CN114627461A - Method and system for high-precision identification of water gauge data based on artificial intelligence - Google Patents
Method and system for high-precision identification of water gauge data based on artificial intelligence Download PDFInfo
- Publication number
- CN114627461A CN114627461A CN202210220358.XA CN202210220358A CN114627461A CN 114627461 A CN114627461 A CN 114627461A CN 202210220358 A CN202210220358 A CN 202210220358A CN 114627461 A CN114627461 A CN 114627461A
- Authority
- CN
- China
- Prior art keywords
- water gauge
- image
- digital
- reading
- character
- 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
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/23—Clustering techniques
-
- 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
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
- G06N3/088—Non-supervised learning, e.g. competitive learning
Abstract
The invention provides a method and a system for identifying water gauge data with high precision based on artificial intelligence. The method comprises the steps that the number on the water gauge is identified by adopting an artificial intelligence technology, and the number position information on the water gauge are identified through supervised training and learning of a large number of pictures marked with numbers; the incomplete character 'E' on the water gauge is recognized, the image is processed by technologies of cluster analysis, noise reduction, angle correction and the like, and the reading of the incomplete character 'E' is calculated by determining the pixel coordinate of the incomplete part at the lowest end of the character 'E'. The water gauge data identification method can theoretically achieve error-free data identification, the error can be controlled within 1cm in practical application, unattended and real-time water gauge reading identification can be achieved at high precision, and the water gauge data identification method is high in transportability and has important practical value.
Description
Technical Field
The invention belongs to the field of artificial intelligence, image recognition and environmental protection, and particularly relates to a method and a system for recognizing water gauge data at high precision based on artificial intelligence, which solve the problem that the water gauge water level data can be recognized without supervision.
Background
The water level is an important mark reflecting the change of water body and water flow, the water level observation can be directly used for hydrologic information forecast, and the water information is provided for the construction, application and management of flood control, drought control, shipping and hydraulic engineering, and the continuous water level data is an important basis for the construction and planning of water conservancy and hydropower, channels, water environment and the like.
At present, automatic water level observation equipment comprises a self-recording water level gauge, a water pressure water level gauge, an ultrasonic water level gauge, a radar water level gauge and the like, and the equipment has high purchase and maintenance cost and great influence on precision by environment; the traditional water gauge is the most direct water level observation equipment, has low installation cost, high precision and easy verification, but needs manual monitoring and is difficult to realize long-term continuous monitoring; the traditional image identification technology is mainly based on gray level images, extracts characteristic points and calculates water gauge reading, and is greatly influenced by environment and image quality. The application of the artificial intelligence technology provides a good solution for identifying the water level of the water gauge. The patent with publication number CN 109508630a discloses an artificial intelligence water gauge identification method, which includes two training models, a water gauge object detection model and a water gauge scale identification model, wherein a water gauge image is input into the water gauge object detection model which is pre-trained, the position of the water gauge is detected, the water gauge image in the original image is intercepted, then the water gauge image is input into the water gauge scale identification model which is deeply trained, and statistical data of numbers and scales in the water gauge image is output. The patent with publication number CN 111680606B discloses a low-power consumption water level remote measuring system based on artificial intelligence identification water gauge, which comprises a remote measuring station, a central station server, a cloud server and the like, wherein the cloud server realizes water gauge data identification by using an intelligent algorithm.
At present, the traditional non-artificial intelligent image detection mainly realizes specific scene recognition based on image gray scale and feature extraction, and has poor adaptability; the existing artificial intelligent water gauge identification method needs to train two models and arrange two data sets, has great workload, has poor identification capability and low precision for the inclined water gauge, the water gauge with the incomplete character E and the water gauge with the background color close to the environmental color, and even can not identify the inclined water gauge and the water gauge with the background color close to the environmental color.
Disclosure of Invention
Aiming at the defects in the prior art, the invention aims to provide a method and a system for identifying water gauge data with high precision based on artificial intelligence.
The technical scheme adopted by the invention is as follows:
a method for identifying water gauge data with high precision based on artificial intelligence comprises the following steps:
step one, acquiring a water gauge image: acquiring a field water gauge image through a high-definition camera, and transmitting the field water gauge image to an indoor computer;
and secondly, identifying the number and the number position of the water gauge: performing supervised training and learning on the image marked with the number through a convolutional neural network to obtain the digital characteristic weight, identifying the image of the field water gauge by using the trained model, obtaining the number and the position information of the number, and determining the lowest number of the water gauge;
thirdly, clustering and analyzing the water gauge images: intercepting the images of the numbers and the characters E at the lowest end of the water gauge according to the number position information, carrying out image clustering analysis in an unsupervised learning mode, comparing the colors of the characters E, and binarizing the intercepted images;
fourthly, identifying the incomplete 'E' scale of the water gauge: performing neighborhood noise reduction and correction on the binarized image, reading image data from the bottom of the image, identifying an RGB value same as that of the character 'E', determining a pixel coordinate of the incomplete character 'E', and calculating the reading of the incomplete character 'E' on a water gauge;
and fifthly, calculating the reading of the water gauge: and calculating the reading of the water gauge according to the number of the lowest end of the water gauge determined in the second step and the reading of the incomplete part E determined in the fourth step.
Further, the first step comprises the steps of:
a high-definition camera is erected in front of a water gauge, an image is shot at certain intervals, and the image is transmitted to a designated folder of a computer.
Further, the second step comprises the following steps:
step 2.1, image annotation: collecting water gauge images or digital images, and carrying out digital labeling to obtain a data set required by artificial intelligence training, verification and testing;
step 2.2, image depth training and learning: carrying out supervised deep training and learning on the marked training set, the marked verification set and the marked test set by utilizing the convolutional neural network to obtain a convolutional neural network model;
step 2.3, identifying the digital information of the water gauge: detecting a field water gauge image by using a convolution neural network model after deep learning, extracting water gauge numbers and digital position information, wherein the digital position information comprises a digital center coordinate and digital image width and height pixel information, and determining the lowest end number of the water gauge according to a digital longitudinal pixel coordinate.
Further, the third step comprises the following steps:
step 3.1, image clipping: intercepting the lowest digit of the water gauge and the image containing the character E at the lower end according to the digit position information in the step 2.3;
step 3.2, image clustering analysis: and converting the intercepted image information into two-dimensional data, performing cluster analysis by using an unsupervised learning mode, and binarizing the intercepted image by setting a threshold value and comparing the color of the character E.
Further, the fourth step comprises the following steps:
step 4.1, image noise reduction correction: performing neighborhood noise reduction and angle correction processing by using a computer image analysis technology to obtain an image without mottle and without inclination of a character E;
step 4.2, calculating the incomplete E scale of the water gauge: reading the image after noise reduction correction, traversing upwards from the bottom of the image, acquiring pixel coordinates with the same RGB value as the character E, and converting the reading of the incomplete character E by using the maximum longitudinal pixel coordinate of the incomplete character E.
A system for high-precision identification of water gauge data based on artificial intelligence comprises:
the water gauge image acquisition module is used for shooting a water gauge image and transmitting the shot field water gauge image to the computer; the water gauge image acquisition module can adopt a high-definition camera;
the water gauge digital and digital position identification module is used for supervised training and learning the image marked with the number through a convolutional neural network, acquiring digital characteristic weight, identifying the field water gauge image by using the trained model, acquiring digital and digital position information and determining the lowest end number of the water gauge;
the water gauge image cluster analysis module is used for intercepting the images of the numbers and the characters E at the lowest end of the water gauge according to the number position information, performing image cluster analysis in an unsupervised learning mode, comparing the colors of the characters E and binarizing the intercepted images;
the water gauge incomplete character E scale identification module is used for performing neighborhood noise reduction and correction on the binarized image, reading image data from the bottom of the image, identifying an RGB value identical to the character E, determining a pixel coordinate of the incomplete character E, and calculating the reading of the incomplete character E on the water gauge;
and the water gauge reading calculation module is used for calculating the water gauge reading according to the lowest end number of the water gauge determined by the digital information identification module and the reading of the incomplete part E determined by the water gauge incomplete part E scale identification module.
The method and the system for identifying the water gauge data with artificial intelligence and high precision provided by the invention are based on the technologies of convolutional neural network, image cluster analysis, image noise reduction correction and the like, provide an automatic and high-precision method and system for identifying the water gauge water level without manual intervention, and have the following beneficial effects:
(1) the invention provides a method and a system for identifying water gauge data with high precision based on artificial intelligence, which mainly aim to solve the problems of low precision of water gauge data acquisition, complex data set acquisition and high labor cost;
(2) based on artificial intelligence technologies such as convolutional neural network and cluster analysis, a large amount of labeled data are trained and learned, the numbers and the number position information of a water gauge in an image are identified by utilizing a trained artificial intelligence model, the image with the numbers and characters E at the lowest end is intercepted according to the number position, and the water level reading of the water gauge is identified through image processing technologies such as cluster analysis, noise reduction and correction;
(3) according to the method provided by the invention, only a digital artificial intelligence recognition model needs to be established, compared with a water gauge and a scale recognition model, a digital artificial intelligence recognition model data set is easier to collect and label, the labor and time cost is greatly saved, and meanwhile, the recognition precision of the model is greatly improved due to obvious digital characteristics;
(4) according to the method provided by the invention, according to digital information extracted by an artificial intelligence model, through image processing technologies such as cluster analysis, noise reduction and correction, the image recognition degree is improved, the image noise point is reduced, and the recognition precision of incomplete characters E is greatly improved;
(5) the method provided by the invention can be programmed into a program, manual intervention is not needed, water level data of the water gauge under different scenes can be continuously identified according to set time intervals aiming at the water gauges with different colors (red, blue or other colors are automatically distinguished), and through practical verification, a program system compiled based on the method is high in identification precision, good in adaptability, high in speed and strong in transportability, and provides technical support for water level data identification of the water gauge.
Drawings
FIG. 1 is a schematic flow chart of a method for identifying water gauge data based on artificial intelligence with high precision according to the present invention;
FIG. 2 is a flowchart illustrating an example of a method for identifying water gauge data based on artificial intelligence with high precision according to the present invention;
FIG. 3 is a schematic diagram of the calculation of the defect "E" in the present invention;
FIG. 4 is a calculation result display interface diagram of the high-precision identification water gauge data system based on artificial intelligence of the present invention;
FIG. 5 is a comparison diagram of results of actual verification of the method for identifying water gauge data based on artificial intelligence with high precision.
Detailed Description
In order to make the technical solutions and advantages of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the embodiments of the present invention and the accompanying drawings, and it is obvious that the described embodiments are some, but not all, embodiments of the present 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 solution of the present invention will be described in further detail with reference to the following embodiments and accompanying drawings.
Referring to fig. 1, an embodiment of the present invention provides a method for identifying water gauge data with high precision based on artificial intelligence, which includes acquiring a field water gauge image through a high definition camera, identifying water gauge numbers and number position information in the image by using an artificial intelligence model, intercepting an image including numbers and characters "E" at the bottom, binarizing the intercepted image through a cluster analysis method, denoising and correcting the image, traversing upwards from the bottom of the image, acquiring pixel coordinates of the same RGB as the characters "E", converting the length represented by the incomplete characters "E", and calculating a water gauge reading by combining the numbers of the bottom of the water gauge. The method specifically comprises the following steps:
step one, acquiring a water gauge image
(1) Image shooting: erecting a high-definition camera in front of a water gauge, shooting an image at certain intervals, and transmitting the image to a designated folder of a computer;
step two, identifying the number and the number position of the water gauge
(1) Image labeling: collecting 1000 digital pictures, and carrying out digital labeling by using LabelImg;
(2) and (3) image depth training and learning: dividing the marked pictures and corresponding marked files into a training set, a verification set and a test set according to the ratio of 8:1:1, and training and learning the data set by using a convolutional neural network model to obtain a convolutional neural network model;
(3) and (3) identifying the digital information of the water gauge: reading the last field water gauge image transmitted into the computer by using the convolutional neural network model after deep learning, identifying the number of the water gauge in the image (as shown in figure 2, the number identified in the present example is 5, 4, 3, 2 and 1) and the corresponding center coordinate of the number and the width and height pixel information of the digital image, and determining the lowest end number of the water gauge (as shown in figure 2, the present example is '1') according to the longitudinal pixel coordinate of the number.
Step three, water gauge image clustering analysis
(1) Image cutting: calculating the pixel coordinate (168, 560) of the upper left corner of the number '1' according to the central coordinate of the number '1' in the original image and the width and height pixel information of the digital image, and expanding the length of 3 characters 'E' downwards and expanding the width of 2 characters 'E' rightwards based on the pixel coordinate of the upper left corner of the number '1' to ensure that the intercepted image comprises the information of all the incomplete characters 'E' (as shown in figure 2);
(2) image clustering analysis: converting the intercepted image into RGB two-dimensional data, setting a cluster center to be 4 (the cluster center is at least 2), performing color cluster analysis by using an unsupervised learning mode, comparing red or blue RGB values, determining the cluster center color of the water gauge number and the character 'E' (the RGB values are 176,53 and 54), setting a threshold value to be 3000, setting the number and the character 'E' in the intercepted image to be black, and setting the other numbers and the character 'E' to be white.
Step four, identifying incomplete 'E' reading of water gauge
(1) And (3) image noise reduction correction: according to a neighborhood noise reduction algorithm, setting a threshold value of the binarized image to be 3, and denoising the image; detecting the boundary of the character E, calculating the inclination angle (the inclination angle of the example is 0.095 degrees), and performing image rotation correction (as shown in FIG. 2);
(2) calculating the incomplete E scale of the water gauge: for the image after noise reduction correction, traversing upwards from the bottom of the image (as shown in fig. 3), acquiring coordinates of pixels with the same RGB values as the characters "E", respectively calculating the pixel heights of the incomplete characters "E", taking the maximum value of the pixel heights, combining the pixel heights (66.25) of the single characters "E", and converting the length (reading is 0.66, unit is dm, and 1dm is 10cm) of the incomplete characters "E";
(3) and (3) calculating the reading of the water gauge: the lowest digit is "1", the incomplete character "E" reads 0.66dm, and the water gauge reads 0.34 dm.
The embodiment of the invention also provides a system for identifying the water gauge data based on artificial intelligence with high precision, which comprises the following steps:
the water gauge image acquisition module is used for shooting a water gauge image and transmitting the shot field water gauge image to the computer; the water gauge image acquisition module can adopt a high-definition camera;
the water gauge number and number position identification module is used for supervised training and learning the image marked with the number through a convolutional neural network, acquiring the number characteristic weight, identifying the image of the field water gauge by using the trained model, acquiring the number and number position information and determining the lowest end number of the water gauge;
the water gauge image cluster analysis module is used for intercepting the images of the numbers and the characters E at the lowest end of the water gauge according to the number position information, performing image cluster analysis in an unsupervised learning mode, comparing the colors of the characters E and binarizing the intercepted images;
the water gauge incomplete character E scale identification module is used for performing neighborhood noise reduction and correction on the binarized image, reading image data from the bottom of the image, identifying an RGB value identical to the character E, determining a pixel coordinate of the incomplete character E, and calculating the reading of the incomplete character E on the water gauge;
and the water gauge reading calculation module is used for calculating the water gauge reading according to the lowest end number of the water gauge determined by the digital information identification module and the reading of the incomplete part E determined by the water gauge incomplete part E scale identification module.
According to the computing thought provided by the invention, a system program (as shown in figure 4) is written by utilizing Python, and the water level identification of the water gauge can be realized without manual supervision and operation. Here, in order to show a system program (fig. 4) written based on the present invention, a method for using the system is described in detail below, and the system includes two modes of automatic recognition and manual recognition, and includes the following operation steps:
(1) preparing basic data: and collecting the pictures containing the numbers, and labeling the pictures by using labeling software.
(2) The method comprises the following operation steps: carrying out supervised deep training and learning on a marked training set, a marked verification set and a marked test set by utilizing a convolutional neural network to obtain a digital feature model; secondly, operating an artificial intelligent water gauge identification system, and opening a program main interface (figure 4), wherein the upper part of the interface is an image display area, the middle part of the interface is a parameter setting area, and the lower part of the interface is a result table display area and an operation area; the automatic identification mode: clicking an 'automatic identification' button to execute functions of a water gauge number and number position identification module, a water gauge image clustering analysis module, a water gauge incomplete 'E' scale identification module and a water gauge reading calculation module, wherein the button becomes unavailable at the moment, a system detects the number of pictures in a preset folder every 1 second, when the number of the pictures is increased or decreased, an artificial intelligent water gauge identification program is activated, water gauge data identification is carried out on the finally transmitted water gauge picture, a result image and data are displayed on a system main interface, the data are automatically stored to a specified file, and the 'stop identification' button is clicked to suspend identification, wherein the 'automatic identification' button is available at the moment and the 'stop identification' button is unavailable; fourthly, manually identifying the mode: clicking a manual identification button to identify the water gauge image for the last picture transmitted into the folder; clicking the 'data save in addition' button to save the data in the main interface form in a self-defined folder.
Description on parameter setting: the default of the system is 4, the default of the number of the clustering centers is 3000, the color difference threshold value is 3000, the set default parameter is a value after comparison and optimization, the requirement of water gauge data identification can be basically met, and meanwhile, the parameter can be adjusted according to the requirement so as to be suitable for different scenes.
Compared with the traditional water level monitoring and water gauge identification method, the water gauge water level identification technology provided by the invention can be used for carrying out high-precision real-time automatic unattended water gauge identification, and the related technology develops a program for carrying out actual verification, and 30 groups of water gauge images are identified, wherein 17 groups of errors are approximately 0, and 13 groups of errors are within 1cm (as shown in figure 5).
The above description is only an embodiment of the present invention, but the scope of the present invention is not limited thereto, and any changes or substitutions that can be easily conceived by those skilled in the art within the technical scope of the present invention are included in the scope of the present invention. Therefore, the protection scope of the present invention shall be subject to the protection scope of the claims.
Claims (6)
1. A method for high-precision identification of water gauge data based on artificial intelligence is characterized by comprising the following steps: the method comprises the following steps:
step one, acquiring a water gauge image: acquiring a field water gauge image through a high-definition camera, and transmitting the field water gauge image to an indoor computer;
and secondly, identifying the number and the number position of the water gauge: through a convolutional neural network, supervised training is carried out, images with numbers marked are learned, digital characteristic weights are obtained, the trained model is used for identifying images of the field water gauge, numbers and digital position information are obtained, and the numbers at the lowest end of the water gauge are determined;
thirdly, clustering and analyzing the water gauge images: intercepting the images of the numbers and the characters E at the lowest end of the water gauge according to the number position information, carrying out image clustering analysis in an unsupervised learning mode, comparing the colors of the characters E, and binarizing the intercepted images;
fourthly, identifying incomplete E scales of the water gauge: performing neighborhood noise reduction and correction on the binarized image, reading image data from the bottom of the image, identifying an RGB value same as that of the character 'E', determining a pixel coordinate of the incomplete character 'E', and calculating the reading of the incomplete character 'E' on a water gauge;
and fifthly, calculating the reading of the water gauge: and calculating the reading of the water gauge according to the number of the lowest end of the water gauge determined in the second step and the reading of the incomplete part E determined in the fourth step.
2. The method for identifying the water gauge data with high precision based on the artificial intelligence as claimed in claim 1, wherein the method comprises the following steps: the first step comprises the following steps:
a high-definition camera is erected in front of a water gauge, an image is shot at certain intervals, and the image is transmitted to a designated folder of a computer.
3. The method for identifying the water gauge data with high precision based on the artificial intelligence as claimed in claim 2, wherein the method comprises the following steps: the second step comprises the following steps:
step 2.1, image annotation: collecting water gauge images or digital images, and carrying out digital labeling to obtain a data set required by artificial intelligence training, verification and testing;
step 2.2, image depth training and learning: carrying out supervised deep training and learning on the marked training set, the marked verification set and the marked test set by using a convolutional neural network to obtain a convolutional neural network model;
step 2.3, identifying the digital information of the water gauge: detecting a field water gauge image by using a convolution neural network model after deep learning, extracting water gauge numbers and digital position information, wherein the digital position information comprises a digital center coordinate and digital image width and height pixel information, and determining the lowest end number of the water gauge according to a digital longitudinal pixel coordinate.
4. The method for identifying the water gauge data with high precision based on the artificial intelligence as claimed in claim 3, wherein the method comprises the following steps: the third step comprises the following steps:
step 3.1, image clipping: intercepting the lowest digit of the water gauge and the image containing the character E at the lower end according to the digit position information in the step 2.3;
step 3.2, image clustering analysis: and converting the intercepted image information into two-dimensional data, performing cluster analysis by using an unsupervised learning mode, and binarizing the intercepted image by setting a threshold value and comparing the color of the character E.
5. The method for identifying the water gauge data with high precision based on the artificial intelligence as claimed in claim 4, wherein the method comprises the following steps: the fourth step comprises the following steps:
step 4.1, image noise reduction correction: performing neighborhood noise reduction and angle correction processing by using a computer image analysis technology to obtain an image without mottle and without inclination of a character E;
step 4.2, calculating the incomplete E scale of the water gauge: reading the image after noise reduction correction, traversing upwards from the bottom of the image, acquiring the pixel coordinate with the same RGB value as the character E, and converting the reading of the incomplete character E by utilizing the maximum longitudinal pixel coordinate of the incomplete character E.
6. The utility model provides a system based on artificial intelligence high accuracy discernment water gauge data which characterized in that: the method comprises the following steps:
the water gauge image acquisition module is used for shooting a water gauge image and transmitting the shot field water gauge image to the computer; the water gauge image acquisition module can adopt a high-definition camera;
the water gauge digital and digital position identification module is used for supervised training and learning the image marked with the number through a convolutional neural network, acquiring digital characteristic weight, identifying the field water gauge image by using the trained model, acquiring digital and digital position information and determining the lowest end number of the water gauge;
the water gauge image cluster analysis module is used for intercepting the images of the numbers and the characters E at the lowest end of the water gauge according to the number position information, performing image cluster analysis in an unsupervised learning mode, comparing the colors of the characters E and binarizing the intercepted images;
the water gauge incomplete character E scale identification module is used for performing neighborhood noise reduction and correction on the binarized image, reading image data from the bottom of the image, identifying an RGB value identical to the character E, determining a pixel coordinate of the incomplete character E, and calculating the reading of the incomplete character E on the water gauge;
and the water gauge reading calculation module is used for calculating the water gauge reading according to the lowest end number of the water gauge determined by the digital information identification module and the reading of the incomplete part E determined by the water gauge incomplete part E scale identification module.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202210220358.XA CN114627461A (en) | 2022-03-08 | 2022-03-08 | Method and system for high-precision identification of water gauge data based on artificial intelligence |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202210220358.XA CN114627461A (en) | 2022-03-08 | 2022-03-08 | Method and system for high-precision identification of water gauge data based on artificial intelligence |
Publications (1)
Publication Number | Publication Date |
---|---|
CN114627461A true CN114627461A (en) | 2022-06-14 |
Family
ID=81900932
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202210220358.XA Pending CN114627461A (en) | 2022-03-08 | 2022-03-08 | Method and system for high-precision identification of water gauge data based on artificial intelligence |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN114627461A (en) |
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN115439861A (en) * | 2022-09-30 | 2022-12-06 | 北京中盛益华科技有限公司 | Water gauge recognition method based on OCR |
CN117053903A (en) * | 2023-10-12 | 2023-11-14 | 江苏南水科技有限公司 | Verification method for zero value error of suspension-hammer type water level gauge |
-
2022
- 2022-03-08 CN CN202210220358.XA patent/CN114627461A/en active Pending
Cited By (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN115439861A (en) * | 2022-09-30 | 2022-12-06 | 北京中盛益华科技有限公司 | Water gauge recognition method based on OCR |
CN117053903A (en) * | 2023-10-12 | 2023-11-14 | 江苏南水科技有限公司 | Verification method for zero value error of suspension-hammer type water level gauge |
CN117053903B (en) * | 2023-10-12 | 2023-12-15 | 江苏南水科技有限公司 | Verification method for zero value error of suspension-hammer type water level gauge |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN112766274B (en) | Water gauge image water level automatic reading method and system based on Mask RCNN algorithm | |
CN111950330B (en) | Pointer instrument indication detection method based on target detection | |
WO2021238030A1 (en) | Water level monitoring method for performing scale recognition on the basis of partitioning by clustering | |
CN109460753B (en) | Method for detecting floating object on water | |
CN113160192B (en) | Visual sense-based snow pressing vehicle appearance defect detection method and device under complex background | |
CN109145830B (en) | Intelligent water gauge identification method | |
US20210374466A1 (en) | Water level monitoring method based on cluster partition and scale recognition | |
CN109409365A (en) | It is a kind of that method is identified and positioned to fruit-picking based on depth targets detection | |
CN114627461A (en) | Method and system for high-precision identification of water gauge data based on artificial intelligence | |
CN111598098B (en) | Water gauge water line detection and effectiveness identification method based on full convolution neural network | |
CN108921165B (en) | Water level identification method based on water gauge image | |
CN112149543B (en) | Building dust recognition system and method based on computer vision | |
CN109858480A (en) | A kind of digital instrument recognition methods | |
CN111652089A (en) | Water level automatic identification method and system based on image processing | |
CN111079826A (en) | SLAM and image processing fused construction progress real-time identification method | |
CN115880683B (en) | Urban waterlogging ponding intelligent water level detection method based on deep learning | |
CN115908774B (en) | Quality detection method and device for deformed materials based on machine vision | |
CN116310263A (en) | Pointer type aviation horizon instrument indication automatic reading implementation method | |
CN115908370A (en) | Method for realizing water level detection based on image inverse perspective transformation | |
CN112699824B (en) | Method and device for detecting constant of electric energy meter and storage medium | |
CN115311447A (en) | Pointer instrument indicating number identification method based on deep convolutional neural network | |
CN113516081A (en) | Method for realizing water level identification of virtual water gauge through deep learning image identification technology | |
CN112990064A (en) | Dial pointer identification method based on color segmentation and probability model | |
CN115345854B (en) | Water level identification method based on multi-region search | |
CN114659595B (en) | Water meter durability intelligent test device and method based on Internet of things |
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 |