CN108009541A - A kind of intelligent substation device panel automatic identifying method and system - Google Patents
A kind of intelligent substation device panel automatic identifying method and system Download PDFInfo
- Publication number
- CN108009541A CN108009541A CN201610935224.0A CN201610935224A CN108009541A CN 108009541 A CN108009541 A CN 108009541A CN 201610935224 A CN201610935224 A CN 201610935224A CN 108009541 A CN108009541 A CN 108009541A
- Authority
- CN
- China
- Prior art keywords
- mrow
- msub
- panel
- images
- recognized
- 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
- G06V20/00—Scenes; Scene-specific elements
- G06V20/60—Type of objects
- G06V20/62—Text, e.g. of license plates, overlay texts or captions on TV images
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/241—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches
-
- 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
- G06V10/28—Quantising the image, e.g. histogram thresholding for discrimination between background and foreground patterns
-
- 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
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y04—INFORMATION OR COMMUNICATION TECHNOLOGIES HAVING AN IMPACT ON OTHER TECHNOLOGY AREAS
- Y04S—SYSTEMS INTEGRATING TECHNOLOGIES RELATED TO POWER NETWORK OPERATION, COMMUNICATION OR INFORMATION TECHNOLOGIES FOR IMPROVING THE ELECTRICAL POWER GENERATION, TRANSMISSION, DISTRIBUTION, MANAGEMENT OR USAGE, i.e. SMART GRIDS
- Y04S10/00—Systems supporting electrical power generation, transmission or distribution
- Y04S10/50—Systems or methods supporting the power network operation or management, involving a certain degree of interaction with the load-side end user applications
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Multimedia (AREA)
- Data Mining & Analysis (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Life Sciences & Earth Sciences (AREA)
- Artificial Intelligence (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Evolutionary Biology (AREA)
- Evolutionary Computation (AREA)
- General Engineering & Computer Science (AREA)
- Image Analysis (AREA)
Abstract
The present invention relates to a kind of intelligent substation device panel automatic identifying method and system, this method to comprise the following steps：(1) acquisition of intelligent substation device panel image information；(2) panel image information pre-processing；(3) panels feature is extracted and selected；(4) panel element and layout grader are designed；(5) the unrelated panel element in visual angle and layout categorised decision.The present invention can automatically identify device panel fundamental and placement rule, so as to reach the automatic test of device panel layout.
Description
Technical field
The present invention relates to a kind of substation digitlization measure and control device to unify the normative measuring technology of panel identification, and in particular to
A kind of intelligent substation device panel automatic identifying method and system.
Background technology
For a long time automation equipment there are production firm it is numerous the problem of, different manufacturers device appearance size, interface connect
Line is widely different, and design, installation, maintenance mode are different, and operation management difficulty is larger.Observing and controlling is digitized according to unified standard
The job requirement of device panel, different types of device should be required according to standard GB/T 19520, be needed by device type and application
Ask, for different type product specification panel fundamental and placement rule, the high width of unified device cabinet and installation interface ruler
It is very little, realize that equipment style liquid crystal panel is laid out, operation button layout is relatively uniform.
Existing panel layout test is carried out by artificial observation and manual metric form, and work efficiency is low.
The content of the invention
To solve above-mentioned deficiency of the prior art, the object of the present invention is to provide a kind of intelligent substation device panel certainly
Dynamic recognition methods and system, the present invention can automatically identify device panel fundamental and placement rule, so as to reach device surface
The automatic test of plate layout.
The purpose of the present invention is what is realized using following technical proposals：
The present invention provides a kind of intelligent substation device panel automatic identifying method, it is improved in that the method
Comprise the following steps：
(1) acquisition of intelligent substation device panel image information；
(2) panel image information pre-processing；
(3) panels feature is extracted and selected；
(4) panel element and layout grader are designed；
(5) the unrelated panel element in visual angle and layout categorised decision.
Further, in the step (1), taken pictures by the camera of hand-held terminal device and obtain intelligent substation device
Panel image information two-dimensional image information.
Further, in the step (2), counter plate image information carries out pretreatment, including A D conversion, at binaryzation
Reason, the smooth of image, conversion, enhancing, recovery and filtering process；Panel layout recognition methods based on different visual angles reaches to not
With the support of visual angle panel identification；According to perspective projection principle, panel elements target template image T0 is subjected to affine transformation, is obtained
The possible form T of target board under different visual angles is obtained, and then index plane plate element target is generated into the template under different visual angles；
And then in identification process, be utilized respectively different affine transformation templates and matched with images to be recognized I regional areas R, know
Other panel layout situation.
Further, in the step (3), for panel and keypad, display screen, nameplate and indicator light element, chain type is calculated
Gradient Features simultaneously carry out feature extraction；
Wherein, the possible form T and images to be recognized I of target board under different visual angles is decomposed into step-length 4, overlapping 6
× 6 fritters, according to set spatial order construction face plate element target template and the image block sequence of images to be recognized；
For each fritter, zoning pixel differential, the corresponding differential direction of each pixel obtain as follows respectively：
Coordinate position (x, y) corresponds to picture in the fritter that target board possible form T and images to be recognized I is decomposed under different visual angles
Plain I (x, y) differential Gx(x,y)、Gy(x, y) is tried to achieve by convolution algorithm：
Gx(x, y)=[- 10 1] * I (x, y)
Gy(x, y)=[- 10 1]T*I(x,y)
It is as follows with direction θ (x, y) that amplitude G (x, y) is calculated respectively：
Wherein：X, y represent that target board possible form T and images to be recognized I under different visual angles is decomposed respectively
Coordinate position in fritter, I (x, y) represent pixel, Gx(x,y)、Gy(x, y) represents the differential of pixel I (x, y) respectively；θ (x, y) table
Show the corresponding differential directions of pixel I (x, y)；
The subject pixel differential of each fritter is further obtained, and discrete to turn to adjacent difference be 30 ° of 12 directions；
Binaryzation is carried out according to each piece of discretization direction of acquisition as a result, constructing chained list according to set spatial order,
The chain type Gradient Features expression of the possible form T of target board under different visual angles and images to be recognized I regional areas R is established respectively
List (T, R) and List (I, R).
Further, in the step (4), design panel and keypad, display screen, nameplate and the classification for indicating light panel element
Device, i.e., express similarity measurements flow function by designing chain type Gradient Features, calculates the possible form T of target board under different visual angles
With the chain type Gradient Features similarity measure of images to be recognized I regional areas R, the classification of the corresponding T for obtaining maximum similarity is made
To identify the result of panel element.
Further, the possible form T and images to be recognized I of target board under different visual angles is decomposed into step-length 4, overlapping
6 × 6 fritters, according to set spatial order construction face plate element target template and the image block sequence of images to be recognized；
For each fritter, zoning pixel differential, the corresponding differential direction of each pixel obtain as follows respectively：
Coordinate position (x, y) corresponds to picture in the fritter that target board possible form T and images to be recognized I is decomposed under different visual angles
Plain I (x, y) differential Gx(x,y)、Gy(x, y) is tried to achieve by convolution algorithm：
Gx(x, y)=[- 10 1] * I (x, y)
Gy(x, y)=[- 10 1]T*I(x,y)
It is as follows with direction θ (x, y) that amplitude G (x, y) is calculated respectively：
Wherein：X, y represent that target board possible form T and images to be recognized I under different visual angles is decomposed respectively
Coordinate position in fritter, I (x, y) represent pixel, Gx(x,y)、Gy(x, y) represents the differential of pixel I (x, y) respectively；θ (x, y) table
Show the corresponding differential directions of pixel I (x, y)；
The subject pixel differential of each fritter is further obtained, and discrete to turn to adjacent difference be 30 ° of 12 directions；
Binaryzation is carried out according to each piece of discretization direction of acquisition as a result, constructing chained list according to set spatial order,
Chain type Gradient Features expression List (T, R) and the List (I, R) of T and I regional areas R is established respectively.
Similarity measurements flow function is built, for all possible regional area R in I, calculates target board under different visual angles
The difference that possible form T, the chain type Gradient Features of images to be recognized I regional areas R are expressed, and then obtain images to be recognized I
In, the position of target template；
Gradual identification method is taken, i.e., for the possible form T of target board under different visual angles, images to be recognized I
Regional area R, only considers that finite part image block rather than all images block are scanned first, special using chain type gradient is simplified
Expression way is levied, quickly ignores impossible position in images to be recognized I, obtains possible position candidate；It is and then progressive
Increase image number of blocks, improve the description precision of chain type Gradient Features expression, until obtaining the position of target template.
Further, divide an image into after fritter, the corresponding differential directions of each pixel I (x, y) in fritter：
Chain type Gradient Features expression similarity measurements flow function is used for calculating images to be recognized I regional areas R and different visual angles
The number of the similar block of differential characteristics, its form of Definition are between the lower possible form T of target board：
Wherein,
Ori (I, c+r) ∈ List (T, R), ori (T, r) ∈ List (T, R), ori (I, c+r) are that images to be recognized I is in place
The discrete main differential direction that c+r just corresponds to fritter is put, ori (T, r) is the possible form T of target board under different visual angles in position
R corresponds to the discrete main differential direction of fritter；C is the center point coordinate of the possible form T of target board under different visual angles, and I is represented
Images to be recognized；δ represents differential characteristics.
Further, to obtain the non-sensitive type that chain type Gradient Features express similarity measurement function pair miniature deformation, into
One step defines metric function ε2Form is：
De (T, r)={ ori (T, l):l∈maxmagk(r) ∧ mag (t, l) ＞ τ }
Wherein, De (T, r) represents the set in T differential directions of maximum intensity in each zonule r；Ori (T, l) is T
Differential direction at pixel l, is worth for mag (t, l), and τ is threshold value, maxmagk(r) k maximum differential amplitude in block r is represented
Position；
In order to make chain type Gradient Features expression similarity measurements flow function obtain the insensitivity for overall small translation, repair
It is following ε to change similarity measurements flow function3Form：
Wherein：W (T, m) defines the operation of the possible form T two-dimension translationals translation m of target board under different visual angles, and m is
Two-dimension translational converts, and M is the conversion displacement set of two-dimension translational conversion.
Further, in the step (5), the panel elemental map under panel element different visual angles is obtained by affine transformation
The possibility form T of picture, and then calculate the dominant gradient of each fritter r in the possibility form T of the panel element image under different visual angles
Direction, chained list is constructed according to set spatial order, obtains the chain type Gradient Features expression of plate element；The counter plate in feature space
Element object is classified, and for the possibility form T of the different visual angles hypograph of different panels element, is calculated it respectively and is known with waiting
The chain type Gradient Features similarity metric function value ε of other image I regional areas R3(I, T, c), calculates：
It is corresponding using C (T1) as classification as a result, wherein C (T1) is the classification of the corresponding panel elements of T1
The coordinate center of images to be recognized I regional areas R, as the position for identifying panel element；For each panel member identified
The set of position of the element in images to be recognized I, as the panel layout classification results being calculated.
The present invention also provides a kind of intelligent substation device panel automatic recognition system, it is improved in that the system
System is used for intelligent substation device panel automatic identification, and the system is intelligent substation device panel detection handheld terminal,
What the handheld terminal included being connected with intelligent substation device panel is used to obtain intelligent substation device panel image information
The camera of two-dimensional image information, panel image information pre-processing module, panels feature extracts and selecting module, panel element and
It is laid out grader and panel layout categorised decision module.
In order to which some aspects of the embodiment to disclosure have a basic understanding, simple summary shown below is.Should
Summarized section is not extensive overview, nor to determine key/critical component or describe the protection domain of these embodiments.
Its sole purpose is that some concepts are presented with simple form, in this, as the preamble of following detailed description.
Compared with the immediate prior art, the excellent effect that technical solution provided by the invention has is：
Based on this automatic identifying method, panel layout test job switchs to automatically working by manual work, so that significantly
Improve panel layout test job efficiency.
For above-mentioned and relevant purpose, one or more embodiments include will be explained in below and in claim
In the feature that particularly points out.Following explanation and attached drawing describe some illustrative aspects in detail, and its instruction is only
Some modes in the utilizable various modes of principle of each embodiment.Other benefits and novel features will be under
The detailed description in face is considered in conjunction with the accompanying and becomes obvious, the disclosed embodiments be will include all these aspects and they
Be equal.
Brief description of the drawings
Fig. 1 is the structure diagram of intelligent substation device panel automatic identifying method provided by the invention.
Embodiment
The embodiment of the present invention is described in further detail below in conjunction with the accompanying drawings.
The following description and drawings fully show specific embodiments of the present invention, to enable those skilled in the art to
Put into practice them.Other embodiments can include structure, logic, it is electric, process and other change.Embodiment
Only represent possible change.Unless explicitly requested, otherwise single component and function are optional, and the order operated can be with
Change.The part of some embodiments and feature can be included in or replace part and the feature of other embodiments.This hair
The scope of bright embodiment includes the gamut of claims, and claims is all obtainable equivalent
Thing.Herein, these embodiments of the invention can individually or generally be represented that this is only with term " invention "
For convenience, and if in fact disclosing the invention more than one, the scope for being not meant to automatically limit the application is to appoint
What single invention or inventive concept.
The present invention provides a kind of intelligent substation device panel automatic identifying method, its structure diagram is as shown in Figure 1, panel
Information is obtained by the camera of hand-held terminal device, and image recognition core algorithm is loaded in hand held equipment terminal, passes through image
Pretreatment, feature extraction, categorised decision and etc. realize the layout of device panel object element detected.
Embodiment one
Image recognition technology is incorporated into the test job of intelligent substation device surface plate element and layout by the present invention first
In, its automatic identifying method step is as follows：
1. the acquisition of device faceplate formation：Taken pictures by the camera of hand-held terminal device and obtain intelligent substation device surface
Plate image information two-dimensional image information.
2. panel image information pre-processing：The development of counter plate image information is handled, including A D, binaryzation, image it is flat
It is sliding, convert, strengthen, recover, filtering etc..Panel layout recognition methods based on different visual angles, which reaches, identifies different visual angles panel
Support.According to perspective projection principle, panel elements target template image T is subjected to affine transformation, is obtained under different visual angles
The possible form of target board, and then index plane plate element target is generated into the template under different visual angles；And then in identification process
In, it is utilized respectively different affine transformation templates and is matched with images to be recognized I regional areas R, identifies panel layout situation.
3. panels feature extracts and selection：For regions such as panel and keypad, display screen, nameplate, indicator lights, chain type ladder is calculated
Degree feature simultaneously carries out feature extraction.
4. panel element and layout classifier design：Design element and the cloth such as panel and keypad, display screen, nameplate, indicator light
Office's grader, chain type Gradient Features expression similarity measurements flow function.
Robust panel elements target layout recognition methods based on chain type Gradient Features
The overall flow of method is：Panel elements target template image T and input images to be recognized I are decomposed into step-length is
4,6 × 6 overlapping fritters, according to set spatial order construction face plate element target template and the image block sequence of images to be recognized
Row.For each fritter, difference zoning pixel differential；Shape occurs for noise, panel to avoid producing in gatherer process
The influence of the factor to recognition result such as become, further obtain the subject pixel differential of each fritter, and discrete turn to adjacent difference
For 30 ° of 12 directions.There is insensitive attribute to small translation in order to ensure to be laid out recognition methods, according to each piece of acquisition
Discretization direction carry out binaryzation as a result, according to set spatial order construct chained list, establish target template image T respectively, treat
Identify chain type Gradient Features expression List (T, R) and the List (I, R) of image I regional areas R.Similarity measurements flow function is built,
For all possible regional area R in images to be recognized I, target template image T, images to be recognized I regional areas R are calculated
The difference of chain type Gradient Features expression, and then obtain in images to be recognized I, the position of target template.
In order to improve the calculating speed of layout recognition methods, gradual identification method is taken.I.e. for target template figure
As T, images to be recognized I regional area R, only consider that finite part image block rather than all images block are scanned, and are adopted first
With chain type Gradient Features expression way is simplified, it can quickly ignore the impossible position of some in image I, obtain possible position
Candidate.And then progressive increase image number of blocks, the description precision of chain type Gradient Features expression is improved, until obtaining target mould
The exact position of plate.
5. the unrelated panel element in visual angle and layout categorised decision：The panel member under different visual angles is designed by affine transformation
Plain template, counter plate element object is classified in feature space, and identifies each object relative location.
Chain type Gradient Features express similarity measurements flow function
For each fritter, zoning pixel differential, the corresponding differential direction of each pixel obtain as follows respectively：
Coordinate position (x, y) corresponds to picture in the fritter that target board possible form T and images to be recognized I is decomposed under different visual angles
Plain I (x, y) differential Gx(x,y)、Gy(x, y) is tried to achieve by convolution algorithm：
Gx(x, y)=[- 10 1] * I (x, y)
Gy(x, y)=[- 10 1]T*I(x,y)
It is as follows with direction θ (x, y) that amplitude G (x, y) is calculated respectively：
Chain type Gradient Features expression similarity measurements flow function is used for calculating images to be recognized I regional areas R and target template
The number of the similar block of differential characteristics between image T.Its form of Definition is：
Wherein,
Ori (I, c+r) ∈ List (T, R), ori (T, r) ∈ List (T, R), ori (I, c+r) are that images to be recognized I is in place
The discrete main differential direction that c+r just corresponds to fritter is put, ori (T, r) is the possible form T of target board under different visual angles in position
R corresponds to the discrete main differential direction of fritter；C is the center point coordinate of the possible form T of target board under different visual angles, and I is represented
Images to be recognized；δ represents differential characteristics.
The non-sensitive type of similarity measurement function pair miniature deformation is expressed for acquisition chain type Gradient Features, further degree of definition
Flow function ε2Form is：
The non-sensitive type of similarity measurement function pair miniature deformation is expressed for acquisition chain type Gradient Features, further degree of definition
Flow function ε2Form is：
De (T, r)={ ori (T, l):l∈maxmagk(r) ∧ mag (t, l) ＞ τ }
Wherein, De (T, r) represents the set in T differential directions of maximum intensity in each zonule r；Ori (T, l) is T
Differential direction at pixel l, is worth for mag (t, l), and τ is threshold value, maxmagk(r) k maximum differential amplitude in block r is represented
Position；
In order to make chain type Gradient Features expression similarity measurements flow function obtain the insensitivity for overall small translation, repair
It is following ε to change similarity measurements flow function3Form：
Wherein：W (T, m) defines the operation of the possible form T two-dimension translationals translation m of target board under different visual angles, and m is
Two-dimension translational converts, and M is the conversion displacement set of two-dimension translational conversion.
Based on this automatic identifying method, panel layout test job switchs to automatically working by manual work, so that significantly
Improve panel layout test job efficiency.
The above embodiments are merely illustrative of the technical scheme of the present invention and are not intended to be limiting thereof, although with reference to above-described embodiment pair
The present invention is described in detail, those of ordinary skill in the art still can to the present invention embodiment into
Row modification either equivalent substitution these without departing from any modification of spirit and scope of the invention or equivalent substitution, applying
Within pending claims of the invention.
Claims (10)
1. a kind of intelligent substation device panel automatic identifying method, it is characterised in that the described method includes following step：
(1) acquisition of intelligent substation device panel image information；
(2) panel image information pre-processing；
(3) panels feature is extracted and selected；
(4) panel element and layout grader are designed；
(5) the unrelated panel element in visual angle and layout categorised decision.
2. intelligent substation device panel automatic identifying method as claimed in claim 1, it is characterised in that the step (1)
In, taken pictures by the camera of hand-held terminal device and obtain intelligent substation device panel image information two-dimensional image information.
3. intelligent substation device panel automatic identifying method as claimed in claim 1, it is characterised in that the step (2)
In, counter plate image information carry out pretreatment, including A D conversion, binary conversion treatment, image it is smooth, conversion, enhancing, recover
And filtering process；Panel layout recognition methods based on different visual angles reaches the support to the identification of different visual angles panel；According to saturating
Depending on projection theory, panel elements target template image T0 is subjected to affine transformation, obtaining the target board under different visual angles may
Form T, and then by index plane plate element target generate different visual angles under template；And then in identification process, it is utilized respectively
Different affine transformation templates is matched with images to be recognized I regional areas R, identifies panel layout situation.
4. intelligent substation device panel automatic identifying method as claimed in claim 1, it is characterised in that the step (3)
In, for panel and keypad, display screen, nameplate and indicator light element, calculate chain type Gradient Features and carry out feature extraction；
Wherein, the possible form T and images to be recognized I of target board under different visual angles is decomposed into step-length 4, overlapping 6 × 6 small
Block, according to set spatial order construction face plate element target template and the image block sequence of images to be recognized；
For each fritter, zoning pixel differential, the corresponding differential direction of each pixel obtain as follows respectively：It is different
Coordinate position (x, y) respective pixel I in the fritter that target board possible form T and images to be recognized I is decomposed under visual angle
(x, y) differential Gx(x,y)、Gy(x, y) is tried to achieve by convolution algorithm：
Gx(x, y)=[- 10 1] * I (x, y)
Gy(x, y)=[- 10 1]T*I(x,y)
It is as follows with direction θ (x, y) that amplitude G (x, y) is calculated respectively：
<mrow>
<mi>G</mi>
<mrow>
<mo>(</mo>
<mi>x</mi>
<mo>,</mo>
<mi>y</mi>
<mo>)</mo>
</mrow>
<mo>=</mo>
<msqrt>
<mrow>
<msub>
<mi>G</mi>
<mi>x</mi>
</msub>
<msup>
<mrow>
<mo>(</mo>
<mi>x</mi>
<mo>,</mo>
<mi>y</mi>
<mo>)</mo>
</mrow>
<mn>2</mn>
</msup>
<mo>+</mo>
<msub>
<mi>G</mi>
<mi>y</mi>
</msub>
<msup>
<mrow>
<mo>(</mo>
<mi>x</mi>
<mo>,</mo>
<mi>y</mi>
<mo>)</mo>
</mrow>
<mn>2</mn>
</msup>
</mrow>
</msqrt>
</mrow>
<mrow>
<mi>&theta;</mi>
<mrow>
<mo>(</mo>
<mi>x</mi>
<mo>,</mo>
<mi>y</mi>
<mo>)</mo>
</mrow>
<mo>=</mo>
<mi>a</mi>
<mi>r</mi>
<mi>c</mi>
<mi>t</mi>
<mi>a</mi>
<mi>n</mi>
<mrow>
<mo>(</mo>
<mfrac>
<mrow>
<msub>
<mi>G</mi>
<mi>x</mi>
</msub>
<mrow>
<mo>(</mo>
<mi>x</mi>
<mo>,</mo>
<mi>y</mi>
<mo>)</mo>
</mrow>
</mrow>
<mrow>
<msub>
<mi>G</mi>
<mi>y</mi>
</msub>
<mrow>
<mo>(</mo>
<mi>x</mi>
<mo>,</mo>
<mi>y</mi>
<mo>)</mo>
</mrow>
</mrow>
</mfrac>
<mo>)</mo>
</mrow>
</mrow>
Wherein：X, y represent the fritter that target board possible form T and images to be recognized I is decomposed under different visual angles respectively
Middle coordinate position, I (x, y) represent pixel, Gx(x,y)、Gy(x, y) represents the differential of pixel I (x, y) respectively；θ (x, y) represents picture
The corresponding differential direction of plain I (x, y)；
The subject pixel differential of each fritter is further obtained, and discrete to turn to adjacent difference be 30 ° of 12 directions；
Binaryzation is carried out according to each piece of discretization direction of acquisition as a result, constructing chained list according to set spatial order, respectively
Establish the chain type Gradient Features expression List of the possible form T of target board under different visual angles and images to be recognized I regional areas R
(T, R) and List (I, R).
5. intelligent substation device panel automatic identifying method as claimed in claim 1, it is characterised in that the step (4)
In, design panel and keypad, display screen, nameplate and the grader for indicating light panel element, i.e., by designing chain type Gradient Features table
Up to similarity measurements flow function, the chain of the possible form T of target board and images to be recognized I regional areas R under different visual angles is calculated
Formula Gradient Features similarity measure, by the classification of the corresponding T for obtaining maximum similarity, the result as identification panel element.
6. intelligent substation device panel automatic identifying method as claimed in claim 5, it is characterised in that by under different visual angles
Target board possible form T and images to be recognized I is decomposed into step-length 4,6 × 6 overlapping fritters, according to set spatial order structure
Make the image block sequence of panel elements target template and images to be recognized；
For each fritter, zoning pixel differential, the corresponding differential direction of each pixel obtain as follows respectively：It is different
Coordinate position (x, y) respective pixel I in the fritter that target board possible form T and images to be recognized I is decomposed under visual angle
(x, y) differential Gx(x,y)、Gy(x, y) is tried to achieve by convolution algorithm：
Gx(x, y)=[- 10 1] * I (x, y)
Gy(x, y)=[- 10 1]T*I(x,y)
It is as follows with direction θ (x, y) that amplitude G (x, y) is calculated respectively：
<mrow>
<mi>G</mi>
<mrow>
<mo>(</mo>
<mi>x</mi>
<mo>,</mo>
<mi>y</mi>
<mo>)</mo>
</mrow>
<mo>=</mo>
<msqrt>
<mrow>
<msub>
<mi>G</mi>
<mi>x</mi>
</msub>
<msup>
<mrow>
<mo>(</mo>
<mi>x</mi>
<mo>,</mo>
<mi>y</mi>
<mo>)</mo>
</mrow>
<mn>2</mn>
</msup>
<mo>+</mo>
<msub>
<mi>G</mi>
<mi>y</mi>
</msub>
<msup>
<mrow>
<mo>(</mo>
<mi>x</mi>
<mo>,</mo>
<mi>y</mi>
<mo>)</mo>
</mrow>
<mn>2</mn>
</msup>
</mrow>
</msqrt>
</mrow>
<mrow>
<mi>&theta;</mi>
<mrow>
<mo>(</mo>
<mi>x</mi>
<mo>,</mo>
<mi>y</mi>
<mo>)</mo>
</mrow>
<mo>=</mo>
<mi>a</mi>
<mi>r</mi>
<mi>c</mi>
<mi>t</mi>
<mi>a</mi>
<mi>n</mi>
<mrow>
<mo>(</mo>
<mfrac>
<mrow>
<msub>
<mi>G</mi>
<mi>x</mi>
</msub>
<mrow>
<mo>(</mo>
<mi>x</mi>
<mo>,</mo>
<mi>y</mi>
<mo>)</mo>
</mrow>
</mrow>
<mrow>
<msub>
<mi>G</mi>
<mi>y</mi>
</msub>
<mrow>
<mo>(</mo>
<mi>x</mi>
<mo>,</mo>
<mi>y</mi>
<mo>)</mo>
</mrow>
</mrow>
</mfrac>
<mo>)</mo>
</mrow>
</mrow>
Wherein：X, y represent the fritter that target board possible form T and images to be recognized I is decomposed under different visual angles respectively
Middle coordinate position, I (x, y) represent pixel, Gx(x,y)、Gy(x, y) represents the differential of pixel I (x, y) respectively；θ (x, y) represents picture
The corresponding differential direction of plain I (x, y)；
The subject pixel differential of each fritter is further obtained, and discrete to turn to adjacent difference be 30 ° of 12 directions；
Binaryzation is carried out according to each piece of discretization direction of acquisition as a result, constructing chained list according to set spatial order, respectively
Establish chain type Gradient Features expression List (T, R) and the List (I, R) of T and I regional areas R.
Similarity measurements flow function is built, for all possible regional area R in I, calculating target board under different visual angles may
Form T, images to be recognized I regional areas R the expression of chain type Gradient Features difference, and then obtain in images to be recognized I, mesh
Mark the position of template；
Take gradual identification method, i.e., it is local for the possible form T of target board under different visual angles, images to be recognized I
Region R, only considers that finite part image block rather than all images block are scanned, using simplified chain type Gradient Features table first
Up to mode, quickly ignore impossible position in images to be recognized I, obtain possible position candidate；And then progressive increase
Image number of blocks, improves the description precision of chain type Gradient Features expression, until obtaining the position of target template.
7. intelligent substation device panel automatic identifying method as claimed in claim 6, it is characterised in that divide an image into
After fritter, the corresponding differential directions of each pixel I (x, y) in fritter：
<mrow>
<mi>&theta;</mi>
<mrow>
<mo>(</mo>
<mi>x</mi>
<mo>,</mo>
<mi>y</mi>
<mo>)</mo>
</mrow>
<mo>=</mo>
<mi>a</mi>
<mi>r</mi>
<mi>c</mi>
<mi>t</mi>
<mi>a</mi>
<mi>n</mi>
<mrow>
<mo>(</mo>
<mfrac>
<mrow>
<msub>
<mi>G</mi>
<mi>x</mi>
</msub>
<mrow>
<mo>(</mo>
<mi>x</mi>
<mo>,</mo>
<mi>y</mi>
<mo>)</mo>
</mrow>
</mrow>
<mrow>
<msub>
<mi>G</mi>
<mi>y</mi>
</msub>
<mrow>
<mo>(</mo>
<mi>x</mi>
<mo>,</mo>
<mi>y</mi>
<mo>)</mo>
</mrow>
</mrow>
</mfrac>
<mo>)</mo>
</mrow>
</mrow>
Chain type Gradient Features expression similarity measurements flow function is used for calculating images to be recognized I regional areas R and mesh under different visual angles
The number of the similar block of differential characteristics, its form of Definition are between the possible form T of mark panel：
<mrow>
<msub>
<mi>&epsiv;</mi>
<mn>1</mn>
</msub>
<mrow>
<mo>(</mo>
<mi>I</mi>
<mo>,</mo>
<mi>T</mi>
<mo>,</mo>
<mi>c</mi>
<mo>)</mo>
</mrow>
<mo>=</mo>
<munder>
<mo>&Sigma;</mo>
<mrow>
<mi>r</mi>
<mo>&Element;</mo>
<mi>R</mi>
</mrow>
</munder>
<mi>&delta;</mi>
<mrow>
<mo>(</mo>
<mi>o</mi>
<mi>r</mi>
<mi>i</mi>
<mo>(</mo>
<mrow>
<mi>I</mi>
<mo>,</mo>
<mi>c</mi>
<mo>+</mo>
<mi>r</mi>
</mrow>
<mo>)</mo>
<mo>=</mo>
<mi>o</mi>
<mi>r</mi>
<mi>i</mi>
<mo>(</mo>
<mrow>
<mi>T</mi>
<mo>,</mo>
<mi>r</mi>
</mrow>
<mo>)</mo>
<mo>)</mo>
</mrow>
</mrow>
Wherein,
Ori (I, c+r) ∈ List (T, R), ori (T, r) ∈ List (T, R), ori (I, c+r) are images to be recognized I in position c+
The discrete main differential direction of r just corresponding fritters, ori (T, r) be under different visual angles the possible form T of target board r pairs of position
Answer the discrete main differential direction of fritter；C is the center point coordinate of the possible form T of target board under different visual angles, and I represents to wait to know
Other image；δ represents differential characteristics.
8. intelligent substation device panel automatic identifying method as claimed in claim 7, it is characterised in that to obtain chain type ladder
The non-sensitive type of feature representation similarity measurement function pair miniature deformation is spent, further defines metric function ε2Form is：
<mrow>
<msub>
<mi>&epsiv;</mi>
<mn>2</mn>
</msub>
<mrow>
<mo>(</mo>
<mi>I</mi>
<mo>,</mo>
<mi>T</mi>
<mo>,</mo>
<mi>c</mi>
<mo>)</mo>
</mrow>
<mo>=</mo>
<munder>
<mo>&Sigma;</mo>
<mrow>
<mi>r</mi>
<mo>&Element;</mo>
<mi>R</mi>
</mrow>
</munder>
<mi>&delta;</mi>
<mrow>
<mo>(</mo>
<mi>o</mi>
<mi>r</mi>
<mi>i</mi>
<mo>(</mo>
<mrow>
<mi>I</mi>
<mo>,</mo>
<mi>c</mi>
<mo>+</mo>
<mi>r</mi>
</mrow>
<mo>)</mo>
<mo>&Element;</mo>
<mi>D</mi>
<mi>e</mi>
<mo>(</mo>
<mrow>
<mi>T</mi>
<mo>,</mo>
<mi>r</mi>
</mrow>
<mo>)</mo>
<mo>)</mo>
</mrow>
</mrow>
De (T, r)={ ori (T, l):l∈maxmagk(r) ^mag (t, l) ＞ τ }
Wherein, De (T, r) represents the set in T differential directions of maximum intensity in each zonule r；Ori (T, l) is T in picture
Differential direction at plain l, is worth for mag (t, l), and τ is threshold value, maxmagk(r) position of k maximum differential amplitude in block r is represented
Put；
In order to make chain type Gradient Features expression similarity measurements flow function obtain the insensitivity for overall small translation, phase is changed
It is following ε like property metric function3Form：
<mrow>
<msub>
<mi>&epsiv;</mi>
<mn>3</mn>
</msub>
<mrow>
<mo>(</mo>
<mi>I</mi>
<mo>,</mo>
<mi>T</mi>
<mo>,</mo>
<mi>c</mi>
<mo>)</mo>
</mrow>
<mo>=</mo>
<msub>
<mi>max</mi>
<mrow>
<mi>m</mi>
<mo>&Element;</mo>
<mi>M</mi>
</mrow>
</msub>
<msub>
<mi>&epsiv;</mi>
<mn>2</mn>
</msub>
<mrow>
<mo>(</mo>
<mi>I</mi>
<mo>,</mo>
<mi>w</mi>
<mo>(</mo>
<mrow>
<mi>T</mi>
<mo>,</mo>
<mi>m</mi>
</mrow>
<mo>)</mo>
<mo>,</mo>
<mi>c</mi>
<mo>)</mo>
</mrow>
<mo>=</mo>
<msub>
<mi>max</mi>
<mrow>
<mi>m</mi>
<mo>&Element;</mo>
<mi>M</mi>
</mrow>
</msub>
<munder>
<mo>&Sigma;</mo>
<mrow>
<mi>r</mi>
<mo>&Element;</mo>
<mi>R</mi>
</mrow>
</munder>
<mi>&delta;</mi>
<mrow>
<mo>(</mo>
<mi>o</mi>
<mi>r</mi>
<mi>i</mi>
<mo>(</mo>
<mrow>
<mi>I</mi>
<mo>,</mo>
<mi>c</mi>
<mo>+</mo>
<mi>r</mi>
</mrow>
<mo>)</mo>
<mo>&Element;</mo>
<mi>D</mi>
<mi>e</mi>
<mo>(</mo>
<mrow>
<mi>w</mi>
<mrow>
<mo>(</mo>
<mrow>
<mi>T</mi>
<mo>,</mo>
<mi>m</mi>
</mrow>
<mo>)</mo>
</mrow>
<mo>,</mo>
<mi>r</mi>
</mrow>
<mo>)</mo>
<mo>)</mo>
</mrow>
</mrow>
Wherein：W (T, m) defines the operation of the possible form T two-dimension translationals translation m of target board under different visual angles, and m is two dimension
Translation transformation, M are the conversion displacement set of two-dimension translational conversion.
9. intelligent substation device panel automatic identifying method as claimed in claim 1, it is characterised in that the step (5)
In, the possibility form T of the panel element image under panel element different visual angles is obtained by affine transformation, and then calculate difference and regard
The dominant gradient direction of each fritter r in the possibility form T of panel element image under angle, chain is constructed according to set spatial order
Table, obtains the chain type Gradient Features expression of plate element；Counter plate element object is classified in feature space, for different faces
The possibility form T of the different visual angles hypograph of plate element, calculates its chain type gradient with images to be recognized I regional areas R respectively
Characteristic similarity flow function value ε3(I, T, c), calculates：
<mrow>
<mi>T</mi>
<mn>1</mn>
<mo>=</mo>
<mi>arg</mi>
<munder>
<mrow>
<mi>m</mi>
<mi>a</mi>
<mi>x</mi>
</mrow>
<mi>T</mi>
</munder>
<msub>
<mi>&epsiv;</mi>
<mn>3</mn>
</msub>
<mrow>
<mo>(</mo>
<mi>I</mi>
<mo>,</mo>
<mi>T</mi>
<mo>,</mo>
<mi>c</mi>
<mo>)</mo>
</mrow>
</mrow>
It is corresponding to wait to know using C (T1) as classification as a result, wherein C (T1) is the classification of the corresponding panel elements of T1
The coordinate center of other image I regional areas R, as the position for identifying panel element；Exist for each panel element identified
The set of position in images to be recognized I, as the panel layout classification results being calculated.
10. a kind of intelligent substation device panel automatic recognition system, it is characterised in that the system is used for intelligent substation
Device panel automatic identification, the system are intelligent substation device panel detection handheld terminal, the handheld terminal include with
The shooting for being used to obtain intelligent substation device panel image information two-dimensional image information of intelligent substation device panel connection
Head, panel image information pre-processing module, panels feature extract and selecting module, panel element and layout grader and panel cloth
Score of the game class decision-making module.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610935224.0A CN108009541B (en) | 2016-11-01 | 2016-11-01 | Automatic identification method and system for intelligent substation device panel |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610935224.0A CN108009541B (en) | 2016-11-01 | 2016-11-01 | Automatic identification method and system for intelligent substation device panel |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN108009541A true CN108009541A (en) | 2018-05-08 |
| CN108009541B CN108009541B (en) | 2023-07-21 |
Family
ID=62047989
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201610935224.0A Active CN108009541B (en) | 2016-11-01 | 2016-11-01 | Automatic identification method and system for intelligent substation device panel |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN108009541B (en) |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102412627A (en) * | 2011-11-29 | 2012-04-11 | 安徽继远电网技术有限责任公司 | Image identification-based intelligent transformer substation state monitoring system |
| CN103324943A (en) * | 2013-06-18 | 2013-09-25 | 中国人民解放军第二炮兵工程大学 | Identification method of complex device panel image multi-sub zone state |
| US20140331198A1 (en) * | 2011-11-29 | 2014-11-06 | Siemens Aktiengesellschaft | Method for designing a physical layout of a photovoltaic system |
| CN104392432A (en) * | 2014-11-03 | 2015-03-04 | 深圳市华星光电技术有限公司 | Histogram of oriented gradient-based display panel defect detection method |
| US20150371111A1 (en) * | 2014-06-20 | 2015-12-24 | Qualcomm Incorporated | Systems and methods for obtaining structural information from a digital image |
-
2016
- 2016-11-01 CN CN201610935224.0A patent/CN108009541B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102412627A (en) * | 2011-11-29 | 2012-04-11 | 安徽继远电网技术有限责任公司 | Image identification-based intelligent transformer substation state monitoring system |
| US20140331198A1 (en) * | 2011-11-29 | 2014-11-06 | Siemens Aktiengesellschaft | Method for designing a physical layout of a photovoltaic system |
| CN103324943A (en) * | 2013-06-18 | 2013-09-25 | 中国人民解放军第二炮兵工程大学 | Identification method of complex device panel image multi-sub zone state |
| US20150371111A1 (en) * | 2014-06-20 | 2015-12-24 | Qualcomm Incorporated | Systems and methods for obtaining structural information from a digital image |
| CN104392432A (en) * | 2014-11-03 | 2015-03-04 | 深圳市华星光电技术有限公司 | Histogram of oriented gradient-based display panel defect detection method |
Non-Patent Citations (1)
| Title |
|---|
| 张维等: "智能变电站一次设备智能化", 《经营管理者》 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN108009541B (en) | 2023-07-21 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN105894502B (en) | RGBD image significance detection method based on hypergraph model | |
| JP5699788B2 (en) | Screen area detection method and system | |
| US20180082178A1 (en) | Information processing device | |
| Mathavan et al. | Use of a self-organizing map for crack detection in highly textured pavement images | |
| CN103345755A (en) | Chessboard angular point sub-pixel extraction method based on Harris operator | |
| CN106127690A (en) | A kind of quick joining method of unmanned aerial vehicle remote sensing image | |
| CN109801301A (en) | A kind of automatic collection method of tile work progress msg based on BIM and computer vision | |
| CN108596975A (en) | A kind of Stereo Matching Algorithm for weak texture region | |
| CN108876781A (en) | Surface defect recognition method based on SSD algorithm | |
| CN112288758B (en) | Infrared and visible light image registration method for power equipment | |
| CN109389165A (en) | Oil level gauge for transformer recognition methods based on crusing robot | |
| CN103793894A (en) | Cloud model cellular automata corner detection-based substation remote viewing image splicing method | |
| CN106022337B (en) | A kind of planar target detection method based on continuous boundary feature | |
| CN103743750B (en) | A kind of generation method of distribution diagram of surface damage of heavy calibre optical element | |
| CN105354816B (en) | A kind of electronic units fix method and device | |
| CN106529548A (en) | Sub-pixel level multi-scale Harris corner point detection algorithm | |
| CN103761521A (en) | LBP-based microscopic image definition measuring method | |
| CN102799861A (en) | Method for rapidly identifying reading of instrument by using color | |
| CN102509299A (en) | Image salient area detection method based on visual attention mechanism | |
| CN112561989B (en) | Recognition method for hoisting object in construction scene | |
| CN108009541A (en) | A kind of intelligent substation device panel automatic identifying method and system | |
| CN103208003B (en) | Geometric graphic feature point-based method for establishing shape descriptor | |
| CN107944340A (en) | A kind of combination is directly measured and the pedestrian of indirect measurement recognition methods again | |
| CN107248143A (en) | A kind of depth image restorative procedure split based on image | |
| CN109359646A (en) | Liquid level type Meter recognition method based on crusing robot |
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 |