CN109508709A - A kind of simple pointer meter reading method based on machine vision - Google Patents
A kind of simple pointer meter reading method based on machine vision Download PDFInfo
- Publication number
- CN109508709A CN109508709A CN201811152252.0A CN201811152252A CN109508709A CN 109508709 A CN109508709 A CN 109508709A CN 201811152252 A CN201811152252 A CN 201811152252A CN 109508709 A CN109508709 A CN 109508709A
- Authority
- CN
- China
- Prior art keywords
- meter
- pointer
- angle
- prototype drawing
- feature
- 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.)
- Granted
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
- G06V10/20—Image preprocessing
- G06V10/22—Image preprocessing by selection of a specific region containing or referencing a pattern; Locating or processing of specific regions to guide the detection or recognition
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/40—Extraction of image or video features
- G06V10/44—Local feature extraction by analysis of parts of the pattern, e.g. by detecting edges, contours, loops, corners, strokes or intersections; Connectivity analysis, e.g. of connected components
-
- 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
Abstract
The simple pointer meter reading method based on machine vision that the invention discloses a kind of, it include: acquisition module figure, Prototype drawing is modeled, feature detection is carried out to the meter region of Prototype drawing, the meter of different scales is classified, then one two dimensional code of of a sort generation saves modeling information, characteristic information and two-dimensional barcode information;It treats identification Instrument image and carries out feature detection, complete the Feature Points Matching between Instrument image and Prototype drawing to be identified using characteristic matching technology, obtain the positioning in meter region in meter diagram to be identified, and be Prototype drawing state by meter regional calibration to be identified;Determine pointer angle;The two dimensional code on meter is read, the distribution of meter scale is obtained;Meter reading is determined using preset angle configuration.The advantages of method of the invention has real-time high, and calculating speed is fast, adapts to various different scale distribution meters.
Description
Technical field
The present invention relates to digital image processing techniques, and in particular to a kind of simple pointer meter reading side based on machine vision
Method.
Background technique
In power industry, most of position is equipped with pointer instrument.Enter to read using traditional artificial number reading method
When taking, it may appear that the inefficiency and instrument installation site having artificially measures difficulty, artificial reading can not be carried out directly, with meter
The development of calculation machine and the rise of artificial intelligence, more and more the meter reading method based on image procossing is suggested.
Patent CN201611055991.9 proposes a kind of method of meter reading automatic identification, and this method is made with front view
For Prototype drawing, complete to protect the center location of the angle instrument of the minimax graduation mark of instrument in Prototype drawing using straight-line detection
There are in Prototype drawing, meter positioning and calibration then are completed using the information in SIFT feature matching and Prototype drawing, is finally utilized
Hough transformation detects pointer straight line, completes to read using preset angle configuration.The SIFT algorithm characteristics that this method is used in characteristic matching
Detection speed is slow, has lower rotation robustness, Scale invariant robustness and real-time, can not fit under the conditions of some inspections
With.And every class meter may have the non-uniform problem of scale in substation inspection, which does not simultaneously solve this problem.
Patent CN201711168146.7 proposes a kind of simple pointer meter reading method based on machine vision, this method
After acquisition module figure, the detection of SURF feature is carried out to the target area of Prototype drawing, saves template characteristic information.Acquisition to
After mapping, treats mapping and carry out the detection of SURF feature, then matched, oriented to mapping with the characteristic information of Prototype drawing
In target area, and the state of Prototype drawing will be calibrated to using perspective transformation matrix to mapping.The feature of this method detects
SURF algorithm is used, which has obtained very big mention relative to SIFT algorithm Scale invariant robustness and rotation robustness
Rise, but taken a long time on calculating the time, real-time it is lower, using effect is less desirable under certain outdoor bad borders.
In consideration of it, the prior art could be improved.
Summary of the invention
In view of the deficiencies of the prior art, the present invention provides a kind of simple pointer meter reading method based on machine vision, energy
Enough to overcome the problems, such as that image tilts when detecting meter, the calculating time is extremely short, and real-time is high, and distributes for different scales
Meter reading can be also read from pointer angle offset.
To achieve the above object, the technical solution of the present invention is as follows:
A kind of simple pointer meter reading method based on machine vision, includes the following steps:
Step 1: acquisition module figure: shooting the front view of instrument as Prototype drawing, Prototype drawing is modeled, to Prototype drawing
Meter region carry out feature detection, the meter of different scales is classified, then one two dimensional code of of a sort generation is protected
Deposit modeling information, characteristic information and two-dimensional barcode information;
Step 2: treating identification Instrument image and carry out feature detection, complete Instrument image to be identified using characteristic matching technology
Feature Points Matching between Prototype drawing, obtains the positioning in meter region in meter diagram to be identified, and by meter to be identified
Regional calibration is Prototype drawing state;
Step 3: determining pointer angle: gray processing and filtering processing are carried out to the meter region after calibration;When extracting modeling
The calibration points of selection and cursor backward center;Simplify meter information, making it includes straight line information as much as possible;Utilize Hough
Transform method carries out straight-line detection and obtains pointer angle;
Step 4: reading the two dimensional code on meter, obtain the distribution of meter scale;
Step 5: determining meter reading using preset angle configuration.
Wherein, the detection of feature described in step 1 and step 2 is detected using improved ORB feature, is specifically included:
(a) feature point extraction: the following steps are included: 1. just extract to meter characteristic point;2. the method for machine learning
Filter out optimal characteristics point in meter;3. non-maxima suppression removes the characteristic point of local comparatively dense;4. the scale of characteristic point is not
Deformation;5. the invariable rotary shape of characteristic point;
(b) it calculates Feature Descriptor: after obtaining characteristic point, description of each characteristic point is calculated using BRIEF algorithm;
(c) Feature Points Matching.
Wherein, step 5 specifically includes: calculating each calibration points and the cursor backward line of centres and positive direction of the x-axis first
Then angle establishes the relational expression of its angle and positive direction of the x-axis;Find pointer straight line to be measured and determine pointer direction, find out to
Then the angle for surveying pointer straight line and positive direction of the x-axis brings angle and reading relational expression into, completes meter reading.
Number reading method of the invention has real-time high, and calculating speed is fast, adapts to various different scale distribution meters
Advantage.
Detailed description of the invention
Fig. 1 is flow chart of the invention.
Specific embodiment
In order to make the foregoing objectives, features and advantages of the present invention clearer and more comprehensible, with reference to the accompanying drawing and specific real
Applying mode, the present invention is described in further detail.
As shown in Figure 1, a kind of simple pointer meter reading method based on machine vision, includes the following steps:
Step 1: acquisition module figure: shooting the front view of instrument as Prototype drawing, model to Prototype drawing, and modeling is specific
Including the centre of gyration for choosing meter region, choosing pointer area, choosing calibration points and instrument;To the meter region of Prototype drawing into
Row feature detection, the meter of different scales is classified, one two dimensional code of of a sort generations, then preservation modeling information,
Characteristic information and two-dimensional barcode information.
Step 2: treating identification Instrument image and carry out feature detection, complete Instrument image to be identified using characteristic matching technology
Feature Points Matching between Prototype drawing, obtains the positioning in meter region in meter diagram to be identified, and by meter to be identified
Regional calibration is Prototype drawing state.
Wherein, feature detection is detected using improved ORB feature, is specifically included:
(a) feature point extraction: the following steps are included: 1. just extract to meter characteristic point;2. the method for machine learning
Filter out optimal characteristics point in meter;3. non-maxima suppression removes the characteristic point of local comparatively dense;4. the scale of characteristic point is not
Deformation;5. the invariable rotary shape of characteristic point;
(b) it calculates Feature Descriptor: after obtaining characteristic point, description of each characteristic point is calculated using BRIEF algorithm;
(c) Feature Points Matching.
Step 3: determine pointer angle: including:
(a) meter after calibration is pre-processed, pretreatment includes gray processing and filtering processing, wherein gray processing
Effect is simplified image data information, and gaussian filtering effect is removal picture noise information;
(b) center of circle is demarcated: extracting the calibration points chosen when modeling and cursor backward center, the step can simplify subsequent line
The process of detection shortens the extraction of straight line time;
(c) Information Simplification: before extracting feature, need to carry out meter Information Simplification processing, including straight line as much as possible
Information removes irrelevant information;
(d) straight-line detection is carried out using Hough transform method obtain pointer angle.
Step 4: reading the two dimensional code on meter, show that meter scale is distributed by the meter scale classification in step 1;
Step 5: determining meter reading, specific steps using preset angle configuration are as follows: calculate each calibration points and cursor backward first
Then the angle of the line of centres and positive direction of the x-axis establishes the relational expression of its angle and positive direction of the x-axis;Find pointer straight line to be measured
And determine pointer direction;Then the angle for finding out pointer straight line and positive direction of the x-axis to be measured brings angle and reading relational expression into, complete
At meter reading.
Wherein, feature detection techniques, characteristic matching technology, the detection of ORB feature, machine learning, BRIEF algorithm and Hough
The specific implementation of transformation is the prior art.
Simply to illustrate that technical concepts and features of the invention, its purpose is allows in the art above-described embodiment
Those of ordinary skill cans understand the content of the present invention and implement it accordingly, and it is not intended to limit the scope of the present invention.It is all
It is the equivalent changes or modifications that the essence of content according to the present invention is made, should be covered by the scope of protection of the present invention.
Claims (3)
1. a kind of simple pointer meter reading method based on machine vision, includes the following steps:
Step 1: acquisition module figure: shooting the front view of instrument as Prototype drawing, Prototype drawing is modeled, to the table of Prototype drawing
It counts region and carries out feature detection, the meter of different scales is classified, then one two dimensional code of of a sort generation is saved and built
Mould information, characteristic information and two-dimensional barcode information;
Step 2: treating identification Instrument image and carry out feature detection, complete Instrument image and mould to be identified using characteristic matching technology
Feature Points Matching between plate figure, obtains the positioning in meter region in meter diagram to be identified, and by meter region to be identified
It is calibrated to Prototype drawing state;
Step 3: determining pointer angle: gray processing and filtering processing are carried out to the meter region after calibration;It is chosen when extracting modeling
Calibration points and cursor backward center;Simplify meter information, making it includes straight line information as much as possible;Utilize Hough transform
Method carries out straight-line detection and obtains pointer angle;
Step 4: reading the two dimensional code on meter, obtain the distribution of meter scale;
Step 5: determining meter reading using preset angle configuration.
2. a kind of simple pointer meter reading method based on machine vision according to claim 1, it is characterised in that: step
1 and step 2 described in feature detection using improved ORB feature detect, specifically include:
(a) feature point extraction: the following steps are included: 1. just extract to meter characteristic point;2. the method for machine learning is screened
Optimal characteristics point in meter out;3. non-maxima suppression removes the characteristic point of local comparatively dense;4. the Scale invariant shape of characteristic point;
5. the invariable rotary shape of characteristic point;
(b) it calculates Feature Descriptor: after obtaining characteristic point, description of each characteristic point is calculated using BRIEF algorithm;
(c) Feature Points Matching.
3. a kind of simple pointer meter reading method based on machine vision according to claim 1, it is characterised in that: step
5 specifically include: calculating the angle of each calibration points Yu the cursor backward line of centres and positive direction of the x-axis first, then establish its angle
The relational expression of degree and positive direction of the x-axis;It finds pointer straight line to be measured and determines pointer direction, finding out pointer straight line to be measured and x-axis just
Then the angle in direction brings angle and reading relational expression into, completes meter reading.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201811152252.0A CN109508709B (en) | 2018-09-29 | 2018-09-29 | Single pointer instrument reading method based on machine vision |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201811152252.0A CN109508709B (en) | 2018-09-29 | 2018-09-29 | Single pointer instrument reading method based on machine vision |
Publications (2)
Publication Number | Publication Date |
---|---|
CN109508709A true CN109508709A (en) | 2019-03-22 |
CN109508709B CN109508709B (en) | 2021-09-21 |
Family
ID=65746271
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201811152252.0A Active CN109508709B (en) | 2018-09-29 | 2018-09-29 | Single pointer instrument reading method based on machine vision |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN109508709B (en) |
Cited By (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN112818973A (en) * | 2021-01-26 | 2021-05-18 | 浙江国自机器人技术股份有限公司 | Positioning and reading rechecking method for meter identification |
CN113313122A (en) * | 2021-06-11 | 2021-08-27 | 曲阜师范大学 | Pointer type instrument automatic reading identification method based on computer vision |
CN113537197A (en) * | 2021-01-26 | 2021-10-22 | 浙江国自机器人技术股份有限公司 | Meter automatic modeling method based on machine vision |
CN114323106A (en) * | 2021-12-24 | 2022-04-12 | 航天科技控股集团股份有限公司 | Automatic automobile instrument pointer vision calibration system capable of sliding left and right |
CN115171091A (en) * | 2022-09-06 | 2022-10-11 | 浩鲸云计算科技股份有限公司 | Meter identification method for substation inspection |
Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN104992481A (en) * | 2015-07-13 | 2015-10-21 | 广东电网有限责任公司惠州供电局 | Metered inspection system and method based on image recognition technology |
CN106017705A (en) * | 2016-05-11 | 2016-10-12 | 红旗仪表(长兴)有限公司 | Marking device of pointer type thermometer dial plate and generating method for dial plate |
CN106570436A (en) * | 2016-11-14 | 2017-04-19 | 三峡大学 | Substation equipment meter reading system and meter reading method based on two-dimensional code |
CN106778823A (en) * | 2016-11-22 | 2017-05-31 | 国网湖北省电力公司宜昌供电公司 | A kind of readings of pointer type meters automatic identifying method |
CN107507174A (en) * | 2017-08-16 | 2017-12-22 | 杭州意能电力技术有限公司 | Power plant's instrument equipment drawing based on hand-held intelligent inspection is as recognition methods and system |
US20180253619A1 (en) * | 2017-03-06 | 2018-09-06 | Intelligent Security Systems Corporation | Systems and methods for evaluating readings of gauge dials |
-
2018
- 2018-09-29 CN CN201811152252.0A patent/CN109508709B/en active Active
Patent Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN104992481A (en) * | 2015-07-13 | 2015-10-21 | 广东电网有限责任公司惠州供电局 | Metered inspection system and method based on image recognition technology |
CN106017705A (en) * | 2016-05-11 | 2016-10-12 | 红旗仪表(长兴)有限公司 | Marking device of pointer type thermometer dial plate and generating method for dial plate |
CN106570436A (en) * | 2016-11-14 | 2017-04-19 | 三峡大学 | Substation equipment meter reading system and meter reading method based on two-dimensional code |
CN106778823A (en) * | 2016-11-22 | 2017-05-31 | 国网湖北省电力公司宜昌供电公司 | A kind of readings of pointer type meters automatic identifying method |
US20180253619A1 (en) * | 2017-03-06 | 2018-09-06 | Intelligent Security Systems Corporation | Systems and methods for evaluating readings of gauge dials |
CN107507174A (en) * | 2017-08-16 | 2017-12-22 | 杭州意能电力技术有限公司 | Power plant's instrument equipment drawing based on hand-held intelligent inspection is as recognition methods and system |
Non-Patent Citations (1)
Title |
---|
柳杨: "《数字图像物体识别理论详解与实战》", 31 March 2018 * |
Cited By (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN112818973A (en) * | 2021-01-26 | 2021-05-18 | 浙江国自机器人技术股份有限公司 | Positioning and reading rechecking method for meter identification |
CN113537197A (en) * | 2021-01-26 | 2021-10-22 | 浙江国自机器人技术股份有限公司 | Meter automatic modeling method based on machine vision |
CN113313122A (en) * | 2021-06-11 | 2021-08-27 | 曲阜师范大学 | Pointer type instrument automatic reading identification method based on computer vision |
CN114323106A (en) * | 2021-12-24 | 2022-04-12 | 航天科技控股集团股份有限公司 | Automatic automobile instrument pointer vision calibration system capable of sliding left and right |
CN115171091A (en) * | 2022-09-06 | 2022-10-11 | 浩鲸云计算科技股份有限公司 | Meter identification method for substation inspection |
Also Published As
Publication number | Publication date |
---|---|
CN109508709B (en) | 2021-09-21 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN109508709A (en) | A kind of simple pointer meter reading method based on machine vision | |
CN109029381B (en) | Tunnel crack detection method and system and terminal equipment | |
CN107808378B (en) | Method for detecting potential defects of complex-structure casting based on vertical longitudinal and transverse line profile features | |
CN109100741A (en) | A kind of object detection method based on 3D laser radar and image data | |
CN106824806B (en) | The detection method of low module plastic gear based on machine vision | |
CN105447512A (en) | Coarse-fine optical surface defect detection method and coarse-fine optical surface defect detection device | |
CN107578400A (en) | A kind of contact net device parameter detection method of BIM and three-dimensional point cloud fusion | |
CN105930852B (en) | A kind of bubble image-recognizing method | |
CN105865344A (en) | Workpiece dimension measuring method and device based on machine vision | |
CN106407924A (en) | Binocular road identifying and detecting method based on pavement characteristics | |
CN108710909B (en) | Counting method for deformable, rotary and invariant boxed objects | |
CN106340010B (en) | A kind of angular-point detection method based on second order profile difference | |
CN103177444A (en) | Automatic SAR (synthetic-aperture radar) image rectification method | |
CN103295232A (en) | SAR (specific absorption rate) image registration method based on straight lines and area | |
CN106127205A (en) | A kind of recognition methods of the digital instrument image being applicable to indoor track machine people | |
CN103593849A (en) | Method for quickly recognizing and tracking image sequence oval artificial target points | |
CN116168028B (en) | High-speed rail original image processing method and system based on edge filtering under low visibility | |
CN105069395A (en) | Road marking automatic identification method based on terrestrial three-dimensional laser scanning technology | |
Princen et al. | A comparison of Hough transform methods | |
CN113705564B (en) | Pointer type instrument identification reading method | |
CN115597494B (en) | Precision detection method and system for prefabricated part preformed hole based on point cloud | |
CN110765993B (en) | SEM graph measuring method based on AI algorithm | |
CN116152532A (en) | Remote sensing image feature extraction and matching method and device and electronic equipment | |
Wang et al. | Improved morphological band-pass filtering algorithm and its application in circle detection | |
CN108447045A (en) | A kind of SAR remote sensing images waters detection method based on SAT integrograms |
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 |