CN112163583A - Method for recognizing digital meter reading, recognition device and computer readable storage medium - Google Patents
Method for recognizing digital meter reading, recognition device and computer readable storage medium Download PDFInfo
- Publication number
- CN112163583A CN112163583A CN202011022129.4A CN202011022129A CN112163583A CN 112163583 A CN112163583 A CN 112163583A CN 202011022129 A CN202011022129 A CN 202011022129A CN 112163583 A CN112163583 A CN 112163583A
- Authority
- CN
- China
- Prior art keywords
- digital
- image
- recognition
- reading
- character code
- 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
- 238000000034 method Methods 0.000 title claims abstract description 60
- 238000003860 storage Methods 0.000 title claims abstract description 16
- 238000012937 correction Methods 0.000 claims abstract description 18
- 238000004590 computer program Methods 0.000 claims description 22
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 claims description 16
- 238000012549 training Methods 0.000 claims description 9
- 238000013528 artificial neural network Methods 0.000 claims description 3
- 238000003062 neural network model Methods 0.000 claims description 3
- 238000013135 deep learning Methods 0.000 description 10
- 238000005516 engineering process Methods 0.000 description 9
- 238000010586 diagram Methods 0.000 description 7
- 230000006870 function Effects 0.000 description 5
- 230000008569 process Effects 0.000 description 3
- 238000012795 verification Methods 0.000 description 3
- 238000005299 abrasion Methods 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 2
- 238000009434 installation Methods 0.000 description 2
- 238000002372 labelling Methods 0.000 description 2
- 238000013519 translation Methods 0.000 description 2
- 238000009825 accumulation Methods 0.000 description 1
- 230000001413 cellular effect Effects 0.000 description 1
- 238000005520 cutting process Methods 0.000 description 1
- 230000003247 decreasing effect Effects 0.000 description 1
- 238000000151 deposition Methods 0.000 description 1
- 230000005611 electricity Effects 0.000 description 1
- 230000007246 mechanism Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 238000012545 processing Methods 0.000 description 1
- 239000007787 solid Substances 0.000 description 1
- 230000007704 transition Effects 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/20—Image preprocessing
- G06V10/26—Segmentation of patterns in the image field; Cutting or merging of image elements to establish the pattern region, e.g. clustering-based techniques; Detection of occlusion
- G06V10/267—Segmentation of patterns in the image field; Cutting or merging of image elements to establish the pattern region, e.g. clustering-based techniques; Detection of occlusion by performing operations on regions, e.g. growing, shrinking or watersheds
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/045—Combinations of networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
-
- 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
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V2201/00—Indexing scheme relating to image or video recognition or understanding
- G06V2201/02—Recognising information on displays, dials, clocks
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- General Health & Medical Sciences (AREA)
- Molecular Biology (AREA)
- Biophysics (AREA)
- Computational Linguistics (AREA)
- Data Mining & Analysis (AREA)
- Evolutionary Computation (AREA)
- Artificial Intelligence (AREA)
- Biomedical Technology (AREA)
- Computing Systems (AREA)
- General Engineering & Computer Science (AREA)
- Life Sciences & Earth Sciences (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Health & Medical Sciences (AREA)
- Multimedia (AREA)
- Character Discrimination (AREA)
Abstract
The invention provides a method for identifying the reading of a digital watch, an identification device and a computer readable storage medium, wherein the identification method comprises the steps of obtaining a character code picture containing a digital watch disc; the first recognition engine stores a first image recognition model, and the first image recognition model recognizes a character code picture and outputs a first recognition result; the second recognition engine stores a second image recognition model, and the second image recognition model recognizes the character code picture and outputs a second recognition result; and judging whether the first recognition result is consistent with the second recognition result, if so, outputting the reading of the digital dial, and if not, inputting the manual correction value into the first recognition engine and the second recognition engine. The method for identifying the digital meter reading can greatly improve the accuracy rate of identifying the digital meter picture.
Description
Technical Field
The invention relates to the technical field of image recognition, in particular to a method for recognizing numerical meter reading, a device for recognizing numerical meter reading and a computer readable storage medium.
Background
At present, the similar water meter identification technology only focuses on digital deconstruction of shot images, but the comparison standards are usually a single set of digital standard structure. Although the identification technology can finally reach a certain identification accuracy rate through a long learning process and image accumulation and put the identification technology into use, the application of the technology is obviously limited by the diversity of image differences expressed by water meters of different models, brands, sizes and service lives. Because the same kind of water meter identification technology can not deal with the diversity problem of water meters by using a single identification method, identification blind areas are easy to appear, and therefore errors of reading with certain probability are caused. More importantly, the recognition result cannot be verified, and the recognition result can be manually corrected only after being discovered by the user and reported, so that the manual examination time is increased, and the normal water consumption of the user is influenced for a long time by depositing the undiscovered recognition error result. In the water meter reading application scene, the data volume to be read is very huge, and other water meter reading identification methods lacking a parallel error correction mechanism need manual verification on each reading identification value to judge that the data is wrong, so that the workload is very huge.
In addition, the existing identification method identifies all water meters with different models and different wear degrees by using a set of fixed digital image structure, and the difference of different water meters is large, so that a long learning process exists and the identification is difficult to achieve high accuracy.
Disclosure of Invention
The invention aims to provide a digital meter reading identification method which can greatly improve the identification accuracy of a digital meter picture.
A second object of the present invention is to provide an identification device of digital meter reading which can greatly improve the accuracy of digital meter picture identification.
A third object of the present invention is to provide a computer-readable storage medium capable of greatly improving the accuracy of digital table picture recognition.
In order to achieve the first object, the invention provides a method for identifying the reading of a digital watch, which comprises the steps of obtaining a character code picture containing a digital watch plate; the first recognition engine stores a first image recognition model, and the first image recognition model recognizes the character code picture and outputs a first recognition result; the second recognition engine stores a second image recognition model, and the second image recognition model recognizes the character code picture and outputs a second recognition result; and judging whether the first recognition result is consistent with the second recognition result, if so, outputting the reading of the digital dial, and if not, inputting the manual correction value into the first recognition engine and the second recognition engine.
According to the scheme, the first recognition engine and the second recognition engine which are used for building two different picture recognition methods are used for recognizing at the same time, the output results of the two engines are compared, and deep learning error correction is performed once the two recognition results are inconsistent. The method for recognizing the reading of the digital meter adopts the double recognition engines to perform double-channel parallel recognition, thereby reducing respective recognition blind areas and improving the recognition accuracy.
In addition, a unique result is output after the results are compared to be consistent, and the readings which are identified and compared to be inconsistent are automatically counted and are brought into the links of manual correction and deep learning. Identification errors are found through automatic comparison results, and the link of deep learning and manual error correction is entered, so that the data discovery rate of the identification errors deposited in the past technology is greatly improved, and the huge workload of manual checking is saved. The method can greatly improve the identification accuracy, cannot output data with identification errors, saves the workload of manual error checking, solves the problem that identification reading needs manual verification, and can automatically refresh and select the error data.
Meanwhile, the recognition method has strong compatibility with different types of digital meters, the existing digital meter recognition technology is mostly applied to the installed traditional digital meters, the digital meters can be provided with different models and sizes by different manufacturers, the installation time of the digital meters is different, and different abrasion degrees exist. For the non-floating difference of the digital table, the invention greatly reduces the identification blind area of a single method by adopting two different image identification methods, thereby being beneficial to identification.
The method for identifying the reading of the digital meter preferably further comprises the steps of constructing a first identification engine; the step of building a first recognition engine comprises: collecting character code pictures endowed with identity codes, wherein the identity codes are numbers corresponding to the character code pictures; carrying out gridding positioning on the character code picture endowed with the identity code; classifying and modeling the character code picture endowed with the identity code by adopting a deep neural network to form a first image recognition model; and taking the character code picture endowed with the identity code as a training sample to carry out deep neural network model training.
Therefore, the first recognition engine receives and stores massive character code pictures of the digital table in the file service, configures corresponding reading of each character code picture as an identity code to be placed in the big data resource pool, and automatically recognizes the input character code pictures of the digital table through comparison with data in the big data resource pool.
The further scheme is that the first recognition engine and the second recognition engine output numbers with large occupied areas according to the occupied areas of the numbers.
Still further, the step of building the first recognition engine further comprises: acquiring an original image containing a digital dial plate; inputting the manual marking value of the original image into a first recognition engine, and performing rotation, stretching and translation operations on the original image; inputting the operated original image and outputting a standard digital table image; and matching the identities of the standard digital table image and the character code image of the first image recognition model, outputting the identity code corresponding to the character code image by the first recognition engine if the matching is successful, and inputting the manual correction value into the first recognition engine if the matching is failed.
Therefore, the resource pool is initialized by manually labeling the original image, the error rate is very high by automatic identification at the beginning, and then the learning and the correction are slowly performed, so that the learning time is very long. The accuracy of the data in the initial resource pool can be basically ensured by manually marking, so that the subsequent error rate basically comes from the problem of diversity of image shooting, and the identification efficiency can be improved.
The further scheme is that the digital dial plate in the original image occupies more than two thirds of the area of the original image.
The method for identifying the reading of the digital meter preferably further comprises the steps of constructing a second identification engine; the step of building a second recognition engine comprises: acquiring an image containing a digital dial plate; splitting a character code picture of an image into 0-9 ten pictures with single character codes; and constructing a second image recognition model of the corresponding digit of the single character code picture.
Preferably, the digital meter comprises a digital gas meter, a digital electric meter or a digital water meter.
In a preferred embodiment, in the step of acquiring the image containing the digital table, a photograph containing the digital table is taken by an image capturing device, or a video containing the digital table is taken, and the photograph containing the digital table is intercepted.
In order to achieve the second object, the present invention provides an apparatus for recognizing digital meter readings, comprising a processor and a memory, wherein the memory stores a computer program, and the computer program realizes the steps of the method for recognizing digital meter readings according to any item above when being executed by the processor.
To achieve the third object, the present invention provides a computer-readable storage medium, on which a computer program is stored, the computer program, when being executed by a controller, implementing the steps of the method for recognizing the reading of the digital table.
Drawings
FIG. 1 is a flow chart of an embodiment of a method of identifying a digital meter reading of the present invention.
FIG. 2 is a schematic representation of a digital meter image in an embodiment of the method of identifying a digital meter reading of the present invention.
FIG. 3 is a schematic diagram of a character code picture and its corresponding identity code in an embodiment of the method for recognizing the reading of the digital meter according to the present invention.
FIG. 4 is a schematic diagram of a grid-based positioning of a character code picture in an embodiment of the identification method of the digital meter reading of the present invention.
FIG. 5 is a schematic diagram of invoking a picture resource in a big data resource pool in an embodiment of the identification method of the digital meter reading of the present invention.
FIG. 6 is a schematic diagram of ten pictures of a single word in an embodiment of the method for identifying a reading of a digital meter of the present invention.
FIG. 7 is a schematic diagram of feature point labeling in an embodiment of the method for identifying readings of a digital meter according to the invention.
FIG. 8 is a schematic diagram of a character code image in an embodiment of the method for recognizing a digital meter reading according to the present invention.
FIG. 9 is a schematic illustration of feature matching in an embodiment of the method of identification of a digital meter reading of the present invention.
FIG. 10 is a schematic diagram of a manual assignment interface in the error correction and deep learning steps in the method for recognizing the reading of the digital meter according to the embodiment of the present invention.
The invention is further explained with reference to the drawings and the embodiments.
Detailed Description
The invention relates to a digital meter reading identification method, which is a computer program applied to digital meter reading identification equipment and used for realizing identification of digital meter reading. The invention also provides a device for recognizing the reading of the digital meter, which comprises a controller, wherein the controller is used for realizing the steps of the method for recognizing the reading of the digital meter when executing the computer program stored in the memory. The present invention also provides a computer readable storage medium having stored thereon a computer program which, when being executed by a controller, carries out the steps of the above-mentioned method of recognizing a reading of a digital meter.
The digital meter is a meter which displays the reading number on the dial plate through the number. Such as water meters, electricity meters, gas meters, etc. The digital meter in this embodiment is a water meter.
Embodiment of method for recognizing digital meter reading
Referring to fig. 1, a method of identifying a digital meter reading includes the following steps.
And S1, building a first recognition engine, wherein the first recognition engine is used as a main recognition engine.
A picture containing the digital dial, or a video containing the digital dial, is taken by the image capture device and the picture containing the digital dial is taken, see fig. 2, in which the digital dial is used to display the digital reading. A total of about 10 million photographs were taken.
Referring to fig. 3, a water meter picture is cut into character code pictures, each character code picture is endowed with an identity code, the identity code is a number corresponding to the character code picture, and the identity code corresponding to each character code picture is manually input.
Referring to fig. 4, the character code picture given the identity code is positioned in a gridding manner, and for the case of transition numbers, numbers with large occupied area are output according to the occupied area of the numbers.
And carrying out classified modeling on the character code pictures endowed with the identity codes by adopting a Deep Neural Network (DNN) to form a first image recognition model, and integrating all model data into a big data resource pool. And taking the character code picture endowed with the identity code as a training sample to carry out deep neural network model training.
And S14, initializing the resource pool.
And acquiring an original image containing a digital dial, wherein the digital dial in the original image occupies more than two thirds of the area of the original image.
The artificial annotation value of the original image is input into a first recognition engine.
And performing rotation, stretching and translation operations on the original image.
Inputting the operated original image and outputting a standard digital table image.
And matching the standard digital table image with the character code picture in the big data resource pool in an identity mode, referring to fig. 5, calling the character code picture in the big data resource pool through an identity code if the matching is successful, outputting a corresponding digital table reading, and manually inputting the reading corresponding to the standard digital table image if the matching is failed.
And S2, building a second recognition engine step.
An image containing a digital dial is acquired.
Referring to fig. 6, the water meter character code picture is split into 0 to 9 pictures of single character codes. The ten pictures of the single character code are obtained by collecting digital tables of different models.
And constructing a second image recognition model of the corresponding digit of the single character code picture. Referring to fig. 7, 0 to 9 pictures of a single character are feature labeled. Based on an SIFT algorithm, namely Scale-invariant feature transform (SIFT), firstly, a Scale space is constructed, extreme points are detected, and Scale invariance is obtained. Then, the feature points are filtered and pinpointed. Then, a direction value is assigned to the feature point. And finally, generating a feature descriptor, endowing an identification value to the picture corresponding to each feature point, and taking the matching of the feature points as a unique basis for picture identification.
And S3, identifying a comparison step.
An image containing a digital dial is acquired.
And cutting the water meter picture into a character code picture containing the digital dial plate.
The first recognition engine recognizes the character code picture, the first recognition engine takes the character code picture as input, the character code picture is recognized through a first image recognition model obtained through pre-training, and a first recognition result is output.
The second recognition engine recognizes the character code picture, the second recognition engine takes the character code picture as input, referring to fig. 8, and utilizes the grid positioning tool to automatically cut the picture, and cuts the picture into a single character code picture. Referring to fig. 9, the character code picture is recognized by the second image recognition model, and a second recognition result is output.
And judging whether the first recognition result is consistent with the second recognition result, if so, outputting the recognition result of the reading of the digital meter, and if not, not outputting the reading, and respectively entering a deep learning error correction step by the first recognition engine and the second recognition engine.
And S4, a deep learning error correction step.
Referring to fig. 10, after the character code pictures with inconsistent recognition results are manually corrected, the corrected values are classified into the respective big data resource pools of the first recognition engine and the second recognition engine, and the first recognition engine and the second recognition engine respectively perform deep learning and then output recognition readings. And the manual correction is to manually input the reading corresponding to the character code picture into the first recognition engine and the second recognition engine.
And updating data of the big data resource pool based on the primary recognition value and the artificial error correction value, performing intensive training of resource accuracy and expansion of resource quantity, and repeatedly performing training to enable recognition to be more accurate and efficient.
Therefore, the first recognition engine and the second recognition engine which are built for two different picture recognition methods are used for simultaneously recognizing, the output results of the two engines are compared, and once the two recognition results are inconsistent, deep learning error correction is carried out. The method for recognizing the reading of the digital meter adopts the double recognition engines to perform double-channel parallel recognition, thereby reducing respective recognition blind areas and improving the recognition accuracy. In addition, a unique result is output after the results are compared to be consistent, and the readings which are identified and compared to be inconsistent are automatically counted and are brought into the links of manual correction and deep learning. Identification errors are found through automatic comparison results, and the link of deep learning and manual error correction is entered, so that the data discovery rate of the identification errors deposited in the past technology is greatly improved, and the huge workload of manual checking is saved. The method can greatly improve the identification accuracy, cannot output data with identification errors, saves the workload of manual error checking, solves the problem that identification reading needs manual verification, and can automatically refresh and select the error data. Meanwhile, the recognition method has strong compatibility with different types of digital meters, the existing digital meter recognition technology is mostly applied to the installed traditional digital meters, the digital meters can be provided with different models and sizes by different manufacturers, the installation time of the digital meters is different, and different abrasion degrees exist. For the non-floating difference of the digital table, the invention greatly reduces the identification blind area of a single method by adopting two different image identification methods, thereby being beneficial to identification.
Identification device embodiment of digital meter reading:
the identification device for the reading of the digital meter of the embodiment comprises a controller, and the controller executes a computer program to realize the steps of the identification method embodiment for the reading of the digital meter.
For example, a computer program may be partitioned into one or more modules, which are stored in a memory and executed by a controller to implement the present invention. One or more of the modules may be a series of computer program instruction segments capable of performing specific functions, the instruction segments being used to describe the execution of the computer program in the identification device of the digital meter reading.
The identification device of the digital meter reading may include, but is not limited to, a controller, a memory. It will be appreciated by those skilled in the art that the identification device of the digital meter reading may comprise more or fewer components, or some components in combination, or different components, e.g. the identification device of the digital meter reading may also comprise an input output device, a network access device, a bus, etc.
For example, the controller may be a Central Processing Unit (CPU), other general purpose controller, a Digital Signal Processor (DSP), an Application Specific Integrated Circuit (ASIC), an off-the-shelf programmable Gate Array (FPGA) or other programmable logic device, discrete Gate or transistor logic, discrete hardware components, and so on. The general controller may be a microcontroller or the controller may be any conventional controller or the like. The controller is a control center of the identification equipment of the digital meter reading, and various interfaces and lines are used for connecting all parts of the identification equipment of the digital meter reading.
The memory may be used to store computer programs and/or modules, and the controller may implement the various functions of the identification device of digital meter readings by running or executing the computer programs and/or modules stored in the memory, as well as by invoking data stored in the memory. For example, the memory may mainly include a storage program area and a storage data area, wherein the storage program area may store an operating system, an application program required for at least one function (e.g., a sound receiving function, a sound-to-text function, etc.), and the like; the storage data area may store data (e.g., audio data, text data, etc.) created according to the use of the cellular phone, etc. In addition, the memory may include high speed random access memory, and may also include non-volatile memory, such as a hard disk, a memory, a plug-in hard disk, a Smart Media Card (SMC), a Secure Digital (SD) Card, a Flash memory Card (Flash Card), at least one magnetic disk storage device, a Flash memory device, or other volatile solid state storage device.
Computer-readable storage medium embodiments:
the computer program stored in the memory of the identification device of the reading of the digital meter can be stored in a computer-readable storage medium if it is implemented in the form of a software functional unit and sold or used as a stand-alone product. Based on such understanding, all or part of the processes in the above embodiments of the method for identifying a digital meter reading may also be implemented by a computer program, which may be stored in a computer readable storage medium and may implement the steps in the above embodiments of the method for identifying a digital meter reading when the computer program is executed by a controller. Wherein the computer program comprises computer program code, which may be in the form of source code, object code, an executable file or some intermediate form, etc. The storage medium may include: any entity or device capable of carrying computer program code, recording medium, U.S. disk, removable hard disk, magnetic disk, optical disk, computer Memory, Read-Only Memory (ROM), Random Access Memory (RAM), electrical carrier wave signals, telecommunications signals, software distribution media, and the like. It should be noted that the computer readable medium may contain other components which may be suitably increased or decreased as required by legislation and patent practice in jurisdictions, for example, in some jurisdictions, in accordance with legislation and patent practice, the computer readable medium does not include electrical carrier signals and telecommunications signals.
Finally, it should be emphasized that the above-described preferred embodiments of the present invention are merely examples of implementations, not limitations, and various changes and modifications may be made by those skilled in the art, without departing from the spirit and scope of the invention, and any changes, equivalents, improvements, etc. made within the spirit and scope of the present invention are intended to be embraced therein.
Claims (10)
1. A method of identifying a digital meter reading, comprising:
acquiring a character code picture containing a digital dial plate;
the first recognition engine stores a first image recognition model, and the first image recognition model recognizes the character code picture and outputs a first recognition result;
the second recognition engine stores a second image recognition model, and the second image recognition model recognizes the character code picture and outputs a second recognition result;
and judging whether the first recognition result is consistent with the second recognition result, if so, outputting the reading of the digital dial, and if not, inputting the manual correction value into the first recognition engine and the second recognition engine.
2. The method of identifying digital meter readings according to claim 1, wherein:
the identification method of the digital meter reading further comprises the step of building a first identification engine;
the step of building a first recognition engine comprises:
collecting character code pictures endowed with identity codes, wherein the identity codes are numbers corresponding to the character code pictures;
carrying out gridding positioning on the character code picture endowed with the identity code;
classifying and modeling the character code picture endowed with the identity code by adopting a deep neural network to form the first image recognition model;
and taking the character code picture endowed with the identity code as a training sample to carry out deep neural network model training.
3. A method of identifying a reading from a digital meter as claimed in claim 2, wherein:
and the first recognition engine and the second recognition engine output numbers with large occupied areas according to the occupied areas of the numbers.
4. A method of identifying a reading from a digital meter as claimed in claim 2, wherein:
the step of building a first recognition engine further comprises:
acquiring an original image containing a digital dial plate;
inputting the artificial annotation value of the original image into the first recognition engine;
rotating, stretching and translating the original image;
inputting the operated original image and outputting a standard digital table image;
and matching the identities of the normalized digital table image and the character code image of the first image recognition model, outputting the identity code corresponding to the character code image by the first recognition engine if the matching is successful, and inputting a manual correction value into the first recognition engine if the matching is failed.
5. The method of identifying digital meter readings according to claim 4, wherein:
and the digital dial plate in the original image accounts for more than two thirds of the area of the original image.
6. Method for the identification of the reading of a digital meter according to any one of claims 1 to 5, characterized in that:
the identification method of the digital meter reading also comprises the step of building a second identification engine;
the step of building a second recognition engine comprises:
acquiring an image containing a digital dial plate;
splitting a character code picture of an image into 0-9 ten pictures with single character codes;
and constructing a second image recognition model of the corresponding digit of the single character code picture.
7. Method for the identification of the reading of a digital meter according to any one of claims 1 to 5, characterized in that:
the digital meter comprises a digital gas meter, a digital electric meter or a digital water meter.
8. Method for the identification of the reading of a digital meter according to any one of claims 1 to 5, characterized in that:
in the step of acquiring the image containing the digital watch, a picture containing the digital watch is taken by an image acquisition device, or
A video containing a digital table is taken and a photograph containing the digital table is captured.
9. An apparatus for identification of digital meter readings, comprising a processor and a memory, said memory storing a computer program, characterized in that said computer program, when executed by said processor, carries out the steps of the method for identification of digital meter readings according to any one of claims 1 to 8.
10. A computer-readable storage medium, on which a computer program is stored, which, when being executed by a controller, carries out the steps of the method for identifying a reading of a digital meter according to any one of claims 1 to 8.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011022129.4A CN112163583A (en) | 2020-09-25 | 2020-09-25 | Method for recognizing digital meter reading, recognition device and computer readable storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011022129.4A CN112163583A (en) | 2020-09-25 | 2020-09-25 | Method for recognizing digital meter reading, recognition device and computer readable storage medium |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN112163583A true CN112163583A (en) | 2021-01-01 |
Family
ID=73862964
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202011022129.4A Pending CN112163583A (en) | 2020-09-25 | 2020-09-25 | Method for recognizing digital meter reading, recognition device and computer readable storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112163583A (en) |
Citations (13)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JPH09251509A (en) * | 1996-03-15 | 1997-09-22 | Sharp Corp | Device for recognizing driver's license |
| US20060115156A1 (en) * | 2004-11-26 | 2006-06-01 | Fuji Photo Film Co. Ltd | Image recognizing system, an image recognizing method, a machine readable medium storing thereon, and a computer program for recognizing images |
| CN103380442A (en) * | 2011-02-17 | 2013-10-30 | 日本电气株式会社 | Information processing device |
| CN104115194A (en) * | 2012-03-22 | 2014-10-22 | 光荣株式会社 | System for reading paper sheet serial number, and method for reading paper sheet serial number |
| CN106169080A (en) * | 2016-07-04 | 2016-11-30 | 重庆大学 | A kind of combustion gas index automatic identifying method based on image |
| CN107729954A (en) * | 2017-10-27 | 2018-02-23 | 朱秋华 | A kind of character recognition method, device, Text region equipment and storage medium |
| CN107862312A (en) * | 2017-11-22 | 2018-03-30 | 朱秋华 | A kind of character recognition method based on tesseract engines, device, equipment and storage medium |
| CN108229480A (en) * | 2017-12-25 | 2018-06-29 | 新智数字科技有限公司 | A kind of recognition methods, device and the equipment of number meter reading |
| US20180211669A1 (en) * | 2017-01-26 | 2018-07-26 | Essence, Inc | Speech Recognition Based on Context and Multiple Recognition Engines |
| CN108388894A (en) * | 2017-12-26 | 2018-08-10 | 新智数字科技有限公司 | A kind of recognition methods, device and the equipment of number meter reading |
| CN109635799A (en) * | 2018-12-17 | 2019-04-16 | 石家庄科林电气股份有限公司 | A kind of recognition methods of gas meter, flow meter wheel digital |
| CN109858453A (en) * | 2019-02-15 | 2019-06-07 | 浪潮通用软件有限公司 | A kind of general multi engine bank slip recognition system and method |
| CN111353502A (en) * | 2020-02-25 | 2020-06-30 | 北京眸视科技有限公司 | Digital table identification method and device and electronic equipment |
-
2020
- 2020-09-25 CN CN202011022129.4A patent/CN112163583A/en active Pending
Patent Citations (13)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JPH09251509A (en) * | 1996-03-15 | 1997-09-22 | Sharp Corp | Device for recognizing driver's license |
| US20060115156A1 (en) * | 2004-11-26 | 2006-06-01 | Fuji Photo Film Co. Ltd | Image recognizing system, an image recognizing method, a machine readable medium storing thereon, and a computer program for recognizing images |
| CN103380442A (en) * | 2011-02-17 | 2013-10-30 | 日本电气株式会社 | Information processing device |
| CN104115194A (en) * | 2012-03-22 | 2014-10-22 | 光荣株式会社 | System for reading paper sheet serial number, and method for reading paper sheet serial number |
| CN106169080A (en) * | 2016-07-04 | 2016-11-30 | 重庆大学 | A kind of combustion gas index automatic identifying method based on image |
| US20180211669A1 (en) * | 2017-01-26 | 2018-07-26 | Essence, Inc | Speech Recognition Based on Context and Multiple Recognition Engines |
| CN107729954A (en) * | 2017-10-27 | 2018-02-23 | 朱秋华 | A kind of character recognition method, device, Text region equipment and storage medium |
| CN107862312A (en) * | 2017-11-22 | 2018-03-30 | 朱秋华 | A kind of character recognition method based on tesseract engines, device, equipment and storage medium |
| CN108229480A (en) * | 2017-12-25 | 2018-06-29 | 新智数字科技有限公司 | A kind of recognition methods, device and the equipment of number meter reading |
| CN108388894A (en) * | 2017-12-26 | 2018-08-10 | 新智数字科技有限公司 | A kind of recognition methods, device and the equipment of number meter reading |
| CN109635799A (en) * | 2018-12-17 | 2019-04-16 | 石家庄科林电气股份有限公司 | A kind of recognition methods of gas meter, flow meter wheel digital |
| CN109858453A (en) * | 2019-02-15 | 2019-06-07 | 浪潮通用软件有限公司 | A kind of general multi engine bank slip recognition system and method |
| CN111353502A (en) * | 2020-02-25 | 2020-06-30 | 北京眸视科技有限公司 | Digital table identification method and device and electronic equipment |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108399405B (en) | Business license identification method and device | |
| CN110210513B (en) | Data classification method and device and terminal equipment | |
| CN112819686B (en) | Image style processing method and device based on artificial intelligence and electronic equipment | |
| WO2017088537A1 (en) | Component classification method and apparatus | |
| CN105426390A (en) | Image recognition-based question search method and system | |
| CN110705651A (en) | Method, device, equipment and medium for testing video identification accuracy | |
| CN113254654B (en) | Model training method, text recognition method, device, equipment and medium | |
| US20220222799A1 (en) | Method for detecting defect in products and electronic device using method | |
| CN104750791A (en) | Image retrieval method and device | |
| CN112528030A (en) | Semi-supervised learning method and system for text classification | |
| CN112560845A (en) | Character recognition method and device, intelligent meal taking cabinet, electronic equipment and storage medium | |
| CN111882034A (en) | Neural network processing and face recognition method, device, equipment and storage medium | |
| CN112527676A (en) | Model automation test method, device and storage medium | |
| CN117173172A (en) | Machine vision-based silica gel molding effect detection method and system | |
| CN114639152A (en) | Multi-modal voice interaction method, device, equipment and medium based on face recognition | |
| CN107071553B (en) | Method, device and computer readable storage medium for modifying video and voice | |
| CN113486715A (en) | Image reproduction identification method, intelligent terminal and computer storage medium | |
| CN112163583A (en) | Method for recognizing digital meter reading, recognition device and computer readable storage medium | |
| CN114693554B (en) | Big data image processing method and system | |
| CN115797291A (en) | Circuit terminal identification method and device, computer equipment and storage medium | |
| CN115578796A (en) | Training method, device, equipment and medium for living body detection model | |
| CN114913350A (en) | Material duplicate checking method, device, equipment and storage medium | |
| CN114821396A (en) | Normative detection method, device and storage medium for LNG unloading operation process | |
| CN115393838A (en) | Pointer instrument reading identification method and device, electronic equipment and storage medium | |
| CN111274369A (en) | English word recognition method and device |
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 |