CN105868722A - Identification method and system of abnormal power equipment images - Google Patents
Identification method and system of abnormal power equipment images Download PDFInfo
- Publication number
- CN105868722A CN105868722A CN201610207490.1A CN201610207490A CN105868722A CN 105868722 A CN105868722 A CN 105868722A CN 201610207490 A CN201610207490 A CN 201610207490A CN 105868722 A CN105868722 A CN 105868722A
- Authority
- CN
- China
- Prior art keywords
- image
- power equipment
- abnormal
- drawing picture
- region
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N7/00—Television systems
- H04N7/18—Closed-circuit television [CCTV] systems, i.e. systems in which the video signal is not broadcast
- H04N7/181—Closed-circuit television [CCTV] systems, i.e. systems in which the video signal is not broadcast for receiving images from a plurality of remote sources
Abstract
The invention discloses an identification method of abnormal power equipment images, and is applied to identifying the abnormal power equipment images from power equipment images. The method comprises the following steps: 1, obtaining the power equipment images; 2, extracting power equipment features from the power equipment images; 3, based on the power equipment features, identifying power equipment and image areas where the power equipment is located; and 4, comparing images of the image areas with corresponding comparison images to determine whether abnormal change exists in the images of the image areas so as to identify the abnormal power equipment images, wherein the abnormal change is correlated with operation state abnormities of the power equipment. Correspondingly, the invention also brings forward an identification system of abnormal power equipment images, for identifying abnormal power equipment images from power equipment images.
Description
Technical field
The present invention relates to recognition methods and the system of a kind of abnormal image, particularly relate to a kind of abnormal power and set
The recognition methods of standby image and system.
Background technology
Along with the continuous expansion of China's electric power networks scale, the safe and reliable operation of the power equipments such as transformer station
Most important, and the running status of power equipment is one of key factor determining its safe and stable operation.
By video monitoring system, power equipment carried out the one that picture control is power equipment running status
Monitor mode.Existing video monitoring system only has video monitoring function and recording function, it is impossible to monitoring
Target carries out intelligentized initiative recognition analysis, is only by substantial amounts of image transmitting to dispatching terminal, needs
Operator's moment observation analysis image, adds the work load of operator virtually;Meanwhile, human eye is easy
Tired weakness and the subjectivity of artificial judgment, had a strong impact on power equipment monitoring running state automatization
The further raising of degree;Additionally, the running status of a lot of high pressure equipmentes is difficult to be converted into the signal of telecommunication,
Signal conversion and transmitting procedure are easily affected by strong-electromagnetic field;The operational factor of visual plant needs in real time
Monitoring, uses manual patrol to be difficult to meet requirement of real-time, and the responsibility of floor manager, working attitude
With the result that mental status has had a strong impact on detection;And, human eye is difficult to differentiate the gray scale of fine image and becomes
Change, it is difficult to the degree of objective judgement power equipment surface defect.
Such as, automation system for the power network dispatching (SCADA) high voltage equipment insulation monitoring, relay protection etc. fill
Put in the safe and reliable operation of transformer station, serve important function, but comparatively speaking, power equipment is transported
Row condition monitoring system is the most perfect, and a many of which difficult problem comes from the existence of high voltage and strong-electromagnetic field,
The acquisition of numerous key parameters is limited by measured, environment and measuring method, needs worry about system
The factors such as the transmission of safety, insulation and weak signal;Additionally, some operational factor and failure symptom signal are very
Difficult it is converted into the signal of telecommunication by contact measurement, microcomputer monitoring even cannot be utilized to obtain, such as transformator leakage of oil,
It is truer that this type of important non-electrical signal utilizes image monitoring to obtain.Current electrical network at different levels has been set up
Overlapping remote viewing system of transformer substation, the operation to power system has played important function more.Due to the change in electrical network
Power station is numerous, and each transformer station needs the image more than one of monitoring, and therefore many scholars concentrate and have studied
The collection of remote image, the compression of data and quick transmission problem, and how to alleviate dispatcher and supervise simultaneously
Survey multiple transformer station and the burden of multiple image frame, and determine the operation residing for transformer station quickly and accurately
State.
Summary of the invention
It is an object of the invention to provide the recognition methods of a kind of abnormal power equipment drawing picture, it can be from electricity
Power equipment drawing picture identifies abnormal power equipment image, thus significantly alleviates monitoring personal observations and analyze
The burden of power equipment image, and improve the objectivity of monitoring, real-time and accuracy, for the fastest
Determine that the running status residing for power equipment provides good basis fastly.
Another object of the present invention is to provide the identification system of a kind of abnormal power equipment drawing picture, this system
There is above-mentioned functions equally.
To achieve these goals, the present invention proposes the recognition methods of a kind of abnormal power equipment drawing picture,
For identifying abnormal power equipment image from power equipment image, it includes step:
(1) power equipment image is obtained;
(2) from described power equipment image, power equipment feature is extracted;
(3) based on described power equipment feature identification power equipment and place image-region thereof;
(4) image of described image-region is compared with corresponding contrast images, to judge described figure
As whether the image in region exists ANOMALOUS VARIATIONS, described ANOMALOUS VARIATIONS is abnormal with power equipment running status
Relevant ANOMALOUS VARIATIONS, to identify abnormal power equipment drawing picture.
The design of the recognition methods of abnormal power equipment drawing picture of the present invention is, by from power equipment
Image extracts power equipment feature, and based on this power equipment feature identification power equipment and place figure thereof
As region, afterwards the image of described image-region is compared with corresponding contrast images, to judge to be
The no ANOMALOUS VARIATIONS that exists, thus identify abnormal power equipment drawing picture.Described power equipment feature is described electricity
Reflecting the characteristics of image of the power equipment particular attribute of its correspondence in power equipment drawing picture, it can be that color is special
Levy, the combination of one or more in textural characteristics and shape facility, thus based on described power equipment
Feature just can identify power equipment and place image-region thereof;Described contrast images can be standard picture
Or history image, described standard picture is first to the power equipment corresponding to described power equipment image
The image gathered when coming into operation and be properly functioning, described history image is to described power equipment image
The image gathered when corresponding power equipment runs after coming into operation first;Standard picture is mainly used in
Judging the degree of ANOMALOUS VARIATIONS, interval time is relatively long, and history image is mainly used in judging ANOMALOUS VARIATIONS
Speed, interval time is relatively short, and both have complementary advantages;Described ANOMALOUS VARIATIONS typically refers to mainly reflect institute
State the change that power equipment running status is abnormal, therefore when judging whether ANOMALOUS VARIATIONS, need to consider
The impact of environmental factors, such as illumination, temperature, humidity, visibility etc..
The recognition methods of abnormal power equipment drawing picture of the present invention, it is by from power equipment image
Identify abnormal power equipment image so that monitoring personnel need not each electric power of observation analysis and set
Standby image, thus significantly alleviate monitoring personal observations and analyze the burden of power equipment image, and improve prison
Objectivity, real-time and the accuracy surveyed, for accurately quickly determining the running status residing for power equipment
Good basis is provided.
Further, in the recognition methods of abnormal power equipment drawing picture of the present invention, described step
(1) further comprise the steps of: the power equipment image to obtaining and carry out pretreatment.
Due to obtain power equipment image equipment itself defect and the impact of the factor such as environment, input
Inevitably containing distortion and noise in power equipment image in computer, this can be to below
Power equipment feature extraction, power equipment identification and power equipment graphical analysis bring serious interference, and
And affect the correctness of result.Therefore, first having to carry out the power equipment image collected must
The pretreatment wanted, such as image denoising, edge enhancing, rim detection etc., for later power equipment feature
Extract and necessary preparation is carried out in power equipment identification, to ensure the accuracy of output result.Noise
Threshold method is a kind of simple noise cancellation method, it for cause because of sensor or channel in
The single-point noise now isolating Discrete Distribution has preferable effect.Noise gate method is used to carry out image denoising
Time, first set threshold value, then each pixel of sequence detection image, by this pixel and its neighborhood
Other interior pixels compare judgement, to determine whether as noise spot;If noise spot, then adjacent with it
In territory, the meansigma methods of all pixel grey scales substitutes, and otherwise, exports with former gray value.
Further, in the recognition methods of above-mentioned abnormal power equipment drawing picture, described pretreatment includes:
Histogram enhancement, image denoising, image sharpening, Image Edge-Detection and image segmentation in one or
Several combinations, thus realize removing picture noise, strengthening the details of image and improve the noise of image
The functions such as ratio.
In such scheme, compare the result of the denoising of algorithms of different, sharpening, rim detection, draw as follows
Conclusion: different noise types is needed to choose different Denoising Algorithm;Enter again after image is sharpened
Row rim detection is better;Utilize the rectangular histogram of gray-scale map, choose suitable threshold value, it may be achieved image
Binarization segmentation, reduces the computing scale in image recognition in the future, improves the real-time of power equipment image recognition
Property.
Further, in the recognition methods of abnormal power equipment drawing picture of the present invention, described electric power
Equipment feature includes: the combination of one or more in color characteristic, textural characteristics and shape facility.
Color is information the most rich and varied in image, and different power equipments has the color being not quite similar,
The power equipment obvious to those color characteristics, can using color characteristic as identify equipment main
Foundation.The color coordinate system (or claiming color space) of often application in image procossing mainly has two kinds:
The color space i.e. rgb space that one is made up of red (R), green (G), blue (B) three primary colors, another kind is
IHS space.Wherein, rgb space all uses this coloured silk of RGB towards hardware, the monitor of the overwhelming majority
Color model.RGB corresponds to three stimulus values, forms three-dimensional reference system, any color in this system
Both fall within RGB color cube.
In RGB color space, the mixed proportion of Red Green Blue defines different colors, and
In systems, the dependency between color channel is the highest, if it is the reddest to define trichromatic coordinate
(l, 0,0), green (0, l, 0), blue (0,0, l).The most all of color all falls by certain mixed proportion
In cube, if the coordinate of yellow is then (1,1,0), white coordinate be (1,1, l).
Power equipment significant for color characteristic, we just can choose its color as knowing another characteristic
Vector, is identified power equipment analyzing.As to close to red transformer, the transformator etc. of Lycoperdon polymorphum Vitt,
First extract the color characteristic of power equipment in image, obtain its residing color property space, know as one
Another characteristic vector, positions the identification target in image, analyzes.
Texture is exactly that textured primitive is rearranged by certain deterministic rule or certain statistical law.
The former is referred to as definitiveness texture, such as artificial texture.The latter is then referred to as randomness texture, such as, natural texture.
The method of texture analysis mainly has two kinds: one to be statistical texture analysis method, and two is structural texture analytic process,
Wherein, statistical texture analysis method is the most frequently used.The parameter describing texture has a lot, such as intensity, the texture of texture
Density, the direction of texture and the degree of roughness etc. of texture.Texture analysis based on neighborhood characteristics statistics
Method is using statistical nature the estimating as image texture of gray scale, mainly side in a certain regional area of calculating
Method includes: maximin method, variance method, absolute difference method, information Entropy Method and gaussian filtering differential technique.
Description to shape facility, general requirement is translating and under the conversion (similarity transformation of shape) of rotation
It is constant, because both conversion do not change the shape of object.The image of power equipment is through denoising, increasing
After strong and rim detection, image border and the region of equipment, and then the shape of acquisition equipment drawing picture are just obtained.
Can reflect that the shape information of equipment drawing picture has three kinds of modes: region, border and skeleton.General, mesh
The pixel on mark intra-zone or border gives " 1 " value, and background and other uninterested area pixel are composed
Giving " 0 " value, form bianry image, bianry image provides target shape clearly.
After pretreatment, choose the image spy that can distinguish power equipment classification and place image-region thereof
Levy, generally include the group of one or more in above-mentioned color characteristic, textural characteristics and shape facility
Close, as input vector when identifying power equipment and place image-region thereof afterwards.Such as, to close
The transformer of redness, the transformator etc. of Lycoperdon polymorphum Vitt, first extract the color of power equipment in power equipment image
Feature, obtains its residing color property space as power equipment feature input vector.Power equipment is special
That levies chooses the impact taking into full account its accuracy on identifying process afterwards with rapidity.
Further, in the recognition methods of abnormal power equipment drawing picture of the present invention, described contrast
Image is standard picture and/or history image.
Further, in the recognition methods of abnormal power equipment drawing picture of the present invention, described exception
Change includes producing burr or the edge of projection, or newly-increased false contouring.
Further, in the step (4) of the recognition methods of abnormal power equipment drawing picture of the present invention,
By the comparison of the image of image-region described in following model realization with corresponding contrast images, to judge
State the image of image-region and whether there is ANOMALOUS VARIATIONS:
Δ Pi (x, y)=Pi (x, y)-P (x, y)
In formula, Pi (x, y) is the image of described image-region, P (x, y) is corresponding contrast images, (x, y)
For pixel coordinate;When this formula result is 0, it is judged that the image of described image-region does not exist ANOMALOUS VARIATIONS,
Otherwise judge that the image of described image-region exists ANOMALOUS VARIATIONS.
Such scheme is generally realized by computer, and wherein contrast images is generally selected from database server
Take.
Correspondingly, the invention allows for the identification system of a kind of abnormal power equipment drawing picture, for from electricity
Identifying abnormal power equipment image in power equipment drawing picture, it includes being sequentially connected with: image acquisition is eventually
End, image pick-up card and computer;Described image acquisition terminal is located near corresponding power equipment;
The identification system of described abnormal power equipment drawing picture realizes abnormal by running program corresponding with following step
The identification of power equipment image:
(1) described computer controls the described image acquisition terminal corresponding power equipment image of acquisition;
(2) described computer extracts corresponding power equipment feature from corresponding power equipment image;
(3) described computer based on the corresponding corresponding power equipment of power equipment feature identification and
Place image-region;
(4) image of described image-region is compared by described computer with corresponding contrast images, with
Judge whether ANOMALOUS VARIATIONS, thus identify abnormal power equipment drawing picture.
The identification system of abnormal power equipment drawing picture of the present invention is based on abnormal power of the present invention
The recognition methods of equipment drawing picture realizes, by running the knowledge with abnormal power equipment drawing picture of the present invention
The corresponding program of step of other method realizes the identification of abnormal power equipment drawing picture.
Further, in the identification system of abnormal power equipment drawing picture of the present invention, described image
Acquisition terminal is multiple, and described image pick-up card is multichannel image capture card, and its multiple passages are with described
Multiple image acquisition terminals correspondence respectively.
Further, in the identification system of abnormal power equipment drawing picture of the present invention, described image
Acquisition terminal is CCD (charge-coupled image sensor, Charge Coupled Device) video camera.
The recognition methods of abnormal power equipment drawing picture of the present invention compared with prior art, has following
Beneficial effect:
1) significantly alleviate monitoring personal observations and analyze the burden of power equipment image;
2) objectivity of monitoring, real-time and accuracy it are effectively improved;
3) for accurately quickly determining the basis that the running status residing for power equipment provides good.
The identification system of abnormal power equipment drawing picture of the present invention has the effect above equally.
Accompanying drawing explanation
Fig. 1 is that the identification system of abnormal power equipment drawing picture of the present invention is under a kind of embodiment
Structural schematic block diagram.
Fig. 2 is that the recognition methods of abnormal power equipment drawing picture of the present invention is under a kind of embodiment
Flow chart.
Detailed description of the invention
Below in conjunction with Figure of description and specific embodiment to abnormal power equipment drawing of the present invention
The recognition methods of picture and system make further explanation and explanation.
Fig. 1 illustrates the identification system of abnormal power equipment drawing picture of the present invention at a kind of embodiment
Under structure.
As it is shown in figure 1, the identification system of the abnormal power equipment drawing picture of the present embodiment, for setting from electric power
Identifying abnormal power equipment image in standby image, it includes being sequentially connected with: as image acquisition eventually
Several ccd video cameras, image pick-up card and the computer of end;Several ccd video cameras are arranged on
Near corresponding power equipment, suitable position, is converted to numeral by the optical image signal of corresponding power equipment
Signal, forms corresponding some acquisition channels, is input to computer through multichannel image capture card, its
In some acquisition channels switching by computer control multichannel image capture card realize;The present embodiment
The identification system of abnormal power equipment drawing picture also includes the keyboard being connected with computer, display, printer,
The external equipment such as alarm device and database server.
Refer to Fig. 2, the identification system of the abnormal power equipment drawing picture of the present embodiment is by running and Fig. 2
The corresponding program of step of signal realizes the identification of abnormal power equipment drawing picture:
(1) computer controls several ccd video cameras to obtain corresponding power equipment image, and to obtaining
The power equipment image taken carries out pretreatment;
Above-mentioned pretreatment includes: histogram enhancement, image denoising, image sharpening, Image Edge-Detection and
Image is split;Principle is followed in this pretreatment: choose different Denoising Algorithm for different noise types;
Rim detection is carried out again after image is sharpened;Utilize the rectangular histogram of gray-scale map, choose suitable threshold value,
Realize the binarization segmentation of image, the computing scale in image recognition after reduction, improves power equipment figure
As the real-time identified;
(2) computer extracts corresponding power equipment feature from above-mentioned corresponding power equipment image;?
In the present embodiment, corresponding power equipment feature is to reflect the electricity of its correspondence in corresponding power equipment image
The characteristics of image of power equipment particular attribute, including the combination of color characteristic, textural characteristics and shape facility;
Color is information the most rich and varied in image, and different power equipments has the color being not quite similar, right
The power equipment that those color characteristics are obvious, can be using color characteristic mainly depending on as the equipment of identification
According to.The color coordinate system (or claiming color space) of often application in image procossing mainly has two kinds: a kind of
The color space i.e. rgb space being made up of red (R), green (G), blue (B) three primary colors, another kind is IHS
Space.Wherein, rgb space all uses this colour of RGB towards hardware, the monitor of the overwhelming majority
Model.RGB corresponds to three stimulus values, forms three-dimensional reference system, and in this system, any color is all
Fall in RGB color cube.
In RGB color space, the mixed proportion of Red Green Blue defines different colors, and
In systems, the dependency between color channel is the highest, if it is the reddest to define trichromatic coordinate
(l, 0,0), green (0, l, 0), blue (0,0, l).The most all of color all falls by certain mixed proportion
In cube, if the coordinate of yellow is then (1,1,0), white coordinate be (1,1, l).
Power equipment significant for color characteristic, we just can choose its color as knowing another characteristic
Vector, is identified power equipment analyzing.As to close to red transformer, the transformator etc. of Lycoperdon polymorphum Vitt,
First extract the color characteristic of power equipment in image, obtain its residing color property space, know as one
Another characteristic vector, positions the identification target in image, analyzes.
Texture is exactly that textured primitive is rearranged by certain deterministic rule or certain statistical law.
The former is referred to as definitiveness texture, such as artificial texture.The latter is then referred to as randomness texture, such as, natural texture.
The method of texture analysis mainly has two kinds: one to be statistical texture analysis method, and two is structural texture analytic process,
Wherein, statistical texture analysis method is the most frequently used.The parameter describing texture has a lot, such as intensity, the texture of texture
Density, the direction of texture and the degree of roughness etc. of texture.Texture analysis based on neighborhood characteristics statistics
Method is using statistical nature the estimating as image texture of gray scale, mainly side in a certain regional area of calculating
Method includes: maximin method, variance method, absolute difference method, information Entropy Method and gaussian filtering differential technique.
Description to shape facility, general requirement is translating and under the conversion (similarity transformation of shape) of rotation
It is constant, because both conversion do not change the shape of object.The image of power equipment is through denoising, increasing
After strong and rim detection, image border and the region of equipment, and then the shape of acquisition equipment drawing picture are just obtained.
Can reflect that the shape information of equipment drawing picture has three kinds of modes: region, border and skeleton.General, mesh
The pixel on mark intra-zone or border gives " 1 " value, and background and other uninterested area pixel are composed
Giving " 0 " value, form bianry image, bianry image provides target shape clearly;
(3) computer is based on the above-mentioned corresponding corresponding power equipment of power equipment feature identification and place thereof
Image-region;The method identification electric power of the present embodiment application color characteristic, textural characteristics and template matching sets
Standby classification;Based on shape facility identification corresponding power equipment place image-region;
(4) computer is by the image of image-region described in following model realization and corresponding contrast images
Relatively, to judge whether the image of described image-region exists ANOMALOUS VARIATIONS:
Δ Pi (x, y)=Pi (x, y)-P (x, y)
In formula, (x, y) be the image of described image-region to Pi, and (x, y) for choosing from database server for P
Corresponding contrast images, (x y) is pixel coordinate;When this formula result is 0, it is judged that described image district
There is not ANOMALOUS VARIATIONS in the image in territory, shows that corresponding power equipment is in normal running status, when this
When formula result is not 0, it is judged that the image of described image-region exists ANOMALOUS VARIATIONS and (shows as certain part to send out
Give birth to sudden change, as created burr, the edge of projection or some false contouring newly-increased etc.), show corresponding electricity
Power equipment faulty (damage of such as outward appearance, electric discharge phenomena and equipment oil leakage etc.), and will be deemed as depositing
At the electricity belonging to the image of the above-mentioned image-region of the change reflecting corresponding power equipment running status exception
Power equipment drawing picture is identified as abnormal power equipment drawing picture.
In the present embodiment, computer be not detected by abnormal power equipment drawing as time do not transmit power equipment
Image, to dispatching terminal computer, only transmits analysis result;COMPUTER DETECTION to abnormal power equipment drawing as time,
Abnormal power equipment drawing picture and alarm signal are sent to dispatching terminal computer, and dispatcher hears alarm sound
After, find corresponding power equipment image to observe and process further according to prompting, thus significantly reduce
The burden of a large amount of real time imaging of dispatcher's Continuous Observation, and need not be communication channel " crowded " again and too much
Consider data compression problem.
In the present embodiment, COMPUTER DETECTION to abnormal power equipment drawing as time, need corresponding electric power is set
For being further analyzed, at this moment realize corresponding power equipment by control multichannel image capture card
The continuous acquisition of image.
The recognition methods abnormal power based on above-mentioned the present embodiment of the abnormal power equipment drawing picture of the present embodiment
The identification system of equipment drawing picture realizes, and the abnormal power equipment drawing of its step and above-mentioned the present embodiment as
The correlation step of identification system is corresponding, therefore repeats no more.
It should be noted that the listed above specific embodiment being only the present invention, it is clear that the invention is not restricted to
Above example, has the similar change of many therewith.If those skilled in the art is public from the present invention
All deformation that the content opened directly derives or associates, all should belong to protection scope of the present invention.
Claims (10)
1. a recognition methods for abnormal power equipment drawing picture, abnormal for identifying from power equipment image
Power equipment image, it is characterised in that include step:
(1) power equipment image is obtained;
(2) from described power equipment image, power equipment feature is extracted;
(3) based on described power equipment feature identification power equipment and place image-region thereof;
(4) image of described image-region is compared with corresponding contrast images, to judge
Whether the image stating image-region exists ANOMALOUS VARIATIONS, and described ANOMALOUS VARIATIONS is to run shape with power equipment
The abnormal relevant ANOMALOUS VARIATIONS of state, to identify abnormal power equipment drawing picture.
2. the recognition methods of abnormal power equipment drawing picture as claimed in claim 1, it is characterised in that described step
Suddenly the power equipment image that (1) further comprises the steps of: obtaining carries out pretreatment.
3. the recognition methods of abnormal power equipment drawing picture as claimed in claim 2, it is characterised in that described pre-
Process includes: histogram enhancement, image denoising, image sharpening, Image Edge-Detection and image divide
The combination of one or more in cutting.
4. the recognition methods of abnormal power equipment drawing picture as claimed in claim 1, it is characterised in that described electricity
Power equipment feature includes: the group of one or more in color characteristic, textural characteristics and shape facility
Close.
5. the recognition methods of abnormal power equipment drawing picture as claimed in claim 1, it is characterised in that described right
It is standard picture and/or history image than image.
6. the recognition methods of abnormal power equipment drawing picture as claimed in claim 1, it is characterised in that described different
Often change includes producing burr or the edge of projection, or newly-increased false contouring.
7. the recognition methods of abnormal power equipment drawing picture as claimed in claim 1, it is characterised in that described step
Suddenly in (4), by image and the corresponding contrast images of image-region described in following model realization
Relatively, to judge whether the image of described image-region exists ANOMALOUS VARIATIONS:
Δ Pi (x, y)=Pi (x, y)-P (x, y)
In formula, Pi (x, y) is the image of described image-region, P (x, y) is corresponding contrast images, (x, y)
For pixel coordinate;When this formula result is 0, it is judged that the image of described image-region does not exist abnormal change
Change, otherwise judge that the image of described image-region exists ANOMALOUS VARIATIONS.
8. an identification system for abnormal power equipment drawing picture is abnormal for identifying from power equipment image
Power equipment image, it is characterised in that include being sequentially connected with: image acquisition terminal, image acquisition
Card and computer;Described image acquisition terminal is located near corresponding power equipment;Described exception is electric
The identification system of power equipment drawing picture realizes abnormal power by operation program corresponding with following step and sets
The identification of standby image:
(1) described computer controls the described image acquisition terminal corresponding power equipment image of acquisition;
(2) described computer extracts corresponding power equipment from corresponding power equipment image
Feature;
(3) described computer is based on the corresponding corresponding power equipment of power equipment feature identification
And place image-region;
(4) image of described image-region is compared by described computer with corresponding contrast images,
To judge whether ANOMALOUS VARIATIONS, thus identify abnormal power equipment drawing picture.
9. the identification system of abnormal power equipment drawing picture as claimed in claim 8, it is characterised in that described figure
As acquisition terminal is multiple, described image pick-up card is multichannel image capture card, its multiple passages with
The plurality of image acquisition terminal correspondence respectively.
10. the identification system of abnormal power equipment drawing picture as claimed in claim 8, it is characterised in that described figure
As acquisition terminal is ccd video camera.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201610207490.1A CN105868722A (en) | 2016-04-05 | 2016-04-05 | Identification method and system of abnormal power equipment images |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201610207490.1A CN105868722A (en) | 2016-04-05 | 2016-04-05 | Identification method and system of abnormal power equipment images |
Publications (1)
Publication Number | Publication Date |
---|---|
CN105868722A true CN105868722A (en) | 2016-08-17 |
Family
ID=56627052
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201610207490.1A Pending CN105868722A (en) | 2016-04-05 | 2016-04-05 | Identification method and system of abnormal power equipment images |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN105868722A (en) |
Cited By (27)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN106296949A (en) * | 2016-09-13 | 2017-01-04 | 成都创慧科达科技有限公司 | A kind of gate control system based on H264 video decoding techniques and method |
CN106340034A (en) * | 2016-08-22 | 2017-01-18 | 成都信息工程大学 | Transformer oil leak detection method |
CN106355187A (en) * | 2016-09-07 | 2017-01-25 | 西华大学 | Application of visual information to electrical equipment monitoring |
CN106680217A (en) * | 2016-12-21 | 2017-05-17 | 四川华雁信息产业股份有限公司 | Method and device for monitoring failure of transformer respirator |
CN107146704A (en) * | 2017-07-10 | 2017-09-08 | 河南远中电力设备有限公司 | A kind of transformer processing unit (plant) and production technology |
CN107180242A (en) * | 2017-05-25 | 2017-09-19 | 上海地铁第运营有限公司 | Security against fire hidden danger image-recognizing method based on embedded system |
CN107609498A (en) * | 2017-09-01 | 2018-01-19 | 湖州师范学院 | A kind of data processing method of computer supervisory control system |
CN107818562A (en) * | 2017-10-23 | 2018-03-20 | 广东电网有限责任公司东莞供电局 | Insulate encapsulated layer crackle online test method in a kind of dry type hollow shunt reactor air channel |
CN108226677A (en) * | 2017-12-29 | 2018-06-29 | 武汉神宇电气有限公司 | Metal cabinet and nonstandard cabinet fault rapid detecting method and system |
CN108764166A (en) * | 2018-05-30 | 2018-11-06 | 天仁民防建筑工程设计有限公司 | A kind of closed guard gate's location status detecting system and method based on supporting rod |
CN108764260A (en) * | 2018-05-30 | 2018-11-06 | 天仁民防建筑工程设计有限公司 | A kind of closed guard gate's location status detecting system and method based on supporting rod mark |
CN108845888A (en) * | 2018-05-24 | 2018-11-20 | 郑州云海信息技术有限公司 | A kind of method and system based on TensorFlow image recognition diagnosis memory failure |
CN108873777A (en) * | 2018-06-29 | 2018-11-23 | 北京航天自动控制研究所 | A kind of ground test launch control system monitor supervision platform and monitoring method |
CN108960256A (en) * | 2018-06-28 | 2018-12-07 | 东软集团股份有限公司 | A kind of determination method, device and equipment of components damage degree |
CN109120906A (en) * | 2018-10-30 | 2019-01-01 | 信利光电股份有限公司 | A kind of intelligent monitor system |
CN109165233A (en) * | 2018-09-06 | 2019-01-08 | 江苏佳源科技有限公司 | A kind of intelligent low-pressure branch monitoring system |
CN109543505A (en) * | 2018-09-29 | 2019-03-29 | 江苏濠汉智能设备有限公司 | A kind of object detection system and method based on video image |
CN109740507A (en) * | 2018-12-29 | 2019-05-10 | 国网浙江省电力有限公司 | A kind of method of transformer equipment abnormality sensing |
CN109784336A (en) * | 2019-01-30 | 2019-05-21 | 云南电网有限责任公司电力科学研究院 | A kind of infrared image fault point recognition methods based on YOLO algorithm of target detection |
CN110188783A (en) * | 2018-02-22 | 2019-08-30 | 北京猎户星空科技有限公司 | A kind of image classification method, device and electronic equipment |
CN111726571A (en) * | 2020-01-05 | 2020-09-29 | 杨文娟 | High-voltage line distribution shape detection system |
CN111900694A (en) * | 2020-07-07 | 2020-11-06 | 贵州电网有限责任公司 | Relay protection equipment information acquisition method and system based on automatic identification |
CN112081714A (en) * | 2019-06-14 | 2020-12-15 | 李晓芳 | Wind power generation control system |
CN112712111A (en) * | 2020-12-23 | 2021-04-27 | 南方电网深圳数字电网研究院有限公司 | Device state detection method, electronic device, and storage medium |
CN113553999A (en) * | 2021-09-17 | 2021-10-26 | 江苏新恒基特种装备股份有限公司 | Method and system for judging molding abnormity based on image recognition |
CN115049650A (en) * | 2022-08-12 | 2022-09-13 | 深圳凯升联合科技有限公司 | Electrical equipment leakage detection method based on image processing |
CN117291911A (en) * | 2023-11-24 | 2023-12-26 | 山东通广电子股份有限公司 | Defect detection method and system for power equipment |
Citations (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103595125A (en) * | 2012-08-13 | 2014-02-19 | 西安元朔科技有限公司 | An electrical device on-line monitoring system based on an image identification technique |
CN103997126A (en) * | 2014-05-28 | 2014-08-20 | 华北电网有限公司 | Fault diagnosis grading method and system based on on-off state |
-
2016
- 2016-04-05 CN CN201610207490.1A patent/CN105868722A/en active Pending
Patent Citations (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103595125A (en) * | 2012-08-13 | 2014-02-19 | 西安元朔科技有限公司 | An electrical device on-line monitoring system based on an image identification technique |
CN103997126A (en) * | 2014-05-28 | 2014-08-20 | 华北电网有限公司 | Fault diagnosis grading method and system based on on-off state |
Cited By (34)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN106340034A (en) * | 2016-08-22 | 2017-01-18 | 成都信息工程大学 | Transformer oil leak detection method |
CN106355187A (en) * | 2016-09-07 | 2017-01-25 | 西华大学 | Application of visual information to electrical equipment monitoring |
CN106296949B (en) * | 2016-09-13 | 2018-11-27 | 成都创慧科达科技有限公司 | A kind of access control system and method based on H264 video decoding techniques |
CN106296949A (en) * | 2016-09-13 | 2017-01-04 | 成都创慧科达科技有限公司 | A kind of gate control system based on H264 video decoding techniques and method |
CN106680217A (en) * | 2016-12-21 | 2017-05-17 | 四川华雁信息产业股份有限公司 | Method and device for monitoring failure of transformer respirator |
CN106680217B (en) * | 2016-12-21 | 2019-09-06 | 四川华雁信息产业股份有限公司 | Transformer breather fault monitoring method and device |
CN107180242A (en) * | 2017-05-25 | 2017-09-19 | 上海地铁第运营有限公司 | Security against fire hidden danger image-recognizing method based on embedded system |
CN107146704A (en) * | 2017-07-10 | 2017-09-08 | 河南远中电力设备有限公司 | A kind of transformer processing unit (plant) and production technology |
CN107609498A (en) * | 2017-09-01 | 2018-01-19 | 湖州师范学院 | A kind of data processing method of computer supervisory control system |
CN107609498B (en) * | 2017-09-01 | 2020-08-18 | 湖州师范学院 | Data processing method of computer monitoring system |
CN107818562A (en) * | 2017-10-23 | 2018-03-20 | 广东电网有限责任公司东莞供电局 | Insulate encapsulated layer crackle online test method in a kind of dry type hollow shunt reactor air channel |
CN108226677A (en) * | 2017-12-29 | 2018-06-29 | 武汉神宇电气有限公司 | Metal cabinet and nonstandard cabinet fault rapid detecting method and system |
CN110188783A (en) * | 2018-02-22 | 2019-08-30 | 北京猎户星空科技有限公司 | A kind of image classification method, device and electronic equipment |
CN108845888A (en) * | 2018-05-24 | 2018-11-20 | 郑州云海信息技术有限公司 | A kind of method and system based on TensorFlow image recognition diagnosis memory failure |
CN108764260A (en) * | 2018-05-30 | 2018-11-06 | 天仁民防建筑工程设计有限公司 | A kind of closed guard gate's location status detecting system and method based on supporting rod mark |
CN108764166A (en) * | 2018-05-30 | 2018-11-06 | 天仁民防建筑工程设计有限公司 | A kind of closed guard gate's location status detecting system and method based on supporting rod |
CN108960256A (en) * | 2018-06-28 | 2018-12-07 | 东软集团股份有限公司 | A kind of determination method, device and equipment of components damage degree |
CN108873777A (en) * | 2018-06-29 | 2018-11-23 | 北京航天自动控制研究所 | A kind of ground test launch control system monitor supervision platform and monitoring method |
CN109165233A (en) * | 2018-09-06 | 2019-01-08 | 江苏佳源科技有限公司 | A kind of intelligent low-pressure branch monitoring system |
CN109165233B (en) * | 2018-09-06 | 2019-07-12 | 江苏佳源科技有限公司 | A kind of intelligent low-pressure branch monitoring system |
CN109543505A (en) * | 2018-09-29 | 2019-03-29 | 江苏濠汉智能设备有限公司 | A kind of object detection system and method based on video image |
CN109120906A (en) * | 2018-10-30 | 2019-01-01 | 信利光电股份有限公司 | A kind of intelligent monitor system |
CN109740507A (en) * | 2018-12-29 | 2019-05-10 | 国网浙江省电力有限公司 | A kind of method of transformer equipment abnormality sensing |
CN109784336A (en) * | 2019-01-30 | 2019-05-21 | 云南电网有限责任公司电力科学研究院 | A kind of infrared image fault point recognition methods based on YOLO algorithm of target detection |
CN112081714A (en) * | 2019-06-14 | 2020-12-15 | 李晓芳 | Wind power generation control system |
CN111726571A (en) * | 2020-01-05 | 2020-09-29 | 杨文娟 | High-voltage line distribution shape detection system |
CN111900694A (en) * | 2020-07-07 | 2020-11-06 | 贵州电网有限责任公司 | Relay protection equipment information acquisition method and system based on automatic identification |
CN111900694B (en) * | 2020-07-07 | 2022-12-27 | 贵州电网有限责任公司 | Relay protection equipment information acquisition method and system based on automatic identification |
CN112712111A (en) * | 2020-12-23 | 2021-04-27 | 南方电网深圳数字电网研究院有限公司 | Device state detection method, electronic device, and storage medium |
CN113553999A (en) * | 2021-09-17 | 2021-10-26 | 江苏新恒基特种装备股份有限公司 | Method and system for judging molding abnormity based on image recognition |
CN113553999B (en) * | 2021-09-17 | 2021-12-17 | 江苏新恒基特种装备股份有限公司 | Method and system for judging molding abnormity based on image recognition |
CN115049650A (en) * | 2022-08-12 | 2022-09-13 | 深圳凯升联合科技有限公司 | Electrical equipment leakage detection method based on image processing |
CN117291911A (en) * | 2023-11-24 | 2023-12-26 | 山东通广电子股份有限公司 | Defect detection method and system for power equipment |
CN117291911B (en) * | 2023-11-24 | 2024-02-09 | 山东通广电子股份有限公司 | Defect detection method and system for power equipment |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN105868722A (en) | Identification method and system of abnormal power equipment images | |
CN110826538A (en) | Abnormal off-duty identification system for electric power business hall | |
CN104408459A (en) | Image identification method applied to power equipment monitoring | |
CN109271844B (en) | Electrical cabinet electrical symbol recognition method based on OpenCV | |
CN110070530A (en) | A kind of powerline ice-covering detection method based on deep neural network | |
CN109118548A (en) | A kind of comprehensive intelligent water quality recognition methods | |
CN106096603A (en) | A kind of dynamic flame detection method merging multiple features and device | |
CN106600580B (en) | Hough transform-based power line anomaly identification method and system | |
CN109740507A (en) | A kind of method of transformer equipment abnormality sensing | |
CN105894015B (en) | A kind of banister state analysis method and system | |
CN110348380A (en) | A kind of visualization grounding switch state video recognition system and method | |
CN100469138C (en) | Power transformer draught fan state recognizing method based on video monitoring and image recognition | |
CN115100562A (en) | Intelligent monitoring system and method for equipment line based on video image and deep learning | |
CN112839200A (en) | Power plant violation behavior identification method and system based on 5G technology and network server | |
CN113536440B (en) | Data processing method based on BIM operation and maintenance management system | |
CN102999994A (en) | Flame detection device based on video image analysis | |
CN103020588A (en) | Flame detection method based on video image analysis | |
CN112508022A (en) | Transformer substation anti-misoperation system and method based on operation and inspection operation full process | |
CN111428987A (en) | Artificial intelligence-based image identification method and system for relay protection device | |
CN116052041A (en) | Indicating lamp state identification method based on depth network | |
CN103927544A (en) | Machine vision grading method for ginned cotton rolling quality | |
CN111652055B (en) | Intelligent switch instrument identification method based on two-stage positioning | |
CN103020587A (en) | View analyzing method based on video image analysis | |
CN111988569A (en) | Method and system for monitoring ignition phenomenon of industrial video monitoring picture of transformer substation | |
Li et al. | Key technology simulation of equipment anomaly detection based on image processing in underground mining |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
C06 | Publication | ||
PB01 | Publication | ||
C10 | Entry into substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20160817 |
|
WD01 | Invention patent application deemed withdrawn after publication |