CN117115488B - Water meter detection method based on image processing - Google Patents

Water meter detection method based on image processing Download PDF

Info

Publication number
CN117115488B
CN117115488B CN202311376869.1A CN202311376869A CN117115488B CN 117115488 B CN117115488 B CN 117115488B CN 202311376869 A CN202311376869 A CN 202311376869A CN 117115488 B CN117115488 B CN 117115488B
Authority
CN
China
Prior art keywords
image
water meter
glass
stain
representing
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
Application number
CN202311376869.1A
Other languages
Chinese (zh)
Other versions
CN117115488A (en
Inventor
蔡敏
王小京
李雪松
韩玉荣
宋健
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Tianjin Tianfei High Tech Valve Co ltd
Original Assignee
Tianjin Tianfei High Tech Valve Co ltd
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Tianjin Tianfei High Tech Valve Co ltd filed Critical Tianjin Tianfei High Tech Valve Co ltd
Priority to CN202311376869.1A priority Critical patent/CN117115488B/en
Publication of CN117115488A publication Critical patent/CN117115488A/en
Application granted granted Critical
Publication of CN117115488B publication Critical patent/CN117115488B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

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
    • G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
    • G06V10/74—Image or video pattern matching; Proximity measures in feature spaces
    • G06V10/75—Organisation of the matching processes, e.g. simultaneous or sequential comparisons of image or video features; Coarse-fine approaches, e.g. multi-scale approaches; using context analysis; Selection of dictionaries
    • 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
    • G06N3/0464—Convolutional networks [CNN, ConvNet]
    • 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
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/0002—Inspection of images, e.g. flaw detection
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/10—Segmentation; Edge detection
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/90—Determination of colour characteristics
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2207/00—Indexing scheme for image analysis or image enhancement
    • G06T2207/20—Special algorithmic details
    • G06T2207/20081—Training; Learning
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2207/00—Indexing scheme for image analysis or image enhancement
    • G06T2207/20—Special algorithmic details
    • G06T2207/20084—Artificial neural networks [ANN]
    • 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
    • G06V2201/07—Target detection
    • 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
    • Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
    • Y02A—TECHNOLOGIES FOR ADAPTATION TO CLIMATE CHANGE
    • Y02A20/00—Water conservation; Efficient water supply; Efficient water use

Landscapes

  • Engineering & Computer Science (AREA)
  • Theoretical Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Evolutionary Computation (AREA)
  • Artificial Intelligence (AREA)
  • General Health & Medical Sciences (AREA)
  • Health & Medical Sciences (AREA)
  • Software Systems (AREA)
  • Computing Systems (AREA)
  • Computational Linguistics (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Biomedical Technology (AREA)
  • Biophysics (AREA)
  • Data Mining & Analysis (AREA)
  • Molecular Biology (AREA)
  • General Engineering & Computer Science (AREA)
  • Mathematical Physics (AREA)
  • Medical Informatics (AREA)
  • Multimedia (AREA)
  • Quality & Reliability (AREA)
  • Databases & Information Systems (AREA)
  • Image Processing (AREA)

Abstract

The invention relates to a water meter detection method based on image processing, which belongs to the technical field of image processing, and comprises the steps of obtaining a water meter surface glass focusing image, adaptively adjusting chromaticity information of a highlight pixel point area, and constructing a glass surface stain model; acquiring a water meter internal focusing image, inputting the water meter internal focusing image and a stain model on the glass surface into a segmentation neural network, and removing background stains to obtain a clear water meter internal image; based on an image matching method, detecting the matching difference of clear internal images of the water meter, and judging whether the water meter needs to be calibrated according to the matching difference.

Description

Water meter detection method based on image processing
Technical Field
The invention belongs to the technical field of image processing, and particularly relates to a water meter detection method based on image processing.
Background
The instrument is an important detection and measurement tool and is also an information source. According to different measuring objects, the instrument converts the measuring signals obtained by the measuring mechanism into digital signals or angular offset through a certain transformation relation.
The pointer water meter has the advantages of visual reading, simple structure, high precision, low manufacturing cost, strong electromagnetic interference resistance, convenient maintenance and the like, and is widely applied to production practice. Usually, the pointer water meter has no digital communication interface, can not convert the measurement signal into a digital signal, and needs to be calibrated manually.
The calibration of the instrument is a tedious, tedious and repeatable task. In some situations requiring a large number of meter readings, the accuracy of obtaining meter readings depends largely on the responsibility and visual fatigue degree of operators, errors and reading errors are easy to occur in the identification process, if errors are found in time, the workload is required to be increased, otherwise serious consequences can be caused.
Because of the differences of appearance and structure of different instruments, according to the implementation principle of the prior art, only the calibration problem of the instrument of the type corresponding to the preset image matching template can be solved, the universality is poor, the use is inflexible, and the reflection and the dirt of the surface glass of the instrument bring great influence on the calibration result
Disclosure of Invention
In order to solve the technical problems, the invention provides a water meter detection method based on image processing, which comprises the following steps:
s1, acquiring a water surface glass focusing image, adaptively adjusting chromaticity information of a highlight pixel point area, and constructing a glass surface stain model;
s2, acquiring a water meter internal focusing image, inputting the water meter internal focusing image and a stain model on the glass surface into a segmentation neural network, and removing background stains to obtain a clear water meter internal image;
and S3, detecting the matching difference of clear internal images of the water meter based on an image matching method, and judging whether the water meter needs to be calibrated according to the matching difference.
Further, step S1 includes:
s11, calculating a brightness difference value of each pixel point in the focusing image of the water surface glass;
s12, adaptively adjusting chromaticity information of the highlight pixel point area;
s13, constructing a stain model of the glass surface according to the self-adaptive adjusted glass focusing image.
Further, in step S2:
extracting an initial characteristic V of an input water meter internal focusing image by using a convolutional neural network Sf, and expanding the initial characteristic V into a concerned characteristic F:
；
k represents a convolutional neural network parameter, and d represents a normalized processing coefficient;
embedding the focus feature and the stain model on the glass surface into the segmentation neural network, obtaining the stain feature mapping, realizing the stain feature mapping and focus feature segmentation, and obtaining the segmentation result of the stain feature removal mapping.
Further, the stain on the glass surface is modeledEmbedding an iterative model->Iterative learning is carried out, and the stain feature mapping Q after each iteration i is as follows:
,i=1,2…，n；
wherein n represents the iteration times, F represents the concerned feature, and each iteration is updated once, so that a new stain feature map is generated and then embedded into the next iteration;
the segmentation result Output of the segmented neural network is represented as follows:
Output=At(Q,V)；
q, V therein represent stain feature map and initial features, respectively; at (Q, V) is a split neural network model, and the split result Output represents that clear internal images of the water meter are obtained after removing stains of corresponding types.
Further, in step S3: set reference picture I R (x, y), the clear internal image of the water meter to be matched is I M (x, y), the image matching model is expressed as:
；
wherein T is opt Representing the matching difference, S represents the reference image I R (x, y) and the image I to be matched M (x, y) similarity between;
according to the difference T of matching opt And detecting a clear internal focusing image of the water meter according to the corresponding numerical value interval range, and judging whether the water meter is required to be calibrated.
Further, let w= { W 1 ,W 2 …,W j …,W N Is the reference picture I R (x, y) and the image I to be matched M A set of pixel feature pairs between (x, y), where W j =[W jR ,W jM ]，W jR Is image I R Pixel point feature in (x, y), W jM Is the image I to be matched M And (3) corresponding pixel point characteristics in (x, y), wherein N represents the number of pixel point characteristic pairs.
Compared with the prior art, the invention has the following beneficial technical effects:
acquiring a water meter surface glass focusing image, adaptively adjusting chromaticity information of a highlight pixel point area, and constructing a glass surface stain model; according to the method, the influence caused by reflection of the water meter surface glass and dirt is fully considered, iterative learning is carried out through the iterative model, and a clear water meter internal image is obtained after the dirt of the corresponding category is removed, so that the clear water meter internal image is obtained.
Based on an image matching method, detecting the matching difference of clear internal images of the water meter, judging whether the water meter needs to be calibrated according to the matching difference, and ensuring accurate calibration basis and accurate calibration result.
Drawings
In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings that are needed in the description of the embodiments will be briefly described below, it being obvious that the drawings in the following description are only some embodiments of the present invention, and that other drawings may be obtained according to these drawings without inventive effort to a person skilled in the art.
FIG. 1 is a flow chart of a water meter detection method based on image processing according to the present invention;
FIG. 2 is a schematic diagram showing the comparison of images of the inside of a water meter before and after treatment by the present invention;
FIG. 3 is a schematic diagram of the pointer and the plane position of the meter face;
fig. 4 is a schematic diagram of an angle matching method.
Detailed Description
For the purposes of making the objects, technical solutions and advantages of the embodiments of the present application more clear, the technical solutions of the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application, and it is apparent that the described embodiments are some embodiments of the present application, but not all embodiments. All other embodiments, which can be made by one of ordinary skill in the art without undue burden from the present disclosure, are within the scope of the present disclosure.
In the drawings of the specific embodiments of the present invention, in order to better and more clearly describe the working principle of each element in the system, the connection relationship of each part in the device is represented, but only the relative positional relationship between each element is clearly distinguished, and the limitations on the signal transmission direction, connection sequence and the structure size, dimension and shape of each part in the element or structure cannot be constructed.
As shown in fig. 1, a flow chart of a water meter detection method based on image processing according to the present invention includes the following steps:
s1, acquiring a water surface glass focusing image, adaptively adjusting chromaticity information of a highlight pixel point area, and constructing a glass surface stain model.
S11, calculating the brightness difference value of each pixel point in the water surface glass focusing image, and determining the pixel point with the brightness difference value larger than the threshold value as the highlight pixel point.
The luminance difference value is obtained by the following formula:
；
wherein S (p) is the brightness difference value of the pixel point p to be solved, Y p For the gray value of the pixel point p in the brightness channel, Y q For the gray value of pixel q in the luminance channel,and representing the gray scale distance, wherein N is the total pixel point of the RGB image.
And determining the pixel points with the brightness difference value larger than the threshold value as highlight pixel points, and determining the rest pixel points as diffuse reflection points.
S12, the chromaticity information of the highlight pixel point area is adaptively adjusted.
Converting a glass focusing image to be detected into a brightness space, wherein the final brightness correction model expression is as follows:
；
t is the brightness ratio, T max And T min Respectively a maximum value and a minimum value of T;is constant, L in L is the brightness of the original image out To output the brightness of the image, C out For the color ratio, C in For color channels (R, G, B), -for color channels (R, G, B)>Is a color correction factor.
The brightness information is not changed, only the color and the chromatic aberration are processed, and the chromatic information of the highlight pixel point area is self-adaptively adjusted by adjusting the color correction factors.
S13, constructing a stain model of the glass surface according to the self-adaptive adjusted glass focusing image.
And adding a positioning function for detecting the glass focusing image stains in an output layer of the basic target detection neural network structure to form an improved target detection neural network structure.
The stain model of the glass surface is as follows:
；
；
wherein L is o Representing a positioning function of the glass stain model;representing a distance vector between a target detection frame and a smudge frame in the glass focused image; c represents an empirical parameter; />Representing a second order norm; cx (cx) a Representing the abscissa, cy, of the center point of the target detection frame in the glass focus image in the image coordinate system a Representing the ordinate, w, of the center point of the target detection frame in the glass focus image in the image coordinate system a Indicating the width of the target detection frame in the glass focusing image, h a Representing the height of a target detection frame in the glass focusing image; cx (cx) b Representing the abscissa, cy, of the center point of the smudge box B in the glass focused image in the image coordinate system b Representing the ordinate, w, of the center point of the smudge box in the glass focused image in the image coordinate system b Indicating the width of a stained frame in a glass focused image, h b The height of the smudge box in the glass focused image is shown.
And inputting the big data of the stains into an improved target detection neural network structure for training, and obtaining a glass stain model through training.
S2, acquiring a water meter internal focusing image, inputting the water meter internal focusing image and a stain model on the glass surface into the segmentation neural network, and removing background stains to obtain a clear water meter internal image.
The segmentation neural network is an important ring of image processing and image understanding in the machine vision technology, classifies each pixel point in an image, and determines the category of each point (such as belonging to a background, a target and the like), so as to perform region segmentation and background removal.
The split neural network is a full convolution network without a full connection layer, performs pixel-level classification on the image, solves the problem of region split of the image, can accept the image size of any size, adopts deconvolution to process the last feature image so as to restore the size of the input image, generates a prediction for each pixel, simultaneously reserves the spatial information in the original input image, and finally performs pixel-by-pixel classification on the up-sampled feature image.
And extracting the initial characteristic V of the input internal focusing image of the water meter by using the convolutional neural network Sf and expanding the initial characteristic V into a concerned characteristic.
Calculating a feature of interest F:
。
K. v represents the convolutional neural network parameters and initial characteristics, and d represents the normalization processing coefficient performed to prevent the result from becoming excessively large.
Embedding the focus feature and the glass stain model into the segmentation neural network to obtain a stain feature map, and segmenting the stain feature map and the focus feature F to obtain a segmentation result of the stain feature map.
Model stains on glass surfacesEmbedding an iterative model->Iterative learning is performed, and the stain feature map Q after each iteration i is expressed as:
,i=1,2…，n；
where n represents the number of iterations and F represents the feature of interest, and each iteration is updated once, a new class is generated for embedding into the next iteration.
The segmentation result Output of the segmented neural network is represented as follows:
Output=At(Q,V) ；
q, V therein represent stain feature maps and initial features, respectively; at (Q, V) is a segmented neural network model.
The segmentation result Output is to remove stains of corresponding categories to obtain clear internal images of the water meter, and the internal images of the water meter before and after the processing are shown in fig. 2.
And S3, detecting the matching difference of clear internal images of the water meter based on an image matching method, and judging whether the water meter needs to be calibrated according to the matching difference.
Since the pointer and the dial of the water meter are not strictly in the same plane, but are respectively in two parallel planes with smaller distances, as shown in fig. 3, after the perspective transformation is performed to perform image correction, the position of the pointer is shifted to a certain extent, and such errors need to be corrected.
For the two images given: reference image I R (x, y) and clear water meter internal focusing image I to be matched M The purpose of the (x, y) image matching is to find an optimal geometrical transformation parameter matrix T so that the clear water meter internal focusing image I to be matched M (x, y) as much as possible with reference image I R (x, y) are identical.
The image matching model is expressed as:
；
wherein T is opt Representing the matching difference, S represents the reference image I R (x, y) and the image I to be matched M Similarity between (x, y).
Let W= { W 1 ,W 2 …,W j …,W N Is the reference picture I R (x, y) and the image I to be matched M A set of pixel feature pairs between (x, y), where W j =[W jR ,W jM ]，W jR Is image I R Pixel point feature in (x, y), W jM Is the image I to be matched M And (3) corresponding pixel point characteristics in (x, y), wherein N represents the number of pixel point characteristic pairs.
According to the difference T of matching opt Detecting the clear internal focusing image of the water meter according to the corresponding numerical range, judging whether the water meter is required to be calibrated, and calibrating the range, namely, matching the difference T opt The corresponding numerical value interval range.
Compared with the approximate neighbor searching method in the prior art, the instrument image matching method has the advantage of high accuracy, the instrument image matching method adopts a pixel-by-pixel characteristic traversing mode to find out the best matching, and the method has the advantages that the global optimal solution is found out, and the matching accuracy is high.
In a preferred embodiment, an angle matching method is used for detecting the angle matching of the clear internal image of the water meter with the standard same angle image, and whether the water meter needs to be calibrated or not is judged according to the angle matching. The angle matching method is a method for reading and calculating according to the angle of the pointer and the angle of the center of gravity of the minimum scale and the center of gravity of the maximum scale and the center of the circle.
As shown in FIG. 4, point C represents the center of a fitting circle, points A and B represent the center of gravity of the scale of the minimum value and the center of gravity of the scale of the maximum value of water respectively, line l represents the fitting straight line of the pointer, and the included angle between CA and CB isThe included angle between CA and l is->. Let the minimum scale value of the instrument be M A Maximum scale value is M B The instrument reading value is I, and then the calculation formula of the instrument angle is:
。
and judging whether the water meter needs to be calibrated or not according to the angle matching property of the clear water meter internal image and the angle matching property of the standard same angle image.
In the above embodiments, it may be implemented in whole or in part by software, hardware, firmware, or any combination thereof. When implemented in software, may be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions. When loaded and executed on a computer, produces a flow or function in accordance with embodiments of the present application, in whole or in part. The computer may be a general purpose computer, a special purpose computer, a computer network, or other programmable apparatus. The computer instructions may be stored in or transmitted across a computer-readable storage medium. The computer readable storage medium may be any available medium that can be accessed by a computer or a data storage device such as a server, data center, etc. that contains an integration of one or more available media. The usable medium may be a magnetic medium (e.g., a floppy disk, a hard disk, a magnetic tape), an optical medium (e.g., a DVD), or a semiconductor medium (e.g., a Solid State Disk (SSD)), or the like.
While the invention has been described with reference to certain preferred embodiments, it will be understood by those skilled in the art that various changes and substitutions of equivalents may be made and equivalents will be apparent to those skilled in the art without departing from the scope of the invention. Therefore, the protection scope of the present application shall be subject to the protection scope of the claims.

Claims (5)

1. The water meter detection method based on image processing is characterized by comprising the following steps:
s1, acquiring a water surface glass focusing image, adaptively adjusting chromaticity information of a highlight pixel point area, and constructing a glass surface stain model;
s11, calculating the brightness difference value of each pixel point in the focusing image of the water surface glass:
；
wherein S (p) is the brightness difference value of the pixel point p to be solved, Y p For the gray value of the pixel point p in the brightness channel, Y q For the gray value of pixel q in the luminance channel,the gray scale distance is represented, and N is the total pixel point of the RGB image; determining the pixel points with the brightness difference value larger than the threshold value as highlight pixel points, and determining the rest pixel points as diffuse reflection points;
s12, adaptively adjusting chromaticity information of the highlight pixel point area;
converting a glass focusing image to be detected into a brightness space, wherein the brightness correction model expression is as follows:
；
t is the brightness ratio, T max And T min Respectively a maximum value and a minimum value of T;is constant, L in L is the brightness of the original image out To output the brightness of the image, C out For the color ratio, C in For color channels (R, G, B), -for color channels (R, G, B)>Is a color correction factor; the chromaticity information of the highlight pixel point area is adaptively adjusted by adjusting the color correction factors;
s13, constructing a glass surface stain model according to the self-adaptively adjusted water surface glass focusing image:
；
；
wherein L is o Representing a positioning function of the glass stain model;representing a distance vector between a target detection frame and a smudge frame in the glass focused image; c represents an empirical parameter; />Representing a second order norm; cx (cx) a Representing the abscissa, cy, of the center point of the target detection frame in the glass focus image in the image coordinate system a Representing the ordinate, w, of the center point of the target detection frame in the glass focus image in the image coordinate system a Indicating the width of the target detection frame in the glass focusing image, h a Representing the height of a target detection frame in the glass focusing image; cx (cx) b Representing the abscissa, cy, of the center point of the smudge box B in the glass focused image in the image coordinate system b Representing the ordinate, w, of the center point of the smudge box in the glass focused image in the image coordinate system b Indicating the width of a stained frame in a glass focused image, h b Representing the height of a smudge box in the glass focused image;
s2, acquiring a water meter internal focusing image, inputting the water meter internal focusing image and a glass surface dirt model into a segmentation neural network, and removing background dirt to obtain a clear water meter internal image;
and S3, detecting the matching difference of clear internal images of the water meter based on an image matching method, and judging whether the water meter needs to be calibrated according to the matching difference.
2. The image processing-based water meter detection method according to claim 1, wherein in step S2:
extracting an initial characteristic V of an input water meter internal focusing image by using a convolutional neural network Sf, and expanding the initial characteristic V into a concerned characteristic F:
；
k represents a convolutional neural network parameter, and d represents a normalized processing coefficient;
embedding the focus feature and the glass surface stain model into the segmentation neural network, obtaining the stain feature mapping, and realizing the segmentation of the stain feature mapping and the focus feature to obtain the segmentation result of the stain removal feature mapping.
3. The method for detecting a water meter based on image processing according to claim 2, wherein,
model glass surface stainsEmbedding an iterative model->Iterative learning is carried out, and the stain feature mapping Q after the ith iteration is as follows:
,i=1,2…，n；
wherein n represents the iteration times, F represents the concerned feature, and each iteration is updated once, so that a new stain feature map is generated and then embedded into the next iteration;
the segmentation result Output of the segmented neural network is as follows:
Output=At(Q,V)；
q, V therein represent stain feature map and initial features, respectively; at (Q, V) is a split neural network model, and the split result Output represents a clear water meter internal image obtained after removing stains of corresponding categories.
4. The image processing-based water meter detection method according to claim 1, wherein in step S3: let the reference picture be I R (x, y), the clear internal image of the water meter to be matched is I M (x, y), the image matching model is expressed as:
；
wherein T is opt Representing the matching difference, S represents the reference image I R (x, y) clear internal image I of water meter to be matched M (x, y) similarity between;
according to the difference T of matching opt And detecting a clear internal focusing image of the water meter according to the corresponding numerical value interval range, and judging whether the water meter is required to be calibrated.
5. The method for detecting a water meter based on image processing according to claim 4, wherein w= { W is set 1 ,W 2 …,W j …,W N Is the reference picture I R (x, y) and the image I to be matched M A set of pixel feature pairs between (x, y), where W j =[W jR ,W jM ]，W jR Is a reference image I R Pixel point feature in (x, y), W jM Is a clear water meter internal image I to be matched M And (3) corresponding pixel point characteristics in (x, y), wherein N represents the number of pixel point characteristic pairs.
CN202311376869.1A 2023-10-24 2023-10-24 Water meter detection method based on image processing Active CN117115488B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202311376869.1A CN117115488B (en) 2023-10-24 2023-10-24 Water meter detection method based on image processing

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202311376869.1A CN117115488B (en) 2023-10-24 2023-10-24 Water meter detection method based on image processing

Publications (2)

Publication Number Publication Date
CN117115488A CN117115488A (en) 2023-11-24
CN117115488B true CN117115488B (en) 2024-02-06

Family

ID=88796973

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202311376869.1A Active CN117115488B (en) 2023-10-24 2023-10-24 Water meter detection method based on image processing

Country Status (1)

Country Link
CN (1) CN117115488B (en)

Citations (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN106802171A (en) * 2017-03-22 2017-06-06 淄博贝林电子有限公司 Shooting direct-reading water meter and automatic pollution removing method with automatic pollution removing function
CN112710362A (en) * 2020-12-16 2021-04-27 杭州欧裴机械科技有限公司 Intelligent water meter cleaning device for solving problem of difficult viewing caused by stain adhesion
CN113567464A (en) * 2021-06-16 2021-10-29 美晟通科技(苏州)有限公司 Transparent medium stain position detection method and device
CN114359538A (en) * 2022-01-04 2022-04-15 重庆邮电大学 Water meter reading positioning and identifying method
CN115791801A (en) * 2022-12-08 2023-03-14 中科(厦门)数据智能研究院 3D glass on-line monitoring platform based on machine vision
CN116343228A (en) * 2023-03-27 2023-06-27 上海第二工业大学 Intelligent reading method and system for water meter
CN116609349A (en) * 2023-05-22 2023-08-18 青岛融合光电科技有限公司 Carrier plate glass foreign matter detection equipment and detection method

Family Cites Families (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2013132042A (en) * 2011-11-25 2013-07-04 Ricoh Co Ltd Image inspection device, image forming apparatus, image inspection method and program

Patent Citations (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN106802171A (en) * 2017-03-22 2017-06-06 淄博贝林电子有限公司 Shooting direct-reading water meter and automatic pollution removing method with automatic pollution removing function
CN112710362A (en) * 2020-12-16 2021-04-27 杭州欧裴机械科技有限公司 Intelligent water meter cleaning device for solving problem of difficult viewing caused by stain adhesion
CN113567464A (en) * 2021-06-16 2021-10-29 美晟通科技(苏州)有限公司 Transparent medium stain position detection method and device
CN114359538A (en) * 2022-01-04 2022-04-15 重庆邮电大学 Water meter reading positioning and identifying method
CN115791801A (en) * 2022-12-08 2023-03-14 中科(厦门)数据智能研究院 3D glass on-line monitoring platform based on machine vision
CN116343228A (en) * 2023-03-27 2023-06-27 上海第二工业大学 Intelligent reading method and system for water meter
CN116609349A (en) * 2023-05-22 2023-08-18 青岛融合光电科技有限公司 Carrier plate glass foreign matter detection equipment and detection method

Also Published As

Publication number Publication date
CN117115488A (en) 2023-11-24

Similar Documents

Publication Publication Date Title
CN110689579B (en) Rapid monocular vision pose measurement method and measurement system based on cooperative target
CN106875381B (en) Mobile phone shell defect detection method based on deep learning
CN112257676B (en) Pointer type instrument reading method and system and inspection robot
CN112818988B (en) Automatic identification reading method and system for pointer instrument
CN109801333B (en) Volume measurement method, device and system and computing equipment
CN110675376A (en) PCB defect detection method based on template matching
CN111368906B (en) Pointer type oil level meter reading identification method based on deep learning
CN110223355B (en) Feature mark point matching method based on dual epipolar constraint
CN111523547B (en) 3D semantic segmentation method and terminal
CN116228780B (en) Silicon wafer defect detection method and system based on computer vision
CN110634137A (en) Bridge deformation monitoring method, device and equipment based on visual perception
CN111239684A (en) Binocular fast distance measurement method based on YoloV3 deep learning
CN112489140A (en) Attitude measurement method
CN115082450A (en) Pavement crack detection method and system based on deep learning network
CN112929626A (en) Three-dimensional information extraction method based on smartphone image
CN114005108A (en) Pointer instrument degree identification method based on coordinate transformation
CN114782451A (en) Workpiece defect detection method and device, electronic equipment and readable storage medium
CN117434568A (en) Intelligent positioning system based on remote sensing satellite
CN117115488B (en) Water meter detection method based on image processing
CN114926417B (en) Microscopic imaging detection method and system for weak scratches on surface of high-density flexible FPC
CN116630662A (en) Feature point mismatching eliminating method applied to visual SLAM
CN116597016A (en) Optical fiber endoscope image calibration method
CN116819561A (en) Point cloud data matching method, system, electronic equipment and storage medium
CN115512343A (en) Method for correcting and recognizing reading of circular pointer instrument
CN115546716A (en) Binocular vision-based method for positioning fire source around power transmission line

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