CN111783727B - Automatic meter reading method and system based on machine vision and edge computing technology - Google Patents
Automatic meter reading method and system based on machine vision and edge computing technology Download PDFInfo
- Publication number
- CN111783727B CN111783727B CN202010677856.8A CN202010677856A CN111783727B CN 111783727 B CN111783727 B CN 111783727B CN 202010677856 A CN202010677856 A CN 202010677856A CN 111783727 B CN111783727 B CN 111783727B
- Authority
- CN
- China
- Prior art keywords
- edge computing
- processing module
- meter reading
- machine vision
- module
- 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.)
- Active
Links
- 238000000034 method Methods 0.000 title claims abstract description 42
- 238000005516 engineering process Methods 0.000 title claims abstract description 38
- 238000012545 processing Methods 0.000 claims abstract description 65
- 238000004891 communication Methods 0.000 claims abstract description 26
- 238000001914 filtration Methods 0.000 claims abstract description 13
- 238000004364 calculation method Methods 0.000 claims abstract description 11
- 238000003062 neural network model Methods 0.000 claims description 8
- 230000008569 process Effects 0.000 claims description 8
- 230000007246 mechanism Effects 0.000 claims description 6
- 230000001502 supplementing effect Effects 0.000 claims description 6
- 230000009467 reduction Effects 0.000 claims description 4
- 238000012795 verification Methods 0.000 claims description 3
- 238000010276 construction Methods 0.000 abstract description 8
- 238000004519 manufacturing process Methods 0.000 abstract description 7
- 230000009545 invasion Effects 0.000 abstract description 3
- 230000009286 beneficial effect Effects 0.000 abstract description 2
- 238000010586 diagram Methods 0.000 description 6
- 230000006872 improvement Effects 0.000 description 6
- 230000005540 biological transmission Effects 0.000 description 5
- 230000006870 function Effects 0.000 description 5
- 238000012423 maintenance Methods 0.000 description 5
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 description 5
- 238000004458 analytical method Methods 0.000 description 3
- 230000008859 change Effects 0.000 description 3
- QVFWZNCVPCJQOP-UHFFFAOYSA-N chloralodol Chemical compound CC(O)(C)CC(C)OC(O)C(Cl)(Cl)Cl QVFWZNCVPCJQOP-UHFFFAOYSA-N 0.000 description 3
- 238000005286 illumination Methods 0.000 description 3
- 238000013528 artificial neural network Methods 0.000 description 2
- 230000001413 cellular effect Effects 0.000 description 2
- 238000011161 development Methods 0.000 description 2
- 210000001503 joint Anatomy 0.000 description 2
- 230000033001 locomotion Effects 0.000 description 2
- 230000006855 networking Effects 0.000 description 2
- 238000007781 pre-processing Methods 0.000 description 2
- 230000002159 abnormal effect Effects 0.000 description 1
- 238000013473 artificial intelligence Methods 0.000 description 1
- 230000003139 buffering effect Effects 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 230000005611 electricity Effects 0.000 description 1
- -1 electricity Substances 0.000 description 1
- 238000005265 energy consumption Methods 0.000 description 1
- 238000000605 extraction Methods 0.000 description 1
- 238000009776 industrial production Methods 0.000 description 1
- 238000007689 inspection Methods 0.000 description 1
- 230000010354 integration Effects 0.000 description 1
- 230000007774 longterm Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000010606 normalization Methods 0.000 description 1
- 238000003672 processing method Methods 0.000 description 1
- 239000000047 product Substances 0.000 description 1
- 230000002035 prolonged effect Effects 0.000 description 1
- 238000012797 qualification Methods 0.000 description 1
- 238000003860 storage Methods 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 239000013589 supplement Substances 0.000 description 1
- 238000012549 training Methods 0.000 description 1
- 238000012546 transfer Methods 0.000 description 1
- 230000001960 triggered effect Effects 0.000 description 1
- 230000000007 visual effect Effects 0.000 description 1
- 239000002699 waste material Substances 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
-
- 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
- G06K—GRAPHICAL DATA READING; PRESENTATION OF DATA; RECORD CARRIERS; HANDLING RECORD CARRIERS
- G06K17/00—Methods or arrangements for effecting co-operative working between equipments covered by two or more of main groups G06K1/00 - G06K15/00, e.g. automatic card files incorporating conveying and reading operations
- G06K17/0022—Methods or arrangements for effecting co-operative working between equipments covered by two or more of main groups G06K1/00 - G06K15/00, e.g. automatic card files incorporating conveying and reading operations arrangements or provisious for transferring data to distant stations, e.g. from a sensing device
-
- 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
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/94—Hardware or software architectures specially adapted for image or video understanding
- G06V10/95—Hardware or software architectures specially adapted for image or video understanding structured as a network, e.g. client-server architectures
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N7/00—Television systems
- H04N7/18—Closed-circuit television [CCTV] systems, i.e. systems in which the video signal is not broadcast
-
- 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)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Multimedia (AREA)
- General Engineering & Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Evolutionary Computation (AREA)
- Software Systems (AREA)
- Life Sciences & Earth Sciences (AREA)
- Artificial Intelligence (AREA)
- Biophysics (AREA)
- Computing Systems (AREA)
- Computational Linguistics (AREA)
- Health & Medical Sciences (AREA)
- Signal Processing (AREA)
- General Health & Medical Sciences (AREA)
- Molecular Biology (AREA)
- Biomedical Technology (AREA)
- Mathematical Physics (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Evolutionary Biology (AREA)
- Arrangements For Transmission Of Measured Signals (AREA)
- Studio Devices (AREA)
Abstract
The invention provides an automatic meter reading method based on machine vision and edge computing technology, which comprises the following steps: s1, starting edge computing equipment; s2, shooting a pointer instrument through a camera module; s3, judging a scene through the image processing module, and judging whether an image or a video stream returned by the camera module contains a pointer instrument or not; s4, positioning the instrument through the image processing module; s5, carrying out numerical value inference through the image processing module and outputting an inferred result to a data processing module on the edge computing equipment; s6, carrying out mean value filtering processing on the inferred result of the image processing module through the data processing module to obtain a meter reading numerical value; and S7, transmitting the filtered meter reading numerical value to a remote server through a communication module. The beneficial effects of the invention are as follows: the construction difficulty is low, and the production is not affected; the device is free of damage and invasion; edge calculation, server pressure is small.
Description
Technical Field
The invention relates to a meter reading method, in particular to an automatic meter reading method and system based on machine vision and edge computing technology.
Background
Today with developed communication technology, many situations requiring manual meter reading, such as residential water, electricity, gas, etc., have been changed into remote meter reading. The main technical scheme of remote meter reading is represented by a narrowband internet of things technology such as NB-IoT, loRa and the like, and the application of a physical network technology greatly reduces the degree of manual participation, so that all things are interconnected to truly fall to the ground. However, the technology needs to integrally replace acquisition equipment such as an ammeter, a water meter, a gas meter and the like, and after replacement, a metering module and a remote communication module (2G/3G) of a meter are matched to realize meter reading and data synchronization, and the details are as follows:
and a metering module: the meter core is generally arranged at the meter core position of the meter, the usage amount is calculated through the rotation angle of the meter core and the formula, and the output value is converted into binary digits through an internal chip and then is transmitted to the remote communication module through the internal memory.
And a remote communication module: the communication module is generally arranged at the top of the meter, a 2G/3G SIM flow card is generally used, key data such as equipment ID, consumption, battery power and the like in the internal memory of the equipment are transmitted to the remote gateway of the Internet of things through the SIM card, and the gateway is forwarded to a corresponding analysis module for analysis and finally enters a database.
In the aspect of power supply of the Internet of things equipment, a 500 milliamp battery is conventionally used, when the electric quantity is exhausted, the battery needs to be replaced, and when the battery is not detachable or has higher integration level, the whole equipment needs to be replaced. Of course, there are also devices using long-term power supply such as DC5V, 12V, depending on the specific use environment.
Therefore, in the implementation process of the Internet of things project, operations such as power failure, water cut-off, gas cut-off and the like are often involved in view of safety, multiparty negotiation is needed, construction time is gradually determined, and normal life of residents is seriously affected. Meanwhile, professional technicians who still need to hold relevant licenses in the construction process carry out, cause whole thing networking to reform transform engineering time long, and the manual work is put into high, and when thing networking equipment breaks down (if equipment normal use, data can not report) need carry out whole change, whole fairly waste time and energy.
In the places where large-scale equipment such as factories, subways and high-speed rails are in cloud collection, many professional equipment does not have replaceable internet of things equipment, and also does not have a data reporting function (such as a barometer of a fire control gas bottle), if large-area internet of things construction is carried out, the operation must be stopped, and a large amount of economic loss is caused. The data items which are required to be focused on simultaneously in industry are up to thousands, however, the number of the replaced internet of things equipment is very few, so that manual meter reading and recording are still used in many places, abnormal conditions cannot be linked with a professional system in time, the scheme has defects, and good control opportunities are missed.
Aiming at the situation, zhejiang Dahua proposes to use an image recognition technology to recognize a pressure pointer, the principle is to design a cup-shaped device which is light-tight, a camera, an LED spotlight and a remote communication module are placed at the bottom, then the whole device is reversely buckled on an instrument, light supplementing and shooting are carried out through the buckled interior, the shot image is conveyed to a remote server for recognition, and data uploading is carried out through the remote server. The whole scheme is economical and practical, the original equipment can be not modified for data reading, but the terminal does not have data processing capability, and the processing pressure of the server is large in large-scale deployment.
The above prior art has the following drawbacks:
1. the implementation cost is high: the adoption of the mode of the narrow-band internet of things for automatic meter reading needs a great deal of time and effort on the coordination of construction time, because the construction involves water cut, power cut and production cut, and huge losses are caused to enterprises, especially production enterprises;
2. the maintenance cost is high: the narrowband internet of things equipment is highly integrated, if any fine module is damaged, the problems that data cannot be uploaded and the like occur, equipment replacement is needed to be carried out by operations such as water cut-off, power failure and the like, and the maintenance cost is gradually increased;
3. poor replaceability: the number of data items which are required to be focused on simultaneously in industry is as high as thousands, however, the number of the replaced Internet of things equipment is quite small, firstly, the related products of the narrowband Internet of things equipment in the scene are not much, secondly, the related working conditions of a plurality of modules in the narrowband Internet of things are not consistent with use, and thirdly, the related qualification can not reach the standard;
4. the personnel overhaul is inconvenient: although the device in the bloom solves the problems to a great extent, when the device is overhauled, workers need to manually read meters and troubleshoot the meters, and the device in the bloom adopts a direct coverage mode, so that the workers cannot directly operate the meters and meters during overhauling, inconvenience is brought to related works, and meanwhile, the risk of damaging the devices is increased;
5. the remote calculation pressure is large: because of the characteristics of the large-scale equipment, the terminal is only responsible for image acquisition and transmission, and the calculation task is carried out at the server, when a plurality of pieces of equipment need to be analyzed at the same time, the server has huge pressure, and downtime is easy to cause threat to other software running on the machine;
6. the information security is poor: because the public network (the operator network) is adopted to transfer data, the use of the bloom equipment is difficult in the scene of needing data confidentiality, the mode of the narrow-band internet of things can be carried out by using the LoRa, the LoRa is paved by adopting the private gateway, the data is transferred from the private network, the safety can be effectively ensured, and the problems of implementation, maintenance, replacement and the like can be solved by using the mode.
Disclosure of Invention
In order to solve the problems in the prior art, the invention provides an automatic meter reading method and system based on machine vision and edge computing technology.
The invention provides an automatic meter reading method based on machine vision and edge computing technology, which comprises the following steps:
s1, starting edge computing equipment, wherein the edge computing equipment is terminal equipment and comprises a camera module, an image processing module, a data processing module and a communication module, the edge computing equipment is used as a computing edge to be registered, and after permission is acquired, the camera module on the edge computing equipment is started;
s2, shooting a pointer instrument through a camera module, and outputting a shot image or video stream to an image processing module on the edge computing equipment;
s3, judging a scene through the image processing module, judging whether an image or a video stream transmitted back by the camera module contains a pointer instrument, if not, returning to the step S2, and if so, entering the next step;
s4, positioning the instrument through the image processing module;
s5, carrying out numerical value inference through the image processing module and outputting an inferred result to a data processing module on the edge computing equipment;
s6, carrying out mean value filtering processing on the inferred result of the image processing module through the data processing module to obtain a meter reading numerical value;
and S7, transmitting the filtered meter reading numerical value to a remote server through a communication module.
As a further development of the invention, in step S3, it is retrieved whether the image or video stream contains a pointer instrument or not, according to the pre-trained neural network model.
As a further refinement of the present invention, the pre-trained neural network model is a LeNet neural network model.
As a further improvement of the present invention, in step S4, the retrieved pointer instrument coordinates are preprocessed, including scaling, noise reduction, gray scale, binarization, to obtain a pointer instrument image;
as a further improvement of the invention, in step S5, the center coordinates of the instrument in the pointer instrument image are determined by the machine vision technology, four image boundaries are drawn by taking the coordinates as (0, 0), then the pointer coordinates are determined, the position of the image boundary is judged, the instrument value, the initial position, the minimum value, the maximum value and the step length are determined, and finally the inferred value is calculated according to the trigonometric function formula.
As a further improvement of the invention, in step S6, the average value filtering processing is performed through the data filter, the data filter is responsible for filtering the generated numerical values, when the numerical values exceed the maximum value and are smaller than the minimum value, and the sudden increase and decrease trigger fusing, namely, operation is performed again, recursion processing is continuously performed, finally meter reading numerical values conforming to the filtering are output, and the data entering the data filter are locally cached for self-learning and parameter adjustment of the data filter.
As a further improvement of the invention, in step S7, the calculation result is subjected to group verification through the communication module, and then the meter reading numerical value is pushed to a remote server by using a communication interface on the edge computing equipment, so that data archiving is completed.
The invention provides an automatic meter reading system based on machine vision and edge computing technology, which is used for realizing the method according to any one of the above.
As a further improvement of the invention, the system comprises a terminal device and a remote server, wherein the terminal device is an edge computing device, the edge computing device comprises a camera module, an image processing module, a data processing module and a communication module, the camera module shoots a pointer instrument, the output end of the camera module is connected with the image processing module, the output end of the image processing module is connected with the data processing module, and the data processing module is connected with the remote server through the communication module.
As a further improvement of the invention, the camera module comprises a camera, a horizontal moving mechanism, a vertical moving mechanism and an LED light supplementing lamp.
The beneficial effects of the invention are as follows:
1. the construction difficulty is small, and the production is not influenced: the pointer type instrument is in butt joint in a visual mode, the original equipment is not required to be replaced, the construction and maintenance difficulties are low, and operations such as production stopping and industry stopping are not required to be performed;
2. no damage, no invasion: when important industrial equipment is in butt joint, the important industrial equipment is not required to be destroyed, and the connected industrial equipment is not perceived;
3. edge calculation, server pressure is small: the terminal equipment completes the key image extraction, instrument positioning and numerical value estimation and completes all calculation processes at the terminal, thereby greatly relieving the information transmission and server data processing pressure.
Drawings
FIG. 1 is a data flow diagram of an automated meter reading method based on machine vision and edge computing technology according to the present invention.
Fig. 2 is a schematic diagram of a camera module of an automated meter reading system based on machine vision and edge computing technology according to the present invention.
Fig. 3 is a schematic diagram of an image processing module of an automated meter reading method based on machine vision and edge computing technology according to the present invention.
Fig. 4 is a schematic diagram of an image processing module of an automated meter reading method based on machine vision and edge computing technology according to the present invention.
Fig. 5 is a schematic diagram of a data processing module of an automated meter reading method based on machine vision and edge computing technology according to the present invention.
Fig. 6 is a schematic diagram of a communication module of an automated meter reading method based on machine vision and edge computing technology according to the present invention.
Detailed Description
The invention is further described with reference to the following description of the drawings and detailed description.
As shown in fig. 1 to 6, an automatic meter reading system based on machine vision and edge computing technology comprises a terminal device and a remote server, wherein the terminal device is an edge computing device, the edge computing device comprises a camera module, an image processing module, a data processing module and a communication module, the camera module shoots a pointer instrument, the output end of the camera module is connected with the image processing module, the output end of the image processing module is connected with the data processing module, and the data processing module is connected with the remote server through the communication module.
The camera module comprises a camera 3, a horizontal moving mechanism 1, a vertical moving mechanism 2 and an LED light supplementing lamp.
The invention integrates the communication module, the camera module and the memory module by using the circuit board, realizes image recognition on hardware, namely meter reading and transmits a final result back to the server, and combines the narrowband internet of things technology with the image recognition technology.
The following steps are carried out based on the system:
step 1, starting edge computing equipment, registering the equipment as computing edge, and starting a camera module on the equipment after obtaining permission;
step 2, retrieving pointer type instruments in a screen according to the pre-trained neural network model;
step 3, preprocessing (scaling, noise reduction, gray level and binarization) the detected instrument coordinates, and outputting pointer instrument images read by a common program;
step 4, data inference is carried out according to a trigonometric function through pointer included angle positioning, circle center positioning, instrument minimum value and step length;
step 5, average filtering processing is carried out on the inferred numerical values, and final reasonable data are output;
and 6, pushing the gigabit network port/4G network card/wifi and the like to a remote server by using the edge computing terminal after the communication module performs group verification on the computing result, and completing data archiving.
The method can automatically read the instrument data without stopping the machine and modifying the machine, and does not influence the normal industrial production requirements (such as manual inspection, etc.), and the detailed technical scheme comprises the following steps:
hardware part (one)
In combination with the actual use scenario, the main functions of the hardware part are as follows: firstly, a basic operation environment can be provided for software; secondly, a data source can be provided for software; thirdly, a mode capable of supporting at least two types of data remote transmission is adopted; fourthly, the device is small enough and can be deployed in any scene; fifthly, the LED can be driven to supplement the light source in the shooting process.
(II) software part
The software is used as a soul of the method, and solves the problems of driving of physical hardware such as cameras, LED lamps and the like, shooting position adjustment, dynamic image acquisition, image content analysis, data processing, remote server communication and the like. The relationship is illustrated in fig. 1.
As can be understood from fig. 1, the present invention is implemented as a physical middleware, which is located between an industrial device and a remote server, and uses a machine vision mode to implement non-modification upgrade on an original industrial device, so as to complete heterogeneous system interconnection, where each module is described as follows:
1. camera module
The camera module shown in fig. 2 is composed of 2 steering engines, 1 camera and 1 LED lamp group, so that equipment image shooting is carried out at different angles and different illumination by using specific resolution.
The resolution range of the camera module is not less than 720P (1280 multiplied by 720), the image shooting of instruments and meters can be realized, the moving angle in the horizontal direction is 0 to 180 degrees, the left and right movement of the camera can be realized, the moving angle in the vertical direction is 0 to 180 degrees, the up and down movement of the camera can be realized, the LED light supplementing is supported, and the light supplementing adjustment can be performed aiming at the scene illumination problem.
2. Image processing module
As shown in fig. 3, the module is responsible for solving the image problem shot by the camera module. The method comprises three parts of target scene judgment, instrument positioning and numerical value inference.
As shown in fig. 4, as core content of the present invention, the specific processing method is as follows:
and (3) scene judgment: the method is used for determining the use scene of the model and the equipment, judging whether the video stream transmitted back by the camera module contains the instrument or not by using the LeNet neural network model, and can greatly reduce the graphic calculation pressure of the terminal. When the scene does not contain any instrument, the terminal equipment does not perform graphic calculation, so that the energy consumption is reduced, and the service life of the equipment is prolonged; according to the specifications of the artificial intelligent neural network, instrument images (including forward direction and reverse direction) are required to be collected as far as possible, label marking is performed manually and training is performed, a hook is used for outputting a model result with an optimal threshold value, and the result is used in the next step;
positioning an instrument: the method comprises the steps of positioning and intercepting meters in an image one by using a preprocessing model, wherein after intercepting, normalization processing, namely size consistency, scaling consistency and the like, is needed due to the fact that the size of the image is in a problem, and noise reduction processing is needed to be carried out on the processed image so as to improve the accuracy of numerical value inference;
numerical value inference: after the instrument is determined, the center coordinates of the instrument are determined by using a machine vision technology, four image boundaries are drawn by using coordinates of (0, 0), and then pointer coordinates are determined by using the technology to judge which image boundary the instrument is positioned in. And taking all the calculated key parameters as a formula I, and finally calculating according to a trigonometric function formula to obtain a final result.
3. Data processing module
As shown in fig. 5, the data processing module is configured to process the result output by the image processing module, where the processing content is digital filtering and local buffering.
A data filter: because the numerical value calculation is a complex process, particularly a plurality of key parameters are needed to calculate each other, numerical value inference is carried out in an image processing module, a certain numerical value deviation can be generated when the influence of illumination, fluctuation and the like occurs, a data filter is responsible for filtering the generated numerical value, and when the numerical value exceeds a maximum value and is smaller than a minimum value, and sudden increase, sudden decrease and the like, fusing is triggered, namely, the numerical value is calculated again, and is subjected to recursive processing continuously, so that the numerical value conforming to the filtering is finally output;
local caching: the data entering the filter is locally cached, and the method is used for key basis of self-learning and parameter adjustment of the filter and is also suitable for a solution for temporary storage under the condition of network fluctuation or no network.
4. Communication module
As shown in fig. 6, the module for communicating with the remote server supports the cellular mobile network, WIFI, gigabit network, bluetooth and other modes to synchronize the locally cached data according to the agreed format, and may also receive the command issued by the remote server to perform simple operation and maintenance, such as angle adjustment, on/off of the lamp and the like.
Packet grouping/unpacking: assembling or disassembling the data according to the message format of the remote server;
checking, encrypting/decrypting: performing RSA asymmetric encryption on the data content, and performing integrity check by using SHA256 to ensure that no message is lost;
event subscription: and processing and corresponding the instruction issued by the remote server.
An automatic meter reading method and system based on machine vision and edge computing technology has the following advantages:
1) The method is non-invasive, does not change the appearance and does not influence the automatic meter reading of the pointer type instrument of the existing industrial equipment/system;
2) The edge computing device is distributed and has unified control and management;
3) The method comprises the steps of identifying and positioning meters based on a neural network, positioning pointers based on machine vision, positioning dials and calculating numerical values of four image boundaries;
4) Wifi, cellular mobile network, gigabit network, bluetooth and other wired/wireless data transmission are supported;
5) And the secondary development of software and information encryption/decryption are supported.
The invention relates to a method for automatically reading meter by a pointer instrument based on edge calculation (combining artificial intelligence and machine vision); the data acquisition strategy of the industrial pointer instrument based on the edge computing equipment is adopted.
The invention discloses a method for automatically reading meter under the conditions of no change of industrial scene, no invasion and no influence on the work of the existing meter. And shooting an instrument image by adopting a camera, finishing logic operation on the edge computing terminal equipment (namely, replacing human eyes to read the pointer instrument by a machine vision method), and finally finishing data transmission by the equipment to realize reporting.
The method is based on the machine vision technology to read the instrument, and the distributed intelligent edge computing technology is used for synchronously collecting data. The intelligent manufacturing system is suitable for various application scenes such as intelligent cities, intelligent communities, intelligent manufacturing, traditional industry and the like, is simple and easy to use, and is rich in imagination and creativity.
The foregoing is a further detailed description of the invention in connection with the preferred embodiments, and it is not intended that the invention be limited to the specific embodiments described. It will be apparent to those skilled in the art that several simple deductions or substitutions may be made without departing from the spirit of the invention, and these should be considered to be within the scope of the invention.
Claims (10)
1. An automatic meter reading method based on machine vision and edge computing technology is characterized by comprising the following steps:
s1, starting edge computing equipment, wherein the edge computing equipment is terminal equipment and comprises a camera module, an image processing module, a data processing module and a communication module, the edge computing equipment is used as a computing edge to be registered, and after permission is acquired, the camera module on the edge computing equipment is started;
s2, shooting a pointer instrument through a camera module, and outputting a shot image or video stream to an image processing module on the edge computing equipment;
s3, judging a scene through the image processing module, judging whether an image or a video stream transmitted back by the camera module contains a pointer instrument, if not, returning to the step S2, and if so, entering the next step;
s4, positioning the instrument through the image processing module;
s5, carrying out numerical value inference through the image processing module and outputting an inferred result to a data processing module on the edge computing equipment;
s6, carrying out mean value filtering processing on the inferred result of the image processing module through the data processing module to obtain a meter reading numerical value;
and S7, transmitting the filtered meter reading numerical value to a remote server through a communication module.
2. The automated meter reading method based on machine vision and edge computing technology of claim 1, wherein: in step S3, whether the image or video stream contains a pointer instrument is retrieved according to the pre-trained neural network model.
3. The automated meter reading method based on machine vision and edge computing technology of claim 2, wherein: the pre-trained neural network model is a LeNet neural network model.
4. The automated meter reading method based on machine vision and edge computing technology of claim 2, wherein: in step S4, the retrieved pointer instrument coordinates are preprocessed, including scaling, noise reduction, gray scale, and binarization, to obtain a pointer instrument image.
5. The automated meter reading method based on machine vision and edge computing technology of claim 4, wherein: in step S5, the central coordinates of the instrument in the pointer instrument image are determined by the machine vision technology, four image boundaries are drawn by taking the coordinates as (0, 0), the pointer coordinates are determined, the position of the image boundary is determined, the instrument value, the initial position, the minimum value, the maximum value and the step size are determined, and finally the inferred value is calculated according to the trigonometric function formula.
6. The automated meter reading method based on machine vision and edge computing technology of claim 1, wherein: in step S6, the average filtering process is performed through the data filter, the data filter is responsible for filtering the generated values, when the value exceeds the maximum value and is smaller than the minimum value, and the sudden increase and decrease trigger fusing is performed, namely, the operation is performed again, the recursion process is continued, finally, the meter reading value conforming to the filtering is output, and the data entering the data filter is locally cached for self-learning and parameter adjustment of the data filter.
7. The automated meter reading method based on machine vision and edge computing technology of claim 1, wherein: in step S7, the communication module performs group verification on the calculation result, and then uses the communication interface on the edge computing device to push the meter reading value to the remote server, so as to complete data archiving.
8. An automatic meter reading system based on machine vision and edge computing technology is characterized in that: for implementing the method of any one of claims 1 to 7.
9. The automated meter reading system based on machine vision and edge computing technology of claim 8, wherein: the system comprises terminal equipment and a remote server, wherein the terminal equipment is edge computing equipment, the edge computing equipment comprises a camera module, an image processing module, a data processing module and a communication module, the camera module shoots a pointer instrument, the output end of the camera module is connected with the image processing module, the output end of the image processing module is connected with the data processing module, and the data processing module is connected with the remote server through the communication module.
10. The automated meter reading system based on machine vision and edge computing technology of claim 9, wherein: the camera module comprises a camera, a horizontal moving mechanism, a vertical moving mechanism and an LED light supplementing lamp.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010677856.8A CN111783727B (en) | 2020-07-15 | 2020-07-15 | Automatic meter reading method and system based on machine vision and edge computing technology |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010677856.8A CN111783727B (en) | 2020-07-15 | 2020-07-15 | Automatic meter reading method and system based on machine vision and edge computing technology |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN111783727A CN111783727A (en) | 2020-10-16 |
| CN111783727B true CN111783727B (en) | 2023-12-26 |
Family
ID=72767786
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010677856.8A Active CN111783727B (en) | 2020-07-15 | 2020-07-15 | Automatic meter reading method and system based on machine vision and edge computing technology |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111783727B (en) |
Families Citing this family (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112556736A (en) * | 2020-12-07 | 2021-03-26 | 苏州爱博斯蒂低碳能源技术有限公司 | Equipment for remote monitoring of digital instrument |
| TWI801830B (en) * | 2021-03-26 | 2023-05-11 | 谷林運算股份有限公司 | Monitoring system and method thereof |
| CN114143446A (en) * | 2021-10-20 | 2022-03-04 | 深圳航天智慧城市***技术研究院有限公司 | Histogram identification method, system, storage medium and equipment based on edge calculation |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106570948A (en) * | 2016-11-01 | 2017-04-19 | 东南大学 | Transformer substation intelligent meter reading system provided with portable device |
| CN106874910A (en) * | 2017-01-19 | 2017-06-20 | 广州优库电子有限公司 | The self-service meter reading terminal of low-power consumption instrument long-distance and method based on OCR identifications |
| CN108921176A (en) * | 2018-06-22 | 2018-11-30 | 河海大学常州校区 | A kind of pointer instrument positioning and recognition methods based on machine vision |
| CN110197131A (en) * | 2019-05-10 | 2019-09-03 | 三峡大学 | A method of remotely monitoring lightning arrester discharge action frequency based on machine vision |
| CN110942058A (en) * | 2019-07-31 | 2020-03-31 | 广东互动电子网络媒体有限公司 | Instrument data reading method based on CCD machine vision recognition |
| KR20200063340A (en) * | 2018-11-22 | 2020-06-05 | 한국클라우드컴퓨팅연구조합 | Method and system that machine learning-based quality inspection using the cloud |
-
2020
- 2020-07-15 CN CN202010677856.8A patent/CN111783727B/en active Active
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106570948A (en) * | 2016-11-01 | 2017-04-19 | 东南大学 | Transformer substation intelligent meter reading system provided with portable device |
| CN106874910A (en) * | 2017-01-19 | 2017-06-20 | 广州优库电子有限公司 | The self-service meter reading terminal of low-power consumption instrument long-distance and method based on OCR identifications |
| CN108921176A (en) * | 2018-06-22 | 2018-11-30 | 河海大学常州校区 | A kind of pointer instrument positioning and recognition methods based on machine vision |
| KR20200063340A (en) * | 2018-11-22 | 2020-06-05 | 한국클라우드컴퓨팅연구조합 | Method and system that machine learning-based quality inspection using the cloud |
| CN110197131A (en) * | 2019-05-10 | 2019-09-03 | 三峡大学 | A method of remotely monitoring lightning arrester discharge action frequency based on machine vision |
| CN110942058A (en) * | 2019-07-31 | 2020-03-31 | 广东互动电子网络媒体有限公司 | Instrument data reading method based on CCD machine vision recognition |
Non-Patent Citations (1)
| Title |
|---|
| 陈众威 ; 张庆辉 ; 王彦辉 ; .基于机器视觉的智能仪表的数字定位技术.数字技术与应用.2016,(11),全文. * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN111783727A (en) | 2020-10-16 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111783727B (en) | Automatic meter reading method and system based on machine vision and edge computing technology | |
| CN109919402A (en) | A kind of intelligent patrol detection management method, system and terminal device | |
| CN110807460B (en) | Transformer substation intelligent patrol system based on image recognition and application method thereof | |
| CN201413534Y (en) | Digitized transformation patrolling device | |
| CN111383429A (en) | Method, system, device and storage medium for detecting dress of workers in construction site | |
| CN108896799A (en) | Very Important Person electric energy meter image-pickup intelligent-reading remote transmission device | |
| CN105005281A (en) | Intelligent transformer substation inspection real-time data display system and method based on Google glasses | |
| CN112418194B (en) | Video monitoring system and method for intelligent community | |
| CN104587630B (en) | A kind of fire extinguisher intelligent management system and management method | |
| CN110048509A (en) | A kind of substation inspection system and method based on MR intelligent safety helmet | |
| CN112508735A (en) | Three-dimensional power station service management system based on data aggregation | |
| CN102736545A (en) | Foolproof monitoring method and device for automated device | |
| CN106791721A (en) | Development machine abnormal work and disaster underground warning system based on infrared image | |
| CN112966552B (en) | Routine inspection method and system based on intelligent identification | |
| CN205028413U (en) | Wisdom city monitored control system | |
| CN208027552U (en) | The wireless kilowatt meter reading-out system with Intelligent supplemental lighting function based on NB-IoT and visual identity | |
| CN203179187U (en) | Multi-polar intelligent anti-external-damage early warning device | |
| CN105513129B (en) | Based on the automatic counter system of bar of laser 3D modeling | |
| CN204469088U (en) | A kind of fire extinguisher intelligent management system | |
| CN203870456U (en) | Valve control system with video monitoring used for natural gas long-distance transport pipe | |
| CN115393340A (en) | AI vision product quality detection system based on 5G algorithm | |
| CN205210974U (en) | Can real -time supervision substation equipment maintain and debugging process patrol and examine robot | |
| CN111145046A (en) | Digital management method and system for hydraulic and hydroelectric engineering construction process | |
| CN106791719A (en) | Development machine abnormal work and calamity forecast system based on image trace monitoring device | |
| CN112128624A (en) | Gas digital twin 3D visual intelligent operation and maintenance system |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |