong environmental suitability and antijamming capability.
Description
Technical field
The present invention relates to image identification technical field, the nameplate recognition methods of more particularly to a kind of substation and device.
Background technology
The nameplate of substation is generally in outdoor at present, since equipment has certain safe distance, it is necessary to be spaced a distance
Carry out nameplate image shooting, since nameplate majority is made of metal material, be placed in throughout the year in outdoor environment, be subject to outdoor dust,
Steel corrosion influence, obtained nameplate picture blur, stroke disconnection；Again since artificial shooting is difficult to the inscription for standardizing, capturing
Board picture is also more with Horizontal Deformation, the perspective deformation even reflective influence of nameplate, therefore, it is difficult to be known with traditional optical character
Not these nameplates of (Optical Character Recognition, referred to as " OCR ") method Direct Recognition.
The content of the invention
In order to solve problem of the prior art, an embodiment of the present invention provides a kind of nameplate recognition methods of substation equipment
And device.The technical solution is as follows：
On the one hand, an embodiment of the present invention provides a kind of nameplate recognition methods of substation equipment, the described method includes：
Obtain the scene image to be measured containing substation equipment nameplate to be measured；
Using default Boundary extracting algorithm, edge extracting processing is carried out to scene image to be measured, is obtained corresponding to be measured
Contour images, and position the nameplate contour area in contour images to be measured；
Nameplate contour area to be measured is cut and correction process, obtain nameplate region image to be measured；
Using default convolutional neural networks, the character in nameplate region image to be measured is identified.
It is described to use default Boundary extracting algorithm in nameplate recognition methods provided in an embodiment of the present invention, to be measured
Scene image carries out edge extracting processing, obtains corresponding contour images to be measured, including：
Using default (Holistically-Nested Edge Detection, referred to as " HED ") network model, treat
Survey scene image and carry out edge extracting processing, obtain corresponding contour images to be measured, and position the nameplate in contour images to be measured
Contour area.
It is described that nameplate contour area to be measured is cut and rectified in nameplate recognition methods provided in an embodiment of the present invention
Positive processing, obtains nameplate region image to be measured, including：
Using Hough transformation line detection method, nameplate contour area to be measured is cut, and uses perspective transform, it is right
Nameplate contour area to be measured after cutting is corrected, and obtains nameplate region image to be measured.
It is described to use default convolutional neural networks in nameplate recognition methods provided in an embodiment of the present invention, to be measured
Nameplate region image in character be identified, including：
Using LeNet-5 convolutional neural networks, the character in nameplate region image to be measured is identified, it is described
LeNet-5 convolutional neural networks are often trained using substation equipment by the use of nameplate character as training data.
It is described before localization process is carried out to detection image in nameplate recognition methods provided in an embodiment of the present invention
Method further includes：
Using single scale Retinex (i.e. retina cerebral cortex theoretical) algorithm for image enhancement, to scene image to be measured into
Row illumination removes pretreatment.
On the other hand, an embodiment of the present invention provides a kind of nameplate identification device of substation equipment, described device to include：
Acquisition module, for obtaining the scene image to be measured containing substation equipment nameplate to be measured；
First processing module, for using default Boundary extracting algorithm, carries out at edge extracting scene image to be measured
Reason, obtains corresponding contour images to be measured, and positions the nameplate contour area in contour images to be measured；
Second processing module, for being cut to nameplate contour area to be measured and correction process, obtains nameplate to be measured
Area image；
Identification module, for using default convolutional neural networks, carries out the character in nameplate region image to be measured
Identification.
In nameplate identification device provided in an embodiment of the present invention, the first processing module, is additionally operable to using default
HED network models, edge extracting processing is carried out to scene image to be measured, obtains corresponding contour images to be measured, and position to be measured
Nameplate contour area in contour images.
In nameplate identification device provided in an embodiment of the present invention, the Second processing module, is additionally operable to become using Hough
Line detection method is changed, nameplate contour area to be measured is cut, and uses perspective transform, to the nameplate wheel to be measured after cutting
Wide region is corrected, and obtains nameplate region image to be measured.
In nameplate identification device provided in an embodiment of the present invention, the identification module, is additionally operable to use LeNet-5 convolution
Neutral net, is identified the character in nameplate region image to be measured, and the LeNet-5 convolutional neural networks use power transformation
Station equipment is often trained by the use of nameplate character as training data.
In nameplate identification device provided in an embodiment of the present invention, further include：
3rd processing module, for using single scale Retinex algorithm for image enhancement, illumination is carried out to scene image to be measured
Remove pretreatment.
The beneficial effect that technical solution provided in an embodiment of the present invention is brought is：
By obtaining the scene image to be measured containing substation equipment nameplate to be measured；Using default Boundary extracting algorithm,
Edge extracting processing is carried out to scene image to be measured, obtains corresponding contour images to be measured, and position in contour images to be measured
Nameplate contour area；Nameplate contour area to be measured is cut and correction process, obtain nameplate region image to be measured；Using
Default convolutional neural networks, are identified the character in nameplate region image to be measured.So inscription of the substation equipment
Board recognition methods, can identify the name plate information of different power equipments in the non-contact case, significant increase inspection
Efficiency and quality, and the identification to name plate information has very strong environmental suitability and antijamming capability.In addition, equipment is engraved
Board image information carries out automatic identification and automatically extracts, and can save the workload of teams and groups personnel tradition machinery formula data acquisition,
Greatly improve the efficiency and quality of inspection.
Brief description of the drawings
To describe the technical solutions in the embodiments of the present invention more clearly, make required in being described below to embodiment
Attached drawing is briefly described, it should be apparent that, drawings in the following description are only some embodiments of the present invention, for
For those of ordinary skill in the art, without creative efforts, other can also be obtained according to these attached drawings
Attached drawing.
Fig. 1 is a kind of nameplate recognition methods flow chart for substation equipment that the embodiment of the present invention one provides；
Fig. 2 is a kind of nameplate identification device structure diagram of substation equipment provided by Embodiment 2 of the present invention.
Embodiment
To make the object, technical solutions and advantages of the present invention clearer, below in conjunction with attached drawing to embodiment party of the present invention
Formula is described in further detail.
Embodiment one
An embodiment of the present invention provides a kind of nameplate recognition methods of substation equipment, for identifying substation equipment
Name plate information, referring to Fig. 1, this method can include：
Step S11, obtains the scene image to be measured containing substation equipment nameplate to be measured.
In the present embodiment, it is not true due to style of shooting when obtaining the nameplate image of substation equipment to be detected
Qualitative, acquired nameplate image not only includes nameplate to be measured, it is also possible to therefore, first comprising the scene residing for nameplate
Obtain the scene image to be measured containing substation equipment nameplate to be measured.
Step S12, using single scale Retinex algorithm for image enhancement, carries out scene image to be measured illumination and removes pre- place
Reason.
In the present embodiment, nameplate image often occur uneven illumination it is even in addition as caused by flash lamp it is strong reflective etc.
Phenomenon, in order to improve picture quality, can use single scale Retinex algorithm for image enhancement to carry out illumination to image and remove pre- place
Reason.
Step S13, using default Boundary extracting algorithm, carries out edge extracting processing to scene image to be measured, obtains phase
The contour images to be measured answered, and position the nameplate contour area in contour images to be measured.
In the present embodiment, it is necessary to be positioned to the nameplate region in scene image to be measured, known with strengthening follow-up nameplate
Other accuracy rate, excludes influence of the non-nameplate region to identification accuracy.Using edge extraction algorithm, can effectively obtain corresponding
The corresponding contour images to be measured of contour images to be measured, and navigate to nameplate contour area.
Specifically, above-mentioned steps S13 can be carried out in the following way：
Using default HED network models, edge extracting processing is carried out to scene image to be measured, acquisition treats measuring wheel accordingly
Wide image, and position the nameplate contour area in contour images to be measured.
In the present embodiment, HED network models can realize training of the image to image, input an image, export this
The edge detection graph of a image, using scene image to be measured as input picture, through processing, can obtain including nameplate profile region
The contour images to be measured in domain.
Step S14, cuts nameplate contour area to be measured and correction process, obtains nameplate region image to be measured.
In the present embodiment, nameplate contour area to be measured is cut and correction process, excludes non-nameplate region and bat
The interference of angle is taken the photograph, the accuracy that nameplate identifies can be caused to be substantially improved.
Specifically, above-mentioned steps S14 can be realized in the following way：
Using Hough transformation line detection method, nameplate contour area to be measured is cut, and uses perspective transform, it is right
Nameplate contour area to be measured after cutting is corrected, and obtains nameplate region image to be measured.
In the present embodiment, since nameplate is substantially rectangle, it is possible to by the way of Hough transformation straight-line detection
Cut, find nameplate four edges and it is found intersection, and after the intersection point of nameplate is determined, using perspective transform, by inscription to be measured
Board area image is corrected.
Step S15, using default convolutional neural networks, is identified the character in nameplate region image to be measured.
Specifically, above-mentioned steps S15 can be realized in the following way：
Using LeNet-5 convolutional neural networks, the character in nameplate region image to be measured is identified, LeNet-5
Convolutional neural networks are often trained using substation equipment by the use of nameplate character as training data.
In the present embodiment, the character on nameplate is mostly the regular coding of electric power application system internal institution, it is non-it is daily should
All words and phrase included in, since the character set on nameplate is relatively fixed, to ensure discrimination, will directly collect
Nameplate sample in word be trained as the training data of LeNet-5 convolutional neural networks.
The embodiment of the present invention is by obtaining the scene image to be measured containing substation equipment nameplate to be measured；Using default side
Edge extraction algorithm, edge extracting processing is carried out to scene image to be measured, obtains corresponding contour images to be measured, and position and treat measuring wheel
Nameplate contour area in wide image；Nameplate contour area to be measured is cut and correction process, obtain nameplate area to be measured
Area image；Using default convolutional neural networks, the character in nameplate region image to be measured is identified.So power transformation
The nameplate recognition methods of station equipment, can identify the name plate information of different power equipments in the non-contact case, greatly carry
The efficiency and quality of inspection are risen, and the identification to name plate information has very strong environmental suitability and antijamming capability.This
Outside, automatic identification is carried out to equipment nameplate image information and is automatically extracted, teams and groups personnel tradition machinery formula data can be saved and adopted
The workload of collection, greatly improves the efficiency and quality of inspection.
Embodiment two
An embodiment of the present invention provides a kind of nameplate identification device of substation equipment, and referring to Fig. 2, which can wrap
Include：Acquisition module 100, first processing module 200, Second processing module 300, identification module 400.
Acquisition module 100, for obtaining the scene image to be measured containing substation equipment nameplate to be measured.
In the present embodiment, it is not true due to style of shooting when obtaining the nameplate image of substation equipment to be detected
Qualitative, acquired nameplate image not only includes nameplate to be measured, it is also possible to therefore, first comprising the scene residing for nameplate
Obtain the scene image to be measured containing substation equipment nameplate to be measured.
First processing module 200, for using default Boundary extracting algorithm, edge extracting is carried out to scene image to be measured
Processing, obtains corresponding contour images to be measured, and positions the nameplate contour area in contour images to be measured.
In the present embodiment, it is necessary to be positioned to the nameplate region in scene image to be measured, known with strengthening follow-up nameplate
Other accuracy rate, excludes influence of the non-nameplate region to identification accuracy.Using edge extraction algorithm, can effectively obtain corresponding
The corresponding contour images to be measured of contour images to be measured, and navigate to nameplate contour area.
Specifically, first processing module 200, are additionally operable to use default HED network models, and scene image to be measured is carried out
Edge extracting processing, obtains corresponding contour images to be measured, and positions the nameplate contour area in contour images to be measured.
In the present embodiment, HED network models can realize training of the image to image, input an image, export this
The edge detection graph of a image, using scene image to be measured as input picture, through processing, can obtain including nameplate profile region
The contour images to be measured in domain.
Second processing module 300, for being cut to nameplate contour area to be measured and correction process, obtains inscription to be measured
Board area image.
In the present embodiment, nameplate contour area to be measured is cut and correction process, excludes non-nameplate region and bat
The interference of angle is taken the photograph, the accuracy that nameplate identifies can be caused to be substantially improved.
Specifically, Second processing module 300, are additionally operable to use Hough transformation line detection method, to nameplate profile to be measured
Region is cut, and uses perspective transform, and the nameplate contour area to be measured after cutting is corrected, obtains nameplate to be measured
Area image.
In the present embodiment, since nameplate is substantially rectangle, it is possible to by the way of Hough transformation straight-line detection
Cut, find nameplate four edges and it is found intersection, and after the intersection point of nameplate is determined, using perspective transform, by inscription to be measured
Board area image is corrected.
Identification module 400, for use default convolutional neural networks, to the character in nameplate region image to be measured into
Row identification.
Specifically, identification module 400, are additionally operable to use LeNet-5 convolutional neural networks, to nameplate region image to be measured
In character be identified, LeNet-5 convolutional neural networks using substation equipment often by the use of nameplate character as training data into
Row training.
In the present embodiment, the character on nameplate is mostly the regular coding of electric power application system internal institution, it is non-it is daily should
All words and phrase included in, since the character set on nameplate is relatively fixed, to ensure discrimination, will directly collect
Nameplate sample in word be trained as the training data of LeNet-5 convolutional neural networks.
Referring to Fig. 2, which further includes：3rd processing module 500.
3rd processing module 500, for using single scale Retinex algorithm for image enhancement, carries out scene image to be measured
Illumination removes pretreatment.
In the present embodiment, nameplate image often occur uneven illumination it is even in addition as caused by flash lamp it is strong reflective etc.
Phenomenon, in order to improve picture quality, can use single scale Retinex algorithm for image enhancement to carry out illumination to image and remove pre- place
Reason.
The embodiment of the present invention is by obtaining the scene image to be measured containing substation equipment nameplate to be measured；Using default side
Edge extraction algorithm, edge extracting processing is carried out to scene image to be measured, obtains corresponding contour images to be measured, and position and treat measuring wheel
Nameplate contour area in wide image；Nameplate contour area to be measured is cut and correction process, obtain nameplate area to be measured
Area image；Using default convolutional neural networks, the character in nameplate region image to be measured is identified.So power transformation
The nameplate recognition methods of station equipment, can identify the name plate information of different power equipments in the non-contact case, greatly carry
The efficiency and quality of inspection are risen, and the identification to name plate information has very strong environmental suitability and antijamming capability.This
Outside, automatic identification is carried out to equipment nameplate image information and is automatically extracted, teams and groups personnel tradition machinery formula data can be saved and adopted
The workload of collection, greatly improves the efficiency and quality of inspection.
The embodiments of the present invention are for illustration only, do not represent the quality of embodiment.
It should be noted that：The nameplate identification device for the substation equipment that above-described embodiment provides is realizing substation equipment
Nameplate identification when, only with the division progress of above-mentioned each function module for example, in practical application, can as needed and incite somebody to action
Above-mentioned function distribution is completed by different function modules, i.e., the internal structure of equipment is divided into different function modules, with complete
Into all or part of function described above.In addition, above-described embodiment provide substation equipment nameplate identification device with
The nameplate recognition methods embodiment of substation equipment belongs to same design, its specific implementation process refers to embodiment of the method, here
Repeat no more.
One of ordinary skill in the art will appreciate that hardware can be passed through by realizing all or part of step of above-described embodiment
To complete, relevant hardware can also be instructed to complete by program, the program can be stored in a kind of computer-readable
In storage medium, storage medium mentioned above can be read-only storage, disk or CD etc..
The foregoing is merely presently preferred embodiments of the present invention, is not intended to limit the invention, it is all the present invention spirit and
Within principle, any modification, equivalent replacement, improvement and so on, should all be included in the protection scope of the present invention.
Claims (10)
- A kind of 1. nameplate recognition methods of substation equipment, it is characterised in that the described method includes：Obtain the scene image to be measured containing substation equipment nameplate to be measured；Using default Boundary extracting algorithm, edge extracting processing is carried out to scene image to be measured, obtains corresponding profile to be measured Image, and position the nameplate contour area in contour images to be measured；Nameplate contour area to be measured is cut and correction process, obtain nameplate region image to be measured；Using default convolutional neural networks, the character in nameplate region image to be measured is identified.
- 2. according to the method described in claim 1, it is characterized in that, described use default Boundary extracting algorithm, to field to be measured Scape image carries out edge extracting processing, obtains corresponding contour images to be measured, including：Using default HED network models, edge extracting processing is carried out to scene image to be measured, obtains corresponding profile diagram to be measured Picture, and position the nameplate contour area in contour images to be measured.
- 3. according to the method described in claim 1, it is characterized in that, described cut and corrected to nameplate contour area to be measured Processing, obtains nameplate region image to be measured, including：Using Hough transformation line detection method, nameplate contour area to be measured is cut, and uses perspective transform, to cutting Nameplate contour area to be measured afterwards is corrected, and obtains nameplate region image to be measured.
- 4. according to the method described in claim 1, it is characterized in that, described use default convolutional neural networks, to be measured Character in nameplate region image is identified, including：Using LeNet-5 convolutional neural networks, the character in nameplate region image to be measured is identified, the LeNet-5 Convolutional neural networks are often trained using substation equipment by the use of nameplate character as training data.
- 5. according to claim 1-4 any one of them methods, it is characterised in that to detection image carry out localization process it Before, the method further includes：Using single scale Retinex algorithm for image enhancement, illumination is carried out to scene image to be measured and removes pretreatment.
- A kind of 6. nameplate identification device of substation equipment, it is characterised in that including：Acquisition module, for obtaining the scene image to be measured containing substation equipment nameplate to be measured；First processing module, for using default Boundary extracting algorithm, carries out edge extracting processing to scene image to be measured, obtains Corresponding contour images to be measured are taken, and position the nameplate contour area in contour images to be measured；Second processing module, for being cut to nameplate contour area to be measured and correction process, obtains nameplate region to be measured Image；Identification module, for using default convolutional neural networks, is identified the character in nameplate region image to be measured.
- 7. device according to claim 6, it is characterised in that the first processing module, is additionally operable to use default HED Network model, edge extracting processing is carried out to scene image to be measured, obtains corresponding contour images to be measured, and position profile to be measured Nameplate contour area in image.
- 8. device according to claim 6, it is characterised in that the Second processing module, is additionally operable to use Hough transformation Line detection method, cuts nameplate contour area to be measured, and uses perspective transform, to the nameplate profile to be measured after cutting Region is corrected, and obtains nameplate region image to be measured.
- 9. device according to claim 6, it is characterised in that the identification module, is additionally operable to using LeNet-5 convolution god Through network, the character in nameplate region image to be measured is identified, the LeNet-5 convolutional neural networks use substation Equipment is often trained by the use of nameplate character as training data.
- 10. according to claim 6-9 any one of them devices, it is characterised in that further include：3rd processing module, for using single scale Retinex algorithm for image enhancement, illumination removal is carried out to scene image to be measured Pretreatment.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201711436199.2A CN108009547A (en) | 2017-12-26 | 2017-12-26 | A kind of nameplate recognition methods of substation equipment and device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201711436199.2A CN108009547A (en) | 2017-12-26 | 2017-12-26 | A kind of nameplate recognition methods of substation equipment and device |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN108009547A true CN108009547A (en) | 2018-05-08 |
Family
ID=62061465
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201711436199.2A Pending CN108009547A (en) | 2017-12-26 | 2017-12-26 | A kind of nameplate recognition methods of substation equipment and device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN108009547A (en) |
Cited By (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109409355A (en) * | 2018-08-13 | 2019-03-01 | 国网陕西省电力公司 | A kind of method and device of novel transformer nameplate identification |
| CN109522834A (en) * | 2018-11-10 | 2019-03-26 | 国网电力科学研究院武汉南瑞有限责任公司 | A kind of nameplate recognition methods of power equipment |
| CN109766891A (en) * | 2018-12-14 | 2019-05-17 | 北京上格云技术有限公司 | Obtain the method and computer readable storage medium of installations and facilities information |
| CN110334647A (en) * | 2019-07-03 | 2019-10-15 | 云南电网有限责任公司信息中心 | A kind of parameter format method based on image recognition |
| CN110647784A (en) * | 2018-06-27 | 2020-01-03 | 中国移动通信集团浙江有限公司 | Equipment asset management method and device based on deep learning |
| CN110956171A (en) * | 2019-11-06 | 2020-04-03 | 广州供电局有限公司 | Automatic nameplate identification method and device, computer equipment and storage medium |
| CN112257547A (en) * | 2020-10-19 | 2021-01-22 | 国网浙江杭州市萧山区供电有限公司 | Transformer substation safety measure identification method based on deep learning |
| CN112668567A (en) * | 2020-12-25 | 2021-04-16 | 深圳太极云软技术有限公司 | Image clipping algorithm based on deep learning |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101697196A (en) * | 2009-10-29 | 2010-04-21 | 上海索广电子有限公司 | Digital identification system and method for serial numbers of name plate of camera |
| CN103473531A (en) * | 2013-09-04 | 2013-12-25 | 上海索广电子有限公司 | Digit image recognition and error correction method based on name board digit recognition |
| CN104268541A (en) * | 2014-09-15 | 2015-01-07 | 青岛高校信息产业有限公司 | Intelligent image identification method of device nameplate and energy efficiency label |
| CN106096602A (en) * | 2016-06-21 | 2016-11-09 | 苏州大学 | A kind of Chinese licence plate recognition method based on convolutional neural networks |
| CN106599890A (en) * | 2015-10-14 | 2017-04-26 | 山东鲁能智能技术有限公司 | Transformer substation patrol robot digital type instrument identification algorithm |
| CN107181319A (en) * | 2017-05-03 | 2017-09-19 | 贵州电网有限责任公司 | A kind of hard pressing plate condition intelligent method for inspecting of transformer station |
| CN107273882A (en) * | 2017-05-03 | 2017-10-20 | 贵州电网有限责任公司 | A kind of non-intrusion type power screen cabinet identity description and intelligent identification Method |
-
2017
- 2017-12-26 CN CN201711436199.2A patent/CN108009547A/en active Pending
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101697196A (en) * | 2009-10-29 | 2010-04-21 | 上海索广电子有限公司 | Digital identification system and method for serial numbers of name plate of camera |
| CN103473531A (en) * | 2013-09-04 | 2013-12-25 | 上海索广电子有限公司 | Digit image recognition and error correction method based on name board digit recognition |
| CN104268541A (en) * | 2014-09-15 | 2015-01-07 | 青岛高校信息产业有限公司 | Intelligent image identification method of device nameplate and energy efficiency label |
| CN106599890A (en) * | 2015-10-14 | 2017-04-26 | 山东鲁能智能技术有限公司 | Transformer substation patrol robot digital type instrument identification algorithm |
| CN106096602A (en) * | 2016-06-21 | 2016-11-09 | 苏州大学 | A kind of Chinese licence plate recognition method based on convolutional neural networks |
| CN107181319A (en) * | 2017-05-03 | 2017-09-19 | 贵州电网有限责任公司 | A kind of hard pressing plate condition intelligent method for inspecting of transformer station |
| CN107273882A (en) * | 2017-05-03 | 2017-10-20 | 贵州电网有限责任公司 | A kind of non-intrusion type power screen cabinet identity description and intelligent identification Method |
Non-Patent Citations (1)
| Title |
|---|
| SAINING XIE ZHUOWEN TU: "Holistically-Nested Edge Detection", 《2015 IEEE INTERNATIONAL CONFERENCE ON COMPUTER VISION》 * |
Cited By (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110647784A (en) * | 2018-06-27 | 2020-01-03 | 中国移动通信集团浙江有限公司 | Equipment asset management method and device based on deep learning |
| CN109409355A (en) * | 2018-08-13 | 2019-03-01 | 国网陕西省电力公司 | A kind of method and device of novel transformer nameplate identification |
| CN109409355B (en) * | 2018-08-13 | 2021-09-14 | 国网陕西省电力公司 | Novel transformer nameplate identification method and device |
| CN109522834A (en) * | 2018-11-10 | 2019-03-26 | 国网电力科学研究院武汉南瑞有限责任公司 | A kind of nameplate recognition methods of power equipment |
| CN109766891A (en) * | 2018-12-14 | 2019-05-17 | 北京上格云技术有限公司 | Obtain the method and computer readable storage medium of installations and facilities information |
| CN109766891B (en) * | 2018-12-14 | 2020-11-10 | 北京上格云技术有限公司 | Method for acquiring equipment facility information and computer readable storage medium |
| CN110334647A (en) * | 2019-07-03 | 2019-10-15 | 云南电网有限责任公司信息中心 | A kind of parameter format method based on image recognition |
| CN110956171A (en) * | 2019-11-06 | 2020-04-03 | 广州供电局有限公司 | Automatic nameplate identification method and device, computer equipment and storage medium |
| CN112257547A (en) * | 2020-10-19 | 2021-01-22 | 国网浙江杭州市萧山区供电有限公司 | Transformer substation safety measure identification method based on deep learning |
| CN112668567A (en) * | 2020-12-25 | 2021-04-16 | 深圳太极云软技术有限公司 | Image clipping algorithm based on deep learning |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108009547A (en) | A kind of nameplate recognition methods of substation equipment and device | |
| WO2021051885A1 (en) | Target labeling method and apparatus | |
| CN108805076B (en) | Method and system for extracting table characters of environmental impact evaluation report | |
| CN109685075A (en) | A kind of power equipment recognition methods based on image, apparatus and system | |
| CN109409355B (en) | Novel transformer nameplate identification method and device | |
| CN105023014A (en) | Method for extracting tower target in unmanned aerial vehicle routing inspection power transmission line image | |
| CN110991448A (en) | Text detection method and device for nameplate image of power equipment | |
| CN110610483B (en) | Crack image acquisition and detection method, computer equipment and readable storage medium | |
| CN110133443B (en) | Power transmission line component detection method, system and device based on parallel vision | |
| CN106599890B (en) | digital instrument recognition algorithm for substation inspection robot | |
| CN109685847B (en) | Training method and device for visual target detection model | |
| CN103177445A (en) | Outdoor tomato identification method based on subsection threshold image segmentation and light spot identification | |
| CN112487848A (en) | Character recognition method and terminal equipment | |
| CN113095441A (en) | Pig herd bundling detection method, device, equipment and readable storage medium | |
| CN113515655A (en) | Fault identification method and device based on image classification | |
| CN113298077A (en) | Transformer substation foreign matter identification and positioning method and device based on deep learning | |
| CN112802027A (en) | Target object analysis method, storage medium and electronic device | |
| CN106650735B (en) | A kind of LED character automatic positioning recognition methods | |
| CN113111728A (en) | Intelligent identification method and system for power production operation risk in transformer substation | |
| CN116152685A (en) | Pedestrian detection method and system based on unmanned aerial vehicle visual field | |
| CN112581495A (en) | Image processing method, device, equipment and storage medium | |
| CN115063739B (en) | Abnormal behavior detection method, device, equipment and computer storage medium | |
| CN114841980A (en) | Insulator defect detection method and system based on line patrol aerial image | |
| CN107330470B (en) | Method and device for identifying picture | |
| CN105469097A (en) | Transformer station feature extraction method based on nerve network |
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 | ||
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20180508 |