CN115019028A - Intelligent identification method and device for reading of polymorphic pointer meter - Google Patents
Intelligent identification method and device for reading of polymorphic pointer meter Download PDFInfo
- Publication number
- CN115019028A CN115019028A CN202210923465.9A CN202210923465A CN115019028A CN 115019028 A CN115019028 A CN 115019028A CN 202210923465 A CN202210923465 A CN 202210923465A CN 115019028 A CN115019028 A CN 115019028A
- Authority
- CN
- China
- Prior art keywords
- pointer
- meter
- dial
- dial plate
- reading
- 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
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
- G06V10/24—Aligning, centring, orientation detection or correction of the image
- G06V10/243—Aligning, centring, orientation detection or correction of the image by compensating for image skew or non-uniform image deformations
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F17/00—Digital computing or data processing equipment or methods, specially adapted for specific functions
- G06F17/10—Complex mathematical operations
-
- 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
-
- 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
- G06V10/25—Determination of region of interest [ROI] or a volume of interest [VOI]
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
- G06V10/77—Processing image or video features in feature spaces; using data integration or data reduction, e.g. principal component analysis [PCA] or independent component analysis [ICA] or self-organising maps [SOM]; Blind source separation
- G06V10/774—Generating sets of training patterns; Bootstrap methods, e.g. bagging or boosting
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
- G06V10/82—Arrangements for image or video recognition or understanding using pattern recognition or machine learning using neural networks
-
- 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
- Y04—INFORMATION OR COMMUNICATION TECHNOLOGIES HAVING AN IMPACT ON OTHER TECHNOLOGY AREAS
- Y04S—SYSTEMS INTEGRATING TECHNOLOGIES RELATED TO POWER NETWORK OPERATION, COMMUNICATION OR INFORMATION TECHNOLOGIES FOR IMPROVING THE ELECTRICAL POWER GENERATION, TRANSMISSION, DISTRIBUTION, MANAGEMENT OR USAGE, i.e. SMART GRIDS
- Y04S20/00—Management or operation of end-user stationary applications or the last stages of power distribution; Controlling, monitoring or operating thereof
- Y04S20/30—Smart metering, e.g. specially adapted for remote reading
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Evolutionary Computation (AREA)
- Multimedia (AREA)
- Software Systems (AREA)
- Computing Systems (AREA)
- Artificial Intelligence (AREA)
- Health & Medical Sciences (AREA)
- Mathematical Physics (AREA)
- Data Mining & Analysis (AREA)
- Databases & Information Systems (AREA)
- General Health & Medical Sciences (AREA)
- General Engineering & Computer Science (AREA)
- Medical Informatics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Mathematical Analysis (AREA)
- Mathematical Optimization (AREA)
- Computational Mathematics (AREA)
- Algebra (AREA)
- Pure & Applied Mathematics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Biomedical Technology (AREA)
- Biophysics (AREA)
- Computational Linguistics (AREA)
- Molecular Biology (AREA)
- Character Discrimination (AREA)
Abstract
The invention discloses a polymorphic pointer meter reading intelligent identification method and device. The invention provides an intelligent identification method for reading of a polymorphic pointer meter, which adopts a single-stage target detection method based on deep learning to accurately position the positions of a dial and a pointer, and maps the identified angle of the pointer meter to a dial scale by mathematical methods such as irregular dial correction and linear transformation; the single-stage target detection method based on deep learning has the advantages that the task process is simplified, the memory occupation and hardware requirements can be reduced, the reading of the pointer meter can be identified under the condition that the dial plate of the pointer meter has no scale characters, and the accuracy of the identified reading of the pointer meter can be improved under the condition that part of the dial plate has serious deformation.
Description
Technical Field
The invention relates to the technical field of instrument recognition, in particular to an intelligent recognition method and device for reading of a polymorphic pointer meter.
Background
The existing pointer meter reading method based on deep learning mainly adopts a method combining target detection meter positioning and traditional image processing pointer identification, firstly, key positions of a dial and a pointer are positioned through target detection methods such as YOLO, Mask RCNN, fast RCNN and the like, then, the determined position of the pointer is identified through traditional image processing such as Hough transform, least square and the like, and finally, pointer scales are obtained through conversion of the identified pointer position and the key position of the dial. The method has the problem of poor robustness. The field pointer table has more styles and is difficult to satisfy the reading identification of the pointer instrument in a complex environment. Once the pointer table style changes, the model algorithm must be adjusted. In addition, the traditional image recognition has high requirements on image quality, the image of the shooting pointer table is influenced by factors such as climate, light, angle and the like in the environment, so that the image quality fluctuates in a large range, the traditional image recognition method mostly relies on the setting of a threshold value to recognize a target, and the ever-changing environmental factors make the existing method difficult to adapt to the image quality recognition task under various conditions.
The method comprises the steps of firstly detecting and positioning key positions of a dial plate and a pointer through a target, then identifying scales of a dial plate through a character recognition method, and finally converting the scales pointed by the pointer. The improved pointer meter reading method abandons the traditional image recognition algorithm and replaces the traditional image recognition algorithm with a deep learning character string recognition method, solves the problem of poor robustness of the traditional image recognition algorithm to a certain extent, can be suitable for more types of pointer meters, and simultaneously solves the problem of image quality change caused by environmental factors well. However, the method adopts a two-stage model algorithm, the task process is still relatively complicated, and the memory occupation and hardware requirements are higher. In addition, the method adopts a character string recognition method, and can be realized only on the basis of the requirement that scale characters exist on a dial plate of the pointer table. In reality, scale characters do not always exist in a plurality of pointer tables, and only the mark bits of the maximum value and the minimum value exist. In this way, it is difficult to recognize the scale characters and convert them to the final scale. In addition, due to the limitation of shooting conditions, a part of the dial plate can be seriously deformed, and the pointer table scales obtained by the method have relatively large errors.
Disclosure of Invention
The invention aims to provide an intelligent identification method and device for reading of a polymorphic pointer meter, and aims to solve the problems that the task process of the existing pointer meter reading method is complicated, the memory occupation and hardware requirements are high, the identification is difficult under the condition that a dial plate of the pointer meter has no scale characters, part of the dial plate can be seriously deformed, and the scale of the identified pointer meter has large errors.
In a first aspect, the invention provides an intelligent identification method for reading of a polymorphic pointer meter, which comprises the following steps:
acquiring an image of a polymorphic pointer table;
calling a trained target detection model to detect a pointer dial plate and a pointer in the image;
correcting irregular pointer tables in the identified pointer table tables;
recording clockwise and anticlockwise information of pointer readings in the dial plate of each pointer meter and dial plate scale information; aiming at a dial without scales, setting the minimum mark position as 0, setting the maximum mark position as 1, and converting the minimum mark position and the minimum mark position into percentage values according to the proportion;
determining a linear relation between the angle of the pointer and the dial scale information according to clockwise and anticlockwise information of the pointer reading and the dial scale information;
identifying a real-time angle of the pointer;
substituting the real-time angle of the pointer into the linear relational expression to convert real-time dial scale information indicated by the pointer on a dial of the pointer meter, and outputting the real-time dial scale information as a reading result of the pointer meter;
and marking the reading result of the pointer table in the image.
Further, the training mode of the target detection model is as follows:
collecting images of a polymorphic pointer table as a sample set for training;
manually labeling a dial plate and a pointer of the pointer meter in the sample set, wherein the pointer is positioned on the diagonal of a labeling frame of the pointer;
dividing the sample set into a training set and a verification set according to the ratio of 9:1, wherein the training set is sent into a yoloV4 network model for training, and the verification set is used for precision calculation in the training process;
and testing the training effect of the model and optimizing the model.
Further, correct the irregular pointer table dial plate in the pointer table dial plate that discerns, include:
establishing a mathematical relation between the irregular shape of the dial plate of the irregular pointer meter and the standard circle;
and mapping the irregular dial plate of the pointer meter into a standard circular dial plate of the pointer meter according to the mathematical relationship, so that scales on the dial plate of the pointer meter are uniformly distributed.
Further, according to the clockwise and counterclockwise information of the pointer reading and the dial scale information, determining a linear relation between the angle of the pointer and the dial scale information, comprising:
determining the corresponding relation between the maximum scale of the dial plate of the pointer meter and the maximum angle of the pointer and the corresponding relation between the minimum scale of the dial plate of the pointer meter and the minimum angle of the pointer;
and determining the mathematical linear relation between the pointer and the scales of the dial plate of the pointer meter according to the corresponding relation.
In a second aspect, the invention provides an intelligent identification device for reading of a polymorphic pointer meter, comprising:
the acquisition unit is used for acquiring an image of the polymorphic pointer table;
the detection unit is used for calling the trained target detection model and detecting the dial plate and the pointer of the pointer table in the image;
the correction unit is used for correcting the irregular pointer meter dial in the identified pointer meter dial;
the input unit is used for inputting clockwise and anticlockwise information of pointer readings in the dial plates of the pointer meters and dial plate scale information; aiming at a dial without scales, setting the minimum mark position as 0, setting the maximum mark position as 1, and converting the minimum mark position and the minimum mark position into percentage values according to the proportion;
the determining unit is used for determining a linear relation between the angle of the pointer and dial scale information;
the identification unit is used for identifying the real-time angle of the pointer;
the conversion unit is used for substituting the real-time angle of the pointer into the linear relational expression, converting real-time dial scale information indicated by the pointer on a dial plate of the pointer meter, and outputting the real-time dial scale information as a reading result of the pointer meter;
and the marking unit is used for marking the reading result of the pointer meter in the image.
Further, the training mode of the target detection model is as follows:
collecting images of a polymorphic pointer table as a sample set for training;
manually labeling a dial plate and a pointer of the pointer meter in the sample set, wherein the pointer is positioned on the diagonal of a labeling frame of the pointer;
dividing the sample set into a training set and a verification set according to the ratio of 9:1, wherein the training set is sent into a yoloV4 network model for training, and the verification set is used for precision calculation in the training process;
and testing the training effect of the model and optimizing the model.
Further, the correction unit includes:
the establishing unit is used for establishing a mathematical relation between the irregular shape of the dial plate of the irregular pointer meter and the standard circle;
and the mapping unit is used for mapping the irregular pointer meter dial plate into a standard circular pointer meter dial plate according to the mathematical relationship, so that scales on the pointer meter dial plate are uniformly distributed.
Further, the determining unit is used for determining the corresponding relation between the maximum scale of the dial plate of the pointer meter and the maximum angle of the pointer, and between the minimum scale of the dial plate of the pointer meter and the minimum angle of the pointer; and determining the mathematical linear relation between the pointer and the scales of the dial plate of the pointer meter according to the corresponding relation.
The invention has the following beneficial effects: the invention provides a polymorphic pointer meter reading intelligent identification method and device.A single-stage target detection method based on deep learning is adopted to accurately position the positions of a dial plate and a pointer, and the identified pointer meter angle is mapped to a dial scale by mathematical methods such as irregular dial plate correction and linear transformation; the single-stage target detection method based on deep learning has the advantages that the task process is simplified, the memory occupation and hardware requirements can be reduced, the reading of the pointer meter can be identified under the condition that the dial plate of the pointer meter has no scale characters, and the accuracy of the identified reading of the pointer meter can be improved under the condition that part of the dial plate has serious deformation.
Drawings
In order to more clearly illustrate the technical solution of the present invention, the drawings needed to be used in the embodiments will be briefly described below, and it is obvious to those skilled in the art that other drawings can be obtained according to the drawings without any inventive exercise.
FIG. 1 is a flowchart of an intelligent identification method for reading of a polymorphic pointer meter according to an embodiment of the present invention;
fig. 2 is a schematic view of an intelligent reading identification device for a polymorphic pointer meter according to an embodiment of the present invention.
Detailed Description
In order to make the objects, technical solutions and advantages of the present invention more apparent, the technical solutions of the present invention will be clearly and completely described below with reference to the specific embodiments of the present invention and the accompanying drawings. It is to be understood that the described embodiments are merely exemplary of the invention, and not restrictive of the full scope of the invention. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention. The technical solutions provided by the embodiments of the present invention are described in detail below with reference to the accompanying drawings.
Referring to fig. 1, an embodiment of the present invention provides an intelligent identification method for reading of a polymorphic pointer table, including:
s101, acquiring an image of the polymorphic pointer table.
Taking the pointer tables of a wind power plant and a coal power plant as an example, images of the multi-form pointer tables in the wind power plant and the coal power plant can be collected, the multi-form pointer tables can comprise pointer tables in various shapes such as a circle, a square and the like, the pointer tables can be provided with different numbers of pointers, and the most common pointer tables are a single pointer, a double pointer and a triple pointer.
And S102, calling the trained target detection model, and detecting the dial plate and the pointer of the pointer table in the image.
In this embodiment, the goal detection model may adopt yoloV4 network model, and the goal detection model is trained as follows: collecting images of a polymorphic pointer table as a sample set for training; manually labeling a dial plate and a pointer of the pointer meter in the sample set, wherein the pointer is positioned on the diagonal of a labeling frame of the pointer; dividing the sample set into a training set and a verification set according to the ratio of 9:1, wherein the training set is sent into a yoloV4 network model for training, and the verification set is used for precision calculation in the training process; and testing the training effect of the model and optimizing the model.
S103, correcting irregular pointer tables in the identified pointer table tables.
Due to the limitation of shooting conditions, a plurality of circular dials can be shot into irregular shapes during shooting, scales on the dials with irregular shapes are in non-uniform distribution, and the scales of the pointer table cannot be directly output by adopting a linear transformation method. The dial plate correction is correction processing performed on an irregular-shaped dial plate, so that the dial plate is finally mapped to a standard circular dial plate, scales of the dial plate are uniformly distributed, and errors caused by a non-circular dial plate are eliminated.
In this embodiment, correct the irregular pointer table dial in the pointer table dial that discerns, can specifically include: establishing a mathematical relation between the irregular shape of the dial plate of the irregular pointer meter and the standard circle; and mapping the irregular dial plate of the pointer meter into a standard circular dial plate of the pointer meter according to the mathematical relationship, so that scales on the dial plate of the pointer meter are uniformly distributed.
S104, recording clockwise and anticlockwise information of pointer readings in the dial plate of each pointer meter and dial plate scale information; and aiming at the dial without scales, setting the minimum mark position as 0, setting the maximum mark position as 1, and converting the minimum mark position and the minimum mark position into percentage values according to the proportion.
Aiming at a dial plate without scales, only the maximum value and the minimum value are marked on the dial plate, and the reading problem of the pointer meter cannot be well solved by the existing method. The invention provides a method for replacing scale values by percentages, wherein the minimum value is set to be 0, the maximum value is set to be 1, and the middle value is converted into percentage values according to the proportion, such as 0.5, 0.8 and the like.
And S105, determining a linear relation between the angle of the pointer and the dial scale information according to the clockwise and anticlockwise information of the pointer reading and the dial scale information.
In this embodiment, determining a linear relation between the angle of the pointer and the dial scale information according to the clockwise and counterclockwise information of the pointer reading and the dial scale information may specifically include: determining the corresponding relation between the maximum scale of the dial plate of the pointer meter and the maximum angle of the pointer and the corresponding relation between the minimum scale of the dial plate of the pointer meter and the minimum angle of the pointer; and determining the mathematical linear relation between the pointer and the scales of the dial plate of the pointer meter according to the corresponding relation.
S106, identifying the real-time angle of the pointer.
And S107, substituting the real-time angle of the pointer into the linear relational expression to convert real-time dial scale information indicated by the pointer on the dial of the pointer meter, and outputting the real-time dial scale information as a reading result of the pointer meter.
Linear transformation is a method for solving the mapping of pointer identification angle and real scale relation. According to the invention, the identification angle of the pointer can be obtained by determining the linear corresponding relation between the dial scale and the pointer angle and combining the accurate target identification of target detection, so that the pointer scale is converted and output.
And S108, marking the reading result of the pointer table in the image.
The invention provides a pointer table identification method based on deep learning, which is compatible with pointer tables in various forms on the basis of identification results, enhances the application range of the reading of the pointer table, and improves the accuracy of reading calculation.
The method adopts a single-stage target recognition algorithm based on deep learning to accurately position the dial plate and the pointer, and then maps the recognized pointer angle to the dial scale through mathematical methods such as irregular dial plate correction and linear transformation.
Referring to fig. 2, the present invention provides an intelligent reading identification device for a polymorphic pointer meter, comprising:
an acquisition unit 21 configured to acquire an image of a polymorphic pointer table;
the detection unit 22 is used for calling the trained target detection model and detecting the dial plate and the pointer of the pointer table in the image;
a correction unit 23 for correcting an irregular dial plate of the identified dial plate;
the recording unit 24 is used for recording clockwise and anticlockwise information of pointer readings in the dial plates of the pointer meters and dial plate scale information; aiming at a dial without scales, setting the minimum mark position as 0, setting the maximum mark position as 1, and converting the minimum mark position and the minimum mark position into percentage values according to the proportion;
a determining unit 25 for determining a linear relation between the angle of the pointer and the dial scale information;
an identification unit 26 for identifying a real-time angle of the pointer;
the conversion unit 27 is used for substituting the real-time angle of the pointer into the linear relational expression to convert the real-time dial scale information indicated by the pointer on the dial plate of the pointer meter, and outputting the real-time dial scale information as a reading result of the pointer meter;
and the marking unit 28 is used for marking the reading result of the pointer table in the image.
Further, the training mode of the target detection model is as follows: collecting images of a polymorphic pointer table as a sample set for training; manually labeling a dial plate and a pointer of the pointer meter in the sample set, wherein the pointer is positioned on the diagonal of a labeling frame of the pointer; dividing the sample set into a training set and a verification set according to the ratio of 9:1, wherein the training set is sent into a yoloV4 network model for training, and the verification set is used for precision calculation in the training process; and testing the training effect of the model and optimizing the model.
Further, the correction unit includes:
the establishing unit is used for establishing a mathematical relation between the irregular shape of the dial plate of the irregular pointer meter and the standard circle;
and the mapping unit is used for mapping the irregular pointer meter dial plate into a standard circular pointer meter dial plate according to the mathematical relationship, so that scales on the pointer meter dial plate are uniformly distributed.
Further, the determining unit is used for determining the corresponding relation between the maximum scale of the dial plate of the pointer meter and the maximum angle of the pointer, and between the minimum scale of the dial plate of the pointer meter and the minimum angle of the pointer; and determining the mathematical linear relation between the pointer and the scales of the dial plate of the pointer meter according to the corresponding relation.
The embodiment of the invention also provides a storage medium, and the storage medium stores a computer program, and when the computer program is executed by a processor, the computer program realizes part or all of the steps in each embodiment of the polymorphic pointer meter reading intelligent identification method provided by the invention. The storage medium may be a magnetic disk, an optical disk, a Read-only memory (ROM) or a Random Access Memory (RAM).
Those skilled in the art will readily appreciate that the techniques of the embodiments of the present invention may be implemented as software plus a required general purpose hardware platform. Based on such understanding, the technical solutions in the embodiments of the present invention may be essentially or partially implemented in the form of a software product, which may be stored in a storage medium, such as ROM/RAM, magnetic disk, optical disk, etc., and includes several instructions for enabling a computer device (which may be a personal computer, a server, or a network device, etc.) to execute the method according to the embodiments or some parts of the embodiments.
The same and similar parts in the various embodiments in this specification may be referred to each other. Especially, for the embodiment of the intelligent identification device for reading of the polymorphic pointer table, the description is simple because the embodiment is basically similar to the embodiment of the method, and the relevant points can be referred to the description in the embodiment of the method.
The above-described embodiments of the present invention should not be construed as limiting the scope of the present invention.
Claims (8)
1. An intelligent identification method for reading of a polymorphic pointer meter is characterized by comprising the following steps:
acquiring an image of a polymorphic pointer table;
calling a trained target detection model to detect a pointer dial plate and a pointer in the image;
correcting irregular pointer tables in the identified pointer table tables;
recording clockwise and anticlockwise information of pointer readings in the dial plate of each pointer meter and dial plate scale information; aiming at a dial without scales, setting the minimum mark position as 0, setting the maximum mark position as 1, and converting the minimum mark position and the minimum mark position into percentage values according to the proportion;
determining a linear relation between the angle of the pointer and the dial scale information according to clockwise and anticlockwise information of the pointer reading and the dial scale information;
identifying a real-time angle of the pointer;
substituting the real-time angle of the pointer into the linear relational expression to convert real-time dial scale information indicated by the pointer on a dial of the pointer meter, and outputting the real-time dial scale information as a reading result of the pointer meter;
and marking the reading result of the pointer table in the image.
2. The intelligent identification method for reading of polymorphic pointer tables according to claim 1, characterized in that the training mode of the target detection model is as follows:
collecting images of a polymorphic pointer table as a sample set for training;
manually labeling a dial plate and a pointer of the pointer meter in the sample set, wherein the pointer is positioned on the diagonal of a labeling frame of the pointer;
dividing the sample set into a training set and a verification set according to the ratio of 9:1, wherein the training set is sent into a yoloV4 network model for training, and the verification set is used for precision calculation in the training process;
and testing the training effect of the model and optimizing the model.
3. The intelligent identification method for reading of the polymorphic pointer meter as claimed in claim 1, wherein the correction of the irregular pointer meter dial in the identified pointer meter dial comprises:
establishing a mathematical relation between the irregular shape of the dial plate of the irregular pointer meter and the standard circle;
and mapping the irregular pointer meter dial plate into a standard circular pointer meter dial plate according to the mathematical relationship, so that scales on the pointer meter dial plate are uniformly distributed.
4. The intelligent identification method of the reading of the polymorphic pointer meter as claimed in claim 1, wherein the step of determining the linear relation between the angle of the pointer and the dial scale information according to the clockwise and counterclockwise information of the reading of the pointer and the dial scale information comprises the following steps:
determining the corresponding relation between the maximum scale of the dial plate of the pointer meter and the maximum angle of the pointer and the corresponding relation between the minimum scale of the dial plate of the pointer meter and the minimum angle of the pointer;
and determining the mathematical linear relation between the pointer and the scales of the dial plate of the pointer meter according to the corresponding relation.
5. The utility model provides a polymorphic pointer table reading intelligent recognition device which characterized in that includes:
the acquisition unit is used for acquiring an image of the polymorphic pointer table;
the detection unit is used for calling the trained target detection model and detecting the dial plate and the pointer of the pointer table in the image;
the correction unit is used for correcting irregular pointer tables in the identified pointer table tables;
the input unit is used for inputting clockwise and anticlockwise information of pointer readings in the dial plates of the pointer meters and dial plate scale information; aiming at a dial without scales, setting the minimum mark position as 0, setting the maximum mark position as 1, and converting the minimum mark position and the minimum mark position into percentage values according to the proportion;
the determining unit is used for determining a linear relation between the angle of the pointer and dial scale information;
the identification unit is used for identifying the real-time angle of the pointer;
the conversion unit is used for substituting the real-time angle of the pointer into the linear relational expression, converting real-time dial scale information indicated by the pointer on a dial plate of the pointer meter, and outputting the real-time dial scale information as a reading result of the pointer meter;
and the marking unit is used for marking the reading result of the pointer meter in the image.
6. The intelligent recognition device for reading of polymorphic pointer tables according to claim 5, wherein the training mode of the target detection model is as follows:
collecting images of a polymorphic pointer table as a sample set for training;
manually labeling a dial plate and a pointer of the pointer meter in the sample set, wherein the pointer is positioned on the diagonal of a labeling frame of the pointer;
dividing the sample set into a training set and a verification set according to the ratio of 9:1, wherein the training set is sent into a yoloV4 network model for training, and the verification set is used for precision calculation in the training process;
and testing the training effect of the model and optimizing the model.
7. The intelligent identification device for reading of polymorphic pointer meters according to claim 5, wherein the correction unit comprises:
the establishing unit is used for establishing a mathematical relation between the irregular shape of the dial plate of the irregular pointer meter and the standard circle;
and the mapping unit is used for mapping the irregular pointer meter dial plate into a standard circular pointer meter dial plate according to the mathematical relationship, so that scales on the pointer meter dial plate are uniformly distributed.
8. The intelligent identification device of polymorphic pointer meter readings of claim 5, wherein the determining unit is used for determining the corresponding relationship between the maximum scale of the dial plate of the pointer meter and the maximum angle of the pointer, and between the minimum scale of the dial plate of the pointer meter and the minimum angle of the pointer; and determining the mathematical linear relation between the pointer and the scales of the dial plate of the pointer meter according to the corresponding relation.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210923465.9A CN115019028A (en) | 2022-08-02 | 2022-08-02 | Intelligent identification method and device for reading of polymorphic pointer meter |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210923465.9A CN115019028A (en) | 2022-08-02 | 2022-08-02 | Intelligent identification method and device for reading of polymorphic pointer meter |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN115019028A true CN115019028A (en) | 2022-09-06 |
Family
ID=83066077
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210923465.9A Pending CN115019028A (en) | 2022-08-02 | 2022-08-02 | Intelligent identification method and device for reading of polymorphic pointer meter |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115019028A (en) |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109993154A (en) * | 2019-04-19 | 2019-07-09 | 山东大学 | The lithium sulfur type instrument intelligent identification Method of substation's simple pointer formula |
| CN111931776A (en) * | 2020-10-16 | 2020-11-13 | 江西小马机器人有限公司 | Deep learning-based double-pointer instrument reading method |
| CN111950330A (en) * | 2019-05-16 | 2020-11-17 | 杭州测质成科技有限公司 | Pointer instrument indicating number detection method based on target detection |
| CN112508098A (en) * | 2020-12-08 | 2021-03-16 | 南京理工大学 | Dial plate positioning and automatic reading pointer type meter value identification method and system |
| CN112949564A (en) * | 2021-02-02 | 2021-06-11 | 电子科技大学 | Pointer type instrument automatic reading method based on deep learning |
| CN114359573A (en) * | 2021-12-06 | 2022-04-15 | 新奥新智科技有限公司 | Pointer meter reading identification method and device and storage medium |
| CN114663744A (en) * | 2022-05-24 | 2022-06-24 | 西安石油大学 | Instrument automatic identification method and system based on machine learning |
-
2022
- 2022-08-02 CN CN202210923465.9A patent/CN115019028A/en active Pending
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109993154A (en) * | 2019-04-19 | 2019-07-09 | 山东大学 | The lithium sulfur type instrument intelligent identification Method of substation's simple pointer formula |
| CN111950330A (en) * | 2019-05-16 | 2020-11-17 | 杭州测质成科技有限公司 | Pointer instrument indicating number detection method based on target detection |
| CN111931776A (en) * | 2020-10-16 | 2020-11-13 | 江西小马机器人有限公司 | Deep learning-based double-pointer instrument reading method |
| CN112508098A (en) * | 2020-12-08 | 2021-03-16 | 南京理工大学 | Dial plate positioning and automatic reading pointer type meter value identification method and system |
| CN112949564A (en) * | 2021-02-02 | 2021-06-11 | 电子科技大学 | Pointer type instrument automatic reading method based on deep learning |
| CN114359573A (en) * | 2021-12-06 | 2022-04-15 | 新奥新智科技有限公司 | Pointer meter reading identification method and device and storage medium |
| CN114663744A (en) * | 2022-05-24 | 2022-06-24 | 西安石油大学 | Instrument automatic identification method and system based on machine learning |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111950330B (en) | Pointer instrument indication detection method based on target detection | |
| CN112257676B (en) | Pointer type instrument reading method and system and inspection robot | |
| CN108182433B (en) | Meter reading identification method and system | |
| CN114168906B (en) | Mapping geographic information data acquisition system based on cloud computing | |
| CN103759758A (en) | Method for detecting position of automobile meter needle based on mechanical angle and scale identification | |
| CN109508709B (en) | Single pointer instrument reading method based on machine vision | |
| CN111738229B (en) | Automatic reading method for scale of pointer dial | |
| CN109840278A (en) | Histogram data switching control method, device, computer equipment and storage medium | |
| CN110287887A (en) | A kind of scale recognition methods, device and computer storage medium | |
| CN111353502B (en) | Digital table identification method and device and electronic equipment | |
| CN114399677A (en) | Pointer instrument identification method based on text region reading | |
| CN115496893A (en) | Pointer type water meter reading identification method | |
| CN103218820A (en) | Camera calibration error compensation method based on multi-dimensional characteristics | |
| CN115019028A (en) | Intelligent identification method and device for reading of polymorphic pointer meter | |
| CN111652244A (en) | Pointer type meter identification method based on unsupervised feature extraction and matching | |
| CN116612265A (en) | Target encoding and decoding-based dial pointer reading identification method | |
| CN112464986B (en) | Reading method and device for pointer type disc instrument | |
| CN115439864A (en) | Water meter reading identification method and system, computer equipment and storage medium | |
| CN113076941A (en) | Single pointer dial reading identification method based on video frame fusion | |
| CN112560983B (en) | Instrument identification method and device | |
| CN112528566A (en) | Real-time air quality data calibration method and system based on AdaBoost training model | |
| CN110705924A (en) | Anemometry tower anemometry data processing method and device based on wind direction sector | |
| CN115019136B (en) | Training method and detection method of target key point detection model for resisting boundary point drift | |
| CN116302624B (en) | Engineering project mapping data intelligent management system | |
| CN113435527B (en) | Taxi meter verification method and system based on machine vision |
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: 20220906 |