CN106951900A - A kind of automatic identifying method of arrester meter reading - Google Patents
A kind of automatic identifying method of arrester meter reading Download PDFInfo
- Publication number
- CN106951900A CN106951900A CN201710239497.6A CN201710239497A CN106951900A CN 106951900 A CN106951900 A CN 106951900A CN 201710239497 A CN201710239497 A CN 201710239497A CN 106951900 A CN106951900 A CN 106951900A
- Authority
- CN
- China
- Prior art keywords
- image
- area
- reading
- connected domain
- rectangle
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/40—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
- G06V10/443—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 by matching or filtering
-
- 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
- G06V10/24—Aligning, centring, orientation detection or correction of the image
- G06V10/243—Aligning, centring, orientation detection or correction of the image by compensating for image skew or non-uniform image deformations
-
- 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
- G06V10/26—Segmentation of patterns in the image field; Cutting or merging of image elements to establish the pattern region, e.g. clustering-based techniques; Detection of occlusion
- G06V10/267—Segmentation of patterns in the image field; Cutting or merging of image elements to establish the pattern region, e.g. clustering-based techniques; Detection of occlusion by performing operations on regions, e.g. growing, shrinking or watersheds
-
- 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
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Multimedia (AREA)
- Theoretical Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Image Analysis (AREA)
Abstract
The invention discloses a kind of automatic identifying method of arrester meter reading, the identification of lightning-arrest instruments and meters is divided into region segmentation and Recognition of Reading two parts；Instrument dial plate is divided into multiple regions first with connected domain detection algorithm and rectangle fitting algorithm；Then these regions are detected using priori, retains the connected domain for representing pointer area and numeric area；Minimum area rectangle fitting is carried out to two connected domains again, two rectangles with deflection angle are obtained；Rotation correction is carried out to image according to angle, then pointer and numeric area is partitioned into from correction chart picture；Finally, preset angle configuration and convolutional neural networks method are utilized respectively Recognition of Reading is carried out to pointer area and numeric area.The present invention can complete total indicator reading identification and digital Recognition of Reading simultaneously, and effective correction can be carried out to image, the accuracy of reading is improved.
Description
Technical field
The invention belongs to image identification technical field, more particularly to a kind of automatic identifying method of arrester meter reading.
Background technology
Arrester is a kind of electrical equipment dedicated for limitation lightning surge or switching overvoltage in transformer station, internal
The valve block changed containing a resistance value with voltage.Under rated voltage, valve block resistance value is very big, equivalent to one insulation
Body, then flows through the current value very little and stably of the valve block；After both end voltage exceedes threshold value, valve block is switched on, and is had very big
Electric current by valve block, be subsequently poured into the earth, so avoid heavy current impact other equipment in parallel, electricity was completed with this
Pressure protection；When voltage recovers, valve block state reverts to insulation, and current value also recovers therewith.Valve block is living through voltge surge
Afterwards, it is likely that impaired, it is therefore desirable to often detect the working condition of arrester.
Be configured with a lightning-arrest instruments and meters in the working condition of arrester, equipment for the ease of detecting.Contain on instrument dial plate
There are pointer and digital two reading areas：The reading of pointer represents the current value for flowing through arrester, and digital reading represents this and taken shelter from the thunder
Device is by number of lightning strokes (number of times for living through voltge surge).The size and arrester unusual condition of current value are directly related, are
The critical data recorded is needed during inspection；For ensure arrester can normal work, it is necessary to enter after it is by voltge surge
Row once thoroughly hand inspection, therefore, number of lightning strokes needs also exist for record.At present, this writing task is still by artificial complete
Into on the one hand this mode missing inspection easily occurs, and on the other hand relatively large deviation easily occur in the data and True Data of record.Cause
This is, it is necessary to a kind of Meter recognition recording process of automation.
Current Meter recognition, handled object is all the instrument with single reading area.For pointer instrument
Identification, conventional method be by fitting a straight line detect dial plate on pointer, then utilize angle information and priori meter
Calculate reading.This method is applied to pointer length and approached with dial plate length and the obvious instrument of color of pointer texture；Work as instrument
Position is compared with timing, and total indicator reading is more accurate, but when the instrument of shooting is tilted, the result of identification often deviation compared with
Greatly.Therefore, be not suitable for the instrument that putting position differs in processing transformer station, and failure is recognized for the pointer for instruments and meters of taking shelter from the thunder.
For the identification of digital instrument, method is more, including template matching method, statistic decision method, BP neural network method
Deng.Wherein, template matching method is by the way that the template numeral in numeral to be identified and ATL is compared, with similarity most
Numeral corresponding to that big template is used as recognition result；This method effect when recognizing press figure is fine, but easily
By noise jamming, be not suitable for identification outdoor meter.Statistic decision method be difficult to reflection numeral in tiny characteristics, so using compared with
It is few.BP neural network method can constantly learn and change parameters at different levels in the training process, so as to reach to training sample pole
Good classifying quality；But it is overly dependent upon the selection of input feature value.In addition, the digital instrument studied at present,
The segmentation in region need not be carried out, emphasis is all in Classification and Identification；And the numeric area in lightning-arrest instruments and meters is smaller, it is necessary to divide
It could be handled after cutting, so to be improved to existing method.
The content of the invention
The invention aims to provide a kind of self adaptation it is strong, can be while completing total indicator reading identification and numeral is read
The automatic identifying method of the arrester meter reading of number identification.
Therefore, the technical scheme is that：A kind of automatic identifying method of arrester meter reading, including following step
Suddenly：
1）Image preprocessing：Using gaussian pyramid down-sampled images, the Instrument image of input is reduced into resolution ratio, then by coloured silk
Color image is converted to gray level image, and is filtered to gray level image processing；
2）Region segmentation：According to the characteristics of Meter recognition region, determine that the idiographic flow of image segmentation is as follows：
a1）Edge extracting and Morphological scale-space；Rim detection is carried out using Canny operators, a bianry image, Ran Houli is obtained
The edge of fracture is connected with expansion process；
a2）Connected domain is detected；8 fields using pixel are found as syntople and mark the connected domain in image, and entire image will
It is divided into N number of connected domain；
a3）Calculate the boundary rectangle of each connected domain；The positive rectangle of all pixels point comprising some connected domain is exactly this company
The boundary rectangle in logical domain；
a4）Filtering interfering connected domain；According to pointer area and numeric area in the relative position information of dial plate, one can be filtered out
A little incorrect extraneous rectangles such as area is too small, area is excessive, percent information gap is big, namely filter out underproof connection
Domain；If the resolution ratio of image is, the information of extraneous rectangle is：, wherein the first two information generation
The coordinate on table rectangle summit, latter two information represents the width and height held；When this rectangle meets following（I）、（II）、
（III）During either condition, the UNICOM domain corresponding to it is deleted；If remaining connected domain>2, the work of the percentage of occupancy at most
To retain item；
（I）
（II）
（III）
If will meet（I）、（II）、（III）After the UNICOM domain of middle either condition is filtered, remaining connected domain number is 2, then enters
Enter step a6）, otherwise, remaining connected domain is calculated as the following formula：, wherein have in molecules present connected domain
The number of point is imitated, denominator represents the area of extraneous rectangle；Result of calculation is ranked up, maximum of which two is selected, is
Remaining two connected domains；
a5）Minimum area rectangle fitting is carried out to remaining two connected domains；Minimum area rectangle is to refer to include connected domain
The minimum rectangle of the area of interior all pixels point, and this rectangle is likely to inclined；
If the central point of two rectangles of fitting is respectivelyWith, deflection angle is respectivelyWith, calculate；
a6）According toCorrecting colour images and bianry image；School be exactly based on spatial alternation original image is mapped as it is new
Image, original coordinates is converted to as the following formula new coordinate：Wherein, the matrix of two rows three row is referred to as affine matrix；
Instrument image timing only carries out rotation correction, without translating, therefore affine matrix is:；
When carrying out spatial alternation, the position corresponding to input picture is found in turn from the pixel of output image, if the picture of mapping
Vegetarian refreshments non-integer, then utilize the single order interpolation calculation pixel value；If the pixel of mapping exceeds the scope of input picture,
The point is entered as zero；
a7）Pointer area is partitioned into from the bianry image of correction, and it is partitioned into from the coloured image of correction pointer areaAnd numeric area；
3）Recognition of Reading：Lightning-arrest instruments and meters has pointer and digital two reading areas, uses Hough transform detection of straight lines and angle
Method recognizes total indicator reading；Use sciagraphy segmentation numeral, neural network identification numeral；Complete the Recognition of Reading of whole instrument.
Further, the step 2）Middle a2）Described connected domain detection is concretely comprised the following steps：
b1）Image is progressively scanned, pixel value in each row formed a whole for 255 pixel, and marks
Its line number, mark number, starting point, terminal, mark number is entered as by the pixel value of this all pixels point integrally included；
b2）From the second row, whether the region for observing this line UNICOM intersects with the region of lastrow, should if intersecting
The mark number in region is changed to the mark number with his intersecting area, and the pixel value of the pixel covered is also changed to the value of mark number；
Otherwise it is constant；
b3）After image scanning terminates, the pixel value of all non-zero pixels points is updated；
b4）Again image is scanned, regards pixel value identical pixel as a connected domain, such entire image will be by
It is divided into N number of connected domain.
Further, the step 2）Middle a7）Described is partitioned into pointer area, and divide from the coloured image of correction
Cut out pointer areaAnd numeric areaSpecific segmentation step be：
C1) using the bianry image after correction as object, connected domain therein is detected, boundary rectangle and the calculating of connected domain is fitted
The center point coordinate of rectangle；
C2) by all center point coordinates and step a5）'sWithIt is compared, retains coordinate closest
Two rectangles；
C3) it is partitioned into respectively from the image of correction。
Further, the step 3）Described in pointer identification comprise the following steps that：
D1 the boundary line angle of pointer area) is calculated；Utilize Hough transform detection zoneTwo boundary lines, meter
Calculate the angle of two lines section；BecauseIt is the region split from bianry image, only includes the obvious line segment in border, do not have
There is the interference of pointer and graduation mark, can more accurately detect the two lines section for representing instrument border；
D2) to regionCarry out gradation conversion, homomorphic filtering processing and piecewise linear transform processing；Homomorphic filtering passes through instead
Light and incident light separation are penetrated, incident light therein is filtered, so as to reduce the influence that illumination is caused to image；By homomorphic filtering
Picture contrast after processing is low, using piecewise linear transform using the gray average of image as boundary, improves pair of whole image
Degree of ratio；
D3) image binaryzation；Binary-state threshold is calculated using maximum variance between clusters, binary image is then drawn；
D4 the deflection angle of pointer) is calculated；The line segment in binary image is detected first with Hough transform；Then line is screened
Section, in a plurality of line segment of detection, is filtered out and boundary lineThe close line segment in position, then select length from remaining line segment
Maximum one；Finally calculate the angle of this line segment；
D5 reading) is calculated；Scale limitation is, then by calculating, obtain
Go out total indicator reading.
Further, the step 3）Described in numeral identification comprise the following steps that：
E1) with regionFor object, gradation conversion, homomorphic filtering processing and piecewise linear transform processing are carried out；Homomorphic filtering is led to
Cross reflected light and incident light separation, incident light therein is filtered, so as to reduce the influence that illumination is caused to image；By same
Picture contrast after state filtering process is low, using piecewise linear transform using the gray average of image as boundary, improves whole figure
The contrast of picture；Carry out gray proces, filtering process, piecewise linear transform, binaryzation；Image binaryzation；Using between maximum kind
Variance method calculates binary-state threshold, then draws binary image；
E2) individual digit is split；Assuming that the resolution ratio of image is, the image is progressively scanned, one is obtained and containsIt is individual
The one-dimension array of element, wherein, theThe value of individual element representsCapable non-zero pixel number；Scan by column, obtain
Contain to oneThe one-dimension array of individual element, wherein, theThe value of individual element representsThe non-zero pixel of row
Number；AskTwo local minimum positions, retainWithInterior zone, remaining region
It is considered as border；AskLocal minimum, multiple positions can be obtained, according to position digital in instrument
Confidence ceases, and filters out four points for representing digital boundary；Finally, numeral is split by following rectangular information：
With；
E3) the numeral identification based on convolutional neural networks；The structure of convolutional neural networks uses internal 5 Rotating fields, i.e. except defeated
Enter, outside output layer, choose two layers of convolutional layer, two layers of pond layer and one layer of full linking layer；Wherein, input layer will use gray-scale map to make
For input vector, output layer is classified using many classification functions of softmax；Training sample directly uses the number of projection localization
Word, the sample size of each numeral is at least 50；During numeral identification, divided direct using the neutral net trained
Class, completes the Recognition of Reading of numeric area.
The identification of lightning-arrest instruments and meters is divided into region segmentation and Recognition of Reading two parts by the present invention.Examined first with connected domain
Instrument dial plate is divided into multiple regions by method of determining and calculating and rectangle fitting algorithm；Then these regions, mistake are detected using priori
Nonsensical region is filtered, retains the connected domain for representing pointer area and numeric area.Two connected domains are carried out most again
Small area rectangle fitting, obtains two rectangles with deflection angle；Rotation correction is carried out to image according to angle, then from school
Pointer and numeric area are partitioned into positive image.Finally, be utilized respectively preset angle configuration and convolutional neural networks method to pointer area and
Numeric area carries out Recognition of Reading.
The present invention can complete total indicator reading identification and digital Recognition of Reading simultaneously, in addition, the present invention can also overcome room
The harmful effect that external environment is caused to Instrument image, with good adaptability；Meanwhile, it is capable to carry out effective school to image
Just, the accuracy of reading is improved.
Brief description of the drawings
It is described in further detail below in conjunction with accompanying drawing and embodiments of the present invention
Fig. 1 is system block diagram of the invention；
Fig. 2 is region segmentation flow chart of the invention；
Fig. 3 is Recognition of Reading flow chart of the invention.
Embodiment
Referring to accompanying drawing.The automatic identifying method of a kind of arrester meter reading described in the present embodiment, while completing pointer
The recognition methods of Recognition of Reading and digital Recognition of Reading.Two readings are completed using connected domain detection algorithm and rectangle fitting algorithm
The segmentation in region, is then utilized respectively preset angle configuration and convolutional neural networks method and carries out Recognition of Reading to two regions.Overall structure
Figure is as shown in Figure 1.
Specifically include following steps：
1）Image preprocessing：
The Instrument image of input is usually the coloured image that resolution ratio is 1920 × 1080.The mistake pre-processed to input picture
Cheng Wei：
F1) image down sampling.Using gaussian pyramid down-sampled images, after processing, will obtain a resolution ratio is 960 × 540
Coloured image；
F2) coloured image is converted to gray level image；
F3) filtering process.Using gray level image as filtering object, medium filtering is carried out first, the salt-pepper noise in image is removed；So
The Gaussian noise in image is removed using gaussian filtering afterwards, while smooth fine edge.
2）Region segmentation
Region segmentation is to carry out later stage Recognition of Reading precondition, and the quality of its segmentation effect directly affects the effect of Recognition of Reading
Really.The object of the present invention is while having the dual area instrument of pointer area and numeric area, according to the spy in Meter recognition region
Point, determines that the idiographic flow of image segmentation is as follows：
a1）Edge extracting and Morphological scale-space；Rim detection is carried out using Canny operators, a bianry image, Ran Houli is obtained
The edge of fracture is connected with expansion process；
a2）Connected domain is detected；Find as syntople and mark the connected domain in image, idiographic flow in 8 fields using pixel
For：
b1）Image is progressively scanned, pixel value in each row formed a whole for 255 pixel, and marks
Its line number, mark number, starting point, terminal, mark number is entered as by the pixel value of this all pixels point integrally included；
b2）From the second row, whether the region for observing this line UNICOM intersects with the region of lastrow, should if intersecting
The mark number in region is changed to the mark number with his intersecting area, and the pixel value of the pixel covered is also changed to the value of mark number；
Otherwise it is constant；
b3）After image scanning terminates, the pixel value of all non-zero pixels points is updated；
b4）Again image is scanned, regards pixel value identical pixel as a connected domain, such entire image will be by
It is divided into N number of connected domain；
a3）Calculate the boundary rectangle of each connected domain；The positive rectangle of all pixels point comprising some connected domain is exactly this company
The boundary rectangle in logical domain；
a4）Filtering interfering connected domain；According to pointer area and numeric area in the relative position information of dial plate, one can be filtered out
A little incorrect extraneous rectangles such as area is too small, area is excessive, percent information gap is big, namely filter out underproof connection
Domain；If the resolution ratio of image is, the information of extraneous rectangle is：, wherein the first two information
The coordinate on rectangle summit is represented, latter two information represents the width and height held；When this rectangle meets following（I）、（II）、
（III）During either condition, the UNICOM domain corresponding to it is deleted；If remaining connected domain>2, the work of the percentage of occupancy at most
To retain item；
（I）
（II）
（III）
If will meet（I）、（II）、（III）After the UNICOM domain of middle either condition is filtered, remaining connected domain number is 2, then enters
Enter step a6）, otherwise, remaining connected domain is calculated as the following formula：, wherein in molecules present connected domain
The number of available point, denominator represents the area of extraneous rectangle；Result of calculation is ranked up, maximum of which two is selected, i.e.,
For remaining two connected domains；
a5）Minimum area rectangle fitting is carried out to remaining two connected domains；Minimum area rectangle is to refer to include connected domain
The minimum rectangle of the area of interior all pixels point, and this rectangle is likely to inclined；
If the central point of two rectangles of fitting is respectivelyWith, deflection angle is respectivelyWith, calculate；
a6）According toCorrecting colour images and bianry image；School be exactly based on spatial alternation original image is mapped as it is new
Image, original coordinates is converted to as the following formula new coordinate：Wherein, the matrix of two rows three row is referred to as affine matrix；
Instrument image timing only carries out rotation correction, without translating, therefore affine matrix is:；
When carrying out spatial alternation, the position corresponding to input picture is found in turn from the pixel of output image, if the picture of mapping
Vegetarian refreshments non-integer, then utilize the single order interpolation calculation pixel value；If the pixel of mapping exceeds the scope of input picture,
The point is entered as zero；
a7）Pointer area is partitioned into from the bianry image of correction, and it is partitioned into from the coloured image of correction pointer areaAnd numeric area.Specifically segmentation flow is：
C1) using the bianry image after correction as object, connected domain therein is detected, boundary rectangle and the calculating of connected domain is fitted
The center point coordinate of rectangle；
C2) by all center point coordinates and step a5）'sWithIt is compared, retains coordinate and most connect
Two near rectangles；
C3) it is partitioned into respectively from the image of correction。
3）Recognition of Reading
Lightning-arrest instruments and meters has pointer and digital two reading areas, and the present invention is known using Hough transform detection of straight lines and preset angle configuration
Other total indicator reading；Use sciagraphy segmentation numeral, neural network identification numeral；Complete the Recognition of Reading of whole instrument.Specifically
Flow is as follows.
What pointer was recognized comprises the following steps that：
D1 the boundary line angle of pointer area) is calculated；Utilize Hough transform detection zoneTwo boundary lines, calculate
The angle of two lines section；BecauseIt is the region split from bianry image, only includes the obvious line segment in border, do not have
The interference of pointer and graduation mark, can more accurately detect the two lines section for representing instrument border；
D2) to regionCarry out gradation conversion, homomorphic filtering processing and piecewise linear transform processing；Homomorphic filtering passes through instead
Light and incident light separation are penetrated, incident light therein is filtered, so as to reduce the influence that illumination is caused to image；By homomorphic filtering
Picture contrast after processing is low, using piecewise linear transform using the gray average of image as boundary, improves pair of whole image
Degree of ratio；
D3) image binaryzation；Binary-state threshold is calculated using maximum variance between clusters, binary image is then drawn；
D4 the deflection angle of pointer) is calculated；The line segment in binary image is detected first with Hough transform；Then line is screened
Section, in a plurality of line segment of detection, is filtered out and boundary lineThe close line segment in position, then select length from remaining line segment
Maximum one；Finally calculate the angle of this line segment；
D5 reading) is calculated；Scale limitation is, then by calculating, obtain
Go out total indicator reading.
What numeral was recognized comprises the following steps that：
E1) withTo picture, to carry out gray proces, filtering process, piecewise linear transform, binaryzation；The step recognized with pointer
Rapid d2) and step d3)；
E2) individual digit is split；Assuming that the resolution ratio of image is, the image is progressively scanned, one is obtained and containsIt is individual
The one-dimension array of element, wherein, theThe value of individual element representsCapable non-zero pixel number；Scan by column, obtain
Contain to oneThe one-dimension array of individual element, wherein, theThe value of individual element representsThe non-zero pixel of row
Number；AskTwo local minimum positions, retainWithInterior zone, remaining region is regarded
For border；AskLocal minimum, multiple positions can be obtained, believed according to position digital in instrument
Breath, filters out four points for representing digital boundary；Finally, numeral is split by following rectangular information：
With；
E3) the numeral identification based on convolutional neural networks；The structure of convolutional neural networks uses internal 5 Rotating fields, i.e. except defeated
Enter, outside output layer, choose two layers of convolutional layer, two layers of pond layer and one layer of full linking layer；Wherein, input layer will use gray-scale map to make
For input vector, output layer is classified using many classification functions of softmax；Training sample directly uses the number of projection localization
Word, the sample size of each numeral is at least 50；During numeral identification, divided direct using the neutral net trained
Class, completes the Recognition of Reading of numeric area.
Claims (5)
1. a kind of automatic identifying method of arrester meter reading, it is characterised in that：Comprise the following steps：
1）Image preprocessing：Using gaussian pyramid down-sampled images, the Instrument image of input is reduced into resolution ratio, then by coloured silk
Color image is converted to gray level image, and is filtered to gray level image processing；
2）Region segmentation：According to the characteristics of Meter recognition region, determine that the idiographic flow of image segmentation is as follows：
a1）Edge extracting and Morphological scale-space；Rim detection is carried out using Canny operators, a bianry image, Ran Houli is obtained
The edge of fracture is connected with expansion process；
a2）Connected domain is detected；8 fields using pixel are found as syntople and mark the connected domain in image, and entire image will
It is divided into N number of connected domain；
a3）Calculate the boundary rectangle of each connected domain；The positive rectangle of all pixels point comprising some connected domain is exactly this company
The boundary rectangle in logical domain；
a4）Filtering interfering connected domain；According to pointer area and numeric area in the relative position information of dial plate, one can be filtered out
A little incorrect extraneous rectangles such as area is too small, area is excessive, percent information gap is big, namely filter out underproof connection
Domain；If the resolution ratio of image is, the information of extraneous rectangle is：, wherein the first two information represents square
The coordinate on shape summit, latter two information represents the width and height held；When this rectangle meets following（I）、（II）、（III）
During either condition, the UNICOM domain corresponding to it is deleted；If remaining connected domain>2, the conduct that the percentage of occupancy is most retains
；
（I）
（II）
（III）
If will meet（I）、（II）、（III）After the UNICOM domain of middle either condition is filtered, remaining connected domain number is 2, then enters
Enter step a6）, otherwise, remaining connected domain is calculated as the following formula：, wherein have in molecules present connected domain
The number of point is imitated, denominator represents the area of extraneous rectangle；Result of calculation is ranked up, maximum of which two is selected, is
Remaining two connected domains；
a5）Minimum area rectangle fitting is carried out to remaining two connected domains；Minimum area rectangle is to refer to include connected domain
The minimum rectangle of the area of interior all pixels point, and this rectangle is likely to inclined；
If the central point of two rectangles of fitting is respectivelyWith, deflection angle is respectivelyWith, calculate；
a6）According toCorrecting colour images and bianry image；School is exactly based on spatial alternation and original image is mapped as to new figure
Picture, original coordinates is converted to as the following formula new coordinate：Wherein, the matrix of two rows three row is referred to as affine matrix；
Instrument image timing only carries out rotation correction, without translating, therefore affine matrix is:；
When carrying out spatial alternation, the position corresponding to input picture is found in turn from the pixel of output image, if the pixel of mapping
Point non-integer, then utilize the single order interpolation calculation pixel value；, will if the pixel of mapping exceeds the scope of input picture
The point is entered as zero；
a7）Pointer area is partitioned into from the bianry image of correction, and it is partitioned into from the coloured image of correction pointer areaAnd numeric area；
3）Recognition of Reading：Lightning-arrest instruments and meters has pointer and digital two reading areas, uses Hough transform detection of straight lines and angle
Method recognizes total indicator reading；Use sciagraphy segmentation numeral, neural network identification numeral；Complete the Recognition of Reading of whole instrument.
2. a kind of automatic identifying method of arrester meter reading as claimed in claim 1, it is characterised in that：The step 2）
Middle a2）Described connected domain detection is concretely comprised the following steps：
b1）Image is progressively scanned, pixel value in each row formed a whole for 255 pixel, and marks
Its line number, mark number, starting point, terminal, mark number is entered as by the pixel value of this all pixels point integrally included；
b2）From the second row, whether the region for observing this line UNICOM intersects with the region of lastrow, should if intersecting
The mark number in region is changed to the mark number with his intersecting area, and the pixel value of the pixel covered is also changed to the value of mark number；
Otherwise it is constant；
b3）After image scanning terminates, the pixel value of all non-zero pixels points is updated；
b4）Again image is scanned, regards pixel value identical pixel as a connected domain, such entire image will be by
It is divided into N number of connected domain.
3. a kind of automatic identifying method of arrester meter reading as claimed in claim 1, it is characterised in that：The step 2）
Middle a7）Described is partitioned into pointer area, and it is partitioned into from the coloured image of correction pointer areaAnd numeric area's
Specifically segmentation step is：
C1) using the bianry image after correction as object, connected domain therein is detected, boundary rectangle and the calculating of connected domain is fitted
The center point coordinate of rectangle；
C2) by all center point coordinates and step a5）'sWithIt is compared, retains coordinate immediate
Two rectangles；
C3) it is partitioned into respectively from the image of correction。
4. a kind of automatic identifying method of arrester meter reading as claimed in claim 1, it is characterised in that：The step 3）
Described in pointer identification comprise the following steps that：
D1 the boundary line angle of pointer area) is calculated；Utilize Hough transform detection zoneTwo boundary lines, calculate
The angle of two lines section；BecauseIt is the region split from bianry image, only includes the obvious line segment in border, do not have
The interference of pointer and graduation mark, can more accurately detect the two lines section for representing instrument border；
D2) to regionCarry out gradation conversion, homomorphic filtering processing and piecewise linear transform processing；Homomorphic filtering will be by that will reflect
Light and incident light separation, filter incident light therein, so as to reduce the influence that illumination is caused to image；At homomorphic filtering
Picture contrast after reason is low, using piecewise linear transform using the gray average of image as boundary, improves the contrast of whole image
Degree；
D3) image binaryzation；Binary-state threshold is calculated using maximum variance between clusters, binary image is then drawn；
D4 the deflection angle of pointer) is calculated；The line segment in binary image is detected first with Hough transform；Then line is screened
Section, in a plurality of line segment of detection, is filtered out and boundary lineThe close line segment in position, then select length from remaining line segment
Maximum one；Finally calculate the angle of this line segment；
D5 reading) is calculated；Scale limitation is, then by calculating, obtain
Go out total indicator reading.
5. a kind of automatic identifying method of arrester meter reading as claimed in claim 1, it is characterised in that：The step 3）
Described in numeral identification comprise the following steps that：
E1) with regionTo picture, to carry out gradation conversion, homomorphic filtering processing and piecewise linear transform processing；Homomorphic filtering is led to
Cross reflected light and incident light separation, incident light therein is filtered, so as to reduce the influence that illumination is caused to image；By same
Picture contrast after state filtering process is low, using piecewise linear transform using the gray average of image as boundary, improves whole figure
The contrast of picture；Carry out gray proces, filtering process, piecewise linear transform, binaryzation；Image binaryzation；Using between maximum kind
Variance method calculates binary-state threshold, then draws binary image；
E2) individual digit is split；Assuming that the resolution ratio of image is, the image is progressively scanned, one is obtained and containsIndividual member
The one-dimension array of element, wherein, theThe value of individual element representsCapable non-zero pixel number；Scan by column, obtain
One containsThe one-dimension array of individual element, wherein, theThe value of individual element representsThe non-zero pixel number of row；
AskTwo local minimum positions, retainWithInterior zone, remaining region is considered as
Border；AskLocal minimum, multiple positions can be obtained, according to positional information digital in instrument,
Filter out four points for representing digital boundary；Finally, numeral is split by following rectangular information：
With；
E3) the numeral identification based on convolutional neural networks；The structure of convolutional neural networks uses internal 5 Rotating fields, i.e. except defeated
Enter, outside output layer, choose two layers of convolutional layer, two layers of pond layer and one layer of full linking layer；Wherein, input layer will use gray-scale map to make
For input vector, output layer is classified using many classification functions of softmax；Training sample directly uses the number of projection localization
Word, the sample size of each numeral is at least 50；During numeral identification, divided direct using the neutral net trained
Class, completes the Recognition of Reading of numeric area.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710239497.6A CN106951900B (en) | 2017-04-13 | 2017-04-13 | A kind of automatic identifying method of arrester meter reading |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710239497.6A CN106951900B (en) | 2017-04-13 | 2017-04-13 | A kind of automatic identifying method of arrester meter reading |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN106951900A true CN106951900A (en) | 2017-07-14 |
| CN106951900B CN106951900B (en) | 2019-10-22 |
Family
ID=59475911
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710239497.6A Active CN106951900B (en) | 2017-04-13 | 2017-04-13 | A kind of automatic identifying method of arrester meter reading |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN106951900B (en) |
Cited By (34)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107766836A (en) * | 2017-11-07 | 2018-03-06 | 国网黑龙江省电力有限公司检修公司 | A kind of circular deflection readings of pointer type meters antidote |
| CN108009535A (en) * | 2017-11-21 | 2018-05-08 | 武汉中元华电科技股份有限公司 | A kind of simple pointer meter reading method based on machine vision |
| CN108345889A (en) * | 2018-02-27 | 2018-07-31 | 国网上海市电力公司 | A kind of application process carrying out registration identification to communication cabinet using Raspberry Pi |
| CN108388894A (en) * | 2017-12-26 | 2018-08-10 | 新智数字科技有限公司 | A kind of recognition methods, device and the equipment of number meter reading |
| CN108491842A (en) * | 2018-03-27 | 2018-09-04 | 康体佳智能科技（深圳）有限公司 | A kind of dial plate identifying system and recognition methods based on neural network |
| CN108562821A (en) * | 2018-05-08 | 2018-09-21 | 中国电力科学研究院有限公司 | A kind of method and system determining Single-phase Earth-fault Selection in Distribution Systems based on Softmax |
| CN108573261A (en) * | 2018-04-17 | 2018-09-25 | 国家电网公司 | A kind of read out instrument recognition methods suitable for Intelligent Mobile Robot |
| CN108764253A (en) * | 2018-05-15 | 2018-11-06 | 杭州电子科技大学 | Pointer instrument digitizing solution |
| CN108764234A (en) * | 2018-05-10 | 2018-11-06 | 浙江理工大学 | A kind of liquid level instrument Recognition of Reading method based on crusing robot |
| CN108921203A (en) * | 2018-06-13 | 2018-11-30 | 深圳市云识科技有限公司 | A kind of detection and recognition methods of pointer-type water meter |
| CN109146806A (en) * | 2018-07-29 | 2019-01-04 | 国网上海市电力公司 | Gauge pointer position detection recognition methods based on shadow removing optimization in remote monitoriong of electric power |
| CN109271985A (en) * | 2018-09-07 | 2019-01-25 | 广东中粤电力科技有限公司 | A kind of digital instrument reading image-recognizing method and system |
| CN109376749A (en) * | 2018-05-22 | 2019-02-22 | 国网山东省电力公司电力科学研究院 | Power transmission and transforming equipment infrared image temperature wide scope recognition methods based on deep learning |
| CN109389150A (en) * | 2018-08-28 | 2019-02-26 | 东软集团股份有限公司 | Image consistency comparison method, device, storage medium and electronic equipment |
| CN109977944A (en) * | 2019-02-21 | 2019-07-05 | 杭州朗阳科技有限公司 | A kind of recognition methods of digital water meter reading |
| CN110110697A (en) * | 2019-05-17 | 2019-08-09 | 山东省计算中心（国家超级计算济南中心） | More fingerprint segmentation extracting methods, system, equipment and medium based on direction correction |
| CN110135420A (en) * | 2019-05-16 | 2019-08-16 | 北京灵汐科技有限公司 | Dial plate state identification method and device, readable storage medium storing program for executing and electronic equipment |
| CN110210477A (en) * | 2019-05-24 | 2019-09-06 | 四川阿泰因机器人智能装备有限公司 | A kind of digital instrument Recognition of Reading method |
| CN110414510A (en) * | 2019-07-26 | 2019-11-05 | 华中科技大学 | A kind of readings of pointer type meters bearing calibration |
| CN110555839A (en) * | 2019-09-06 | 2019-12-10 | 腾讯云计算（北京）有限责任公司 | Defect detection and identification method and device, computer equipment and storage medium |
| CN111091100A (en) * | 2019-12-21 | 2020-05-01 | 河海大学 | Concrete crack identification method based on complex noise image deep learning |
| CN111325164A (en) * | 2020-02-25 | 2020-06-23 | 北京眸视科技有限公司 | Pointer indication number identification method and device and electronic equipment |
| CN111444781A (en) * | 2020-03-09 | 2020-07-24 | 武汉理工大学 | Water meter reading identification method and equipment and storage medium |
| CN111476787A (en) * | 2020-04-23 | 2020-07-31 | 中科开创（广州）智能科技发展有限公司 | Automatic reading method and device for adaptive distortion of pointer meter |
| CN111598094A (en) * | 2020-05-27 | 2020-08-28 | 深圳市铁越电气有限公司 | Deep learning-based angle regression meter reading identification method, device and system |
| CN111868740A (en) * | 2018-03-23 | 2020-10-30 | 株式会社东芝 | Reading system, reading method and storage medium |
| CN112668578A (en) * | 2020-12-31 | 2021-04-16 | 中广核研究院有限公司 | Pointer instrument reading method and device, computer equipment and storage medium |
| CN112906602A (en) * | 2021-03-04 | 2021-06-04 | 杭州电力设备制造有限公司 | Automatic identification device and identification method for electricity meter of power distribution cabinet based on image processing |
| CN113159027A (en) * | 2021-04-13 | 2021-07-23 | 杭州电子科技大学 | Seven-segment type digital display instrument identification method based on minimum external rectangle variant |
| CN113469162A (en) * | 2021-06-02 | 2021-10-01 | 广东白云学院 | Pointer instrument reading method, device, equipment and medium based on double-scale segmentation |
| CN113642437A (en) * | 2021-08-03 | 2021-11-12 | 中国地质大学（北京） | Quantitative calculation method for content and radius of different components in coal |
| CN116612118A (en) * | 2023-07-19 | 2023-08-18 | 中建五局第三建设有限公司 | Artificial intelligence-based quality detection and evaluation method for building lightning arrester |
| CN117095246A (en) * | 2023-10-20 | 2023-11-21 | 国网江西省电力有限公司超高压分公司 | Polarization imaging-based deep learning pointer instrument reading identification method |
| CN117576800A (en) * | 2023-10-25 | 2024-02-20 | 大唐海口清洁能源发电有限责任公司 | Automatic inspection method and device for thermal power plant, inspection robot and storage medium |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102975826A (en) * | 2012-12-03 | 2013-03-20 | 上海海事大学 | Portable ship water gauge automatic detection and identification method based on machine vision |
| CN103914843A (en) * | 2014-04-04 | 2014-07-09 | 上海交通大学 | Image segmentation method based on watershed algorithm and morphological marker |
| CN103955694A (en) * | 2014-04-09 | 2014-07-30 | 广州邦讯信息系统有限公司 | Image recognition meter reading system and method |
| CN105203148A (en) * | 2015-09-30 | 2015-12-30 | 湖北工业大学 | Visual detection method for automobile combination instrument |
| CN105303168A (en) * | 2015-10-14 | 2016-02-03 | 南京第五十五所技术开发有限公司 | Multi-view pointer type instrument identification method and device |
| CN105740829A (en) * | 2016-02-02 | 2016-07-06 | 暨南大学 | Scanning line processing based automatic reading method for pointer instrument |
| CN106529519A (en) * | 2016-09-19 | 2017-03-22 | 国家电网公司 | Automatic number identification method and system of power pointer type instrument |
-
2017
- 2017-04-13 CN CN201710239497.6A patent/CN106951900B/en active Active
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102975826A (en) * | 2012-12-03 | 2013-03-20 | 上海海事大学 | Portable ship water gauge automatic detection and identification method based on machine vision |
| CN103914843A (en) * | 2014-04-04 | 2014-07-09 | 上海交通大学 | Image segmentation method based on watershed algorithm and morphological marker |
| CN103955694A (en) * | 2014-04-09 | 2014-07-30 | 广州邦讯信息系统有限公司 | Image recognition meter reading system and method |
| CN105203148A (en) * | 2015-09-30 | 2015-12-30 | 湖北工业大学 | Visual detection method for automobile combination instrument |
| CN105303168A (en) * | 2015-10-14 | 2016-02-03 | 南京第五十五所技术开发有限公司 | Multi-view pointer type instrument identification method and device |
| CN105740829A (en) * | 2016-02-02 | 2016-07-06 | 暨南大学 | Scanning line processing based automatic reading method for pointer instrument |
| CN106529519A (en) * | 2016-09-19 | 2017-03-22 | 国家电网公司 | Automatic number identification method and system of power pointer type instrument |
Non-Patent Citations (1)
| Title |
|---|
| 崔家瑞等: "基于图像的换流站用避雷器仪表识别", 《电工技术学报》 * |
Cited By (52)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107766836A (en) * | 2017-11-07 | 2018-03-06 | 国网黑龙江省电力有限公司检修公司 | A kind of circular deflection readings of pointer type meters antidote |
| CN108009535A (en) * | 2017-11-21 | 2018-05-08 | 武汉中元华电科技股份有限公司 | A kind of simple pointer meter reading method based on machine vision |
| CN108388894A (en) * | 2017-12-26 | 2018-08-10 | 新智数字科技有限公司 | A kind of recognition methods, device and the equipment of number meter reading |
| CN108345889B (en) * | 2018-02-27 | 2022-02-11 | 国网上海市电力公司 | Application method for performing reading identification on communication cabinet by utilizing raspberry party |
| CN108345889A (en) * | 2018-02-27 | 2018-07-31 | 国网上海市电力公司 | A kind of application process carrying out registration identification to communication cabinet using Raspberry Pi |
| CN111868740A (en) * | 2018-03-23 | 2020-10-30 | 株式会社东芝 | Reading system, reading method and storage medium |
| CN108491842A (en) * | 2018-03-27 | 2018-09-04 | 康体佳智能科技（深圳）有限公司 | A kind of dial plate identifying system and recognition methods based on neural network |
| CN108573261A (en) * | 2018-04-17 | 2018-09-25 | 国家电网公司 | A kind of read out instrument recognition methods suitable for Intelligent Mobile Robot |
| CN108562821A (en) * | 2018-05-08 | 2018-09-21 | 中国电力科学研究院有限公司 | A kind of method and system determining Single-phase Earth-fault Selection in Distribution Systems based on Softmax |
| CN108562821B (en) * | 2018-05-08 | 2021-09-28 | 中国电力科学研究院有限公司 | Method and system for determining single-phase earth fault line selection of power distribution network based on Softmax |
| CN108764234A (en) * | 2018-05-10 | 2018-11-06 | 浙江理工大学 | A kind of liquid level instrument Recognition of Reading method based on crusing robot |
| CN108764234B (en) * | 2018-05-10 | 2021-10-12 | 浙江理工大学 | Liquid level meter reading identification method based on inspection robot |
| CN108764253A (en) * | 2018-05-15 | 2018-11-06 | 杭州电子科技大学 | Pointer instrument digitizing solution |
| CN109376749A (en) * | 2018-05-22 | 2019-02-22 | 国网山东省电力公司电力科学研究院 | Power transmission and transforming equipment infrared image temperature wide scope recognition methods based on deep learning |
| CN109376749B (en) * | 2018-05-22 | 2021-03-19 | 国网山东省电力公司电力科学研究院 | Power transmission and transformation equipment infrared image temperature wide range identification method based on deep learning |
| CN108921203A (en) * | 2018-06-13 | 2018-11-30 | 深圳市云识科技有限公司 | A kind of detection and recognition methods of pointer-type water meter |
| CN109146806A (en) * | 2018-07-29 | 2019-01-04 | 国网上海市电力公司 | Gauge pointer position detection recognition methods based on shadow removing optimization in remote monitoriong of electric power |
| CN109389150A (en) * | 2018-08-28 | 2019-02-26 | 东软集团股份有限公司 | Image consistency comparison method, device, storage medium and electronic equipment |
| CN109389150B (en) * | 2018-08-28 | 2022-04-05 | 东软集团股份有限公司 | Image consistency comparison method and device, storage medium and electronic equipment |
| CN109271985A (en) * | 2018-09-07 | 2019-01-25 | 广东中粤电力科技有限公司 | A kind of digital instrument reading image-recognizing method and system |
| CN109977944A (en) * | 2019-02-21 | 2019-07-05 | 杭州朗阳科技有限公司 | A kind of recognition methods of digital water meter reading |
| CN109977944B (en) * | 2019-02-21 | 2023-08-01 | 杭州朗阳科技有限公司 | Digital water meter reading identification method |
| CN110135420A (en) * | 2019-05-16 | 2019-08-16 | 北京灵汐科技有限公司 | Dial plate state identification method and device, readable storage medium storing program for executing and electronic equipment |
| CN110135420B (en) * | 2019-05-16 | 2021-11-16 | 北京灵汐科技有限公司 | Dial plate state identification method and device, readable storage medium and electronic equipment |
| CN110110697A (en) * | 2019-05-17 | 2019-08-09 | 山东省计算中心（国家超级计算济南中心） | More fingerprint segmentation extracting methods, system, equipment and medium based on direction correction |
| CN110210477A (en) * | 2019-05-24 | 2019-09-06 | 四川阿泰因机器人智能装备有限公司 | A kind of digital instrument Recognition of Reading method |
| CN110210477B (en) * | 2019-05-24 | 2023-03-24 | 四川阿泰因机器人智能装备有限公司 | Digital instrument reading identification method |
| CN110414510A (en) * | 2019-07-26 | 2019-11-05 | 华中科技大学 | A kind of readings of pointer type meters bearing calibration |
| CN110414510B (en) * | 2019-07-26 | 2021-10-08 | 华中科技大学 | Reading correction method for pointer instrument |
| CN110555839A (en) * | 2019-09-06 | 2019-12-10 | 腾讯云计算（北京）有限责任公司 | Defect detection and identification method and device, computer equipment and storage medium |
| CN111091100A (en) * | 2019-12-21 | 2020-05-01 | 河海大学 | Concrete crack identification method based on complex noise image deep learning |
| CN111325164B (en) * | 2020-02-25 | 2023-11-21 | 北京眸视科技有限公司 | Pointer representation number identification method and device and electronic equipment |
| CN111325164A (en) * | 2020-02-25 | 2020-06-23 | 北京眸视科技有限公司 | Pointer indication number identification method and device and electronic equipment |
| CN111444781B (en) * | 2020-03-09 | 2023-08-29 | 武汉理工大学 | Water meter reading identification method, device and storage medium |
| CN111444781A (en) * | 2020-03-09 | 2020-07-24 | 武汉理工大学 | Water meter reading identification method and equipment and storage medium |
| CN111476787A (en) * | 2020-04-23 | 2020-07-31 | 中科开创（广州）智能科技发展有限公司 | Automatic reading method and device for adaptive distortion of pointer meter |
| CN111598094B (en) * | 2020-05-27 | 2023-08-18 | 深圳市铁越电气有限公司 | Angle regression instrument reading identification method, equipment and system based on deep learning |
| CN111598094A (en) * | 2020-05-27 | 2020-08-28 | 深圳市铁越电气有限公司 | Deep learning-based angle regression meter reading identification method, device and system |
| CN112668578B (en) * | 2020-12-31 | 2023-11-07 | 中广核研究院有限公司 | Pointer type instrument reading method, pointer type instrument reading device, computer equipment and storage medium |
| CN112668578A (en) * | 2020-12-31 | 2021-04-16 | 中广核研究院有限公司 | Pointer instrument reading method and device, computer equipment and storage medium |
| CN112906602A (en) * | 2021-03-04 | 2021-06-04 | 杭州电力设备制造有限公司 | Automatic identification device and identification method for electricity meter of power distribution cabinet based on image processing |
| CN112906602B (en) * | 2021-03-04 | 2023-08-25 | 杭州电力设备制造有限公司 | Automatic identification device and identification method for electric quantity meter of power distribution cabinet based on image processing |
| CN113159027A (en) * | 2021-04-13 | 2021-07-23 | 杭州电子科技大学 | Seven-segment type digital display instrument identification method based on minimum external rectangle variant |
| CN113159027B (en) * | 2021-04-13 | 2024-02-09 | 杭州电子科技大学 | Seven-segment digital display instrument identification method based on minimum external rectangular variant |
| CN113469162B (en) * | 2021-06-02 | 2023-09-26 | 广东白云学院 | Pointer instrument identification method, device, equipment and medium based on double-scale segmentation |
| CN113469162A (en) * | 2021-06-02 | 2021-10-01 | 广东白云学院 | Pointer instrument reading method, device, equipment and medium based on double-scale segmentation |
| CN113642437A (en) * | 2021-08-03 | 2021-11-12 | 中国地质大学（北京） | Quantitative calculation method for content and radius of different components in coal |
| CN113642437B (en) * | 2021-08-03 | 2022-05-31 | 中国地质大学（北京） | Quantitative calculation method for content and radius of different components in coal |
| CN116612118A (en) * | 2023-07-19 | 2023-08-18 | 中建五局第三建设有限公司 | Artificial intelligence-based quality detection and evaluation method for building lightning arrester |
| CN116612118B (en) * | 2023-07-19 | 2023-10-03 | 中建五局第三建设有限公司 | Artificial intelligence-based quality detection and evaluation method for building lightning arrester |
| CN117095246A (en) * | 2023-10-20 | 2023-11-21 | 国网江西省电力有限公司超高压分公司 | Polarization imaging-based deep learning pointer instrument reading identification method |
| CN117576800A (en) * | 2023-10-25 | 2024-02-20 | 大唐海口清洁能源发电有限责任公司 | Automatic inspection method and device for thermal power plant, inspection robot and storage medium |
Also Published As
| Publication number | Publication date |
|---|---|
| CN106951900B (en) | 2019-10-22 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN106951900B (en) | A kind of automatic identifying method of arrester meter reading | |
| CN106529537B (en) | A kind of digital instrument reading image-recognizing method | |
| CN111626190B (en) | Water level monitoring method for scale recognition based on clustering partition | |
| CN112149667B (en) | Automatic reading method of pointer instrument based on deep learning | |
| CN110363182B (en) | Deep learning-based lane line detection method | |
| CN112699876B (en) | Automatic reading method for various meters of gas collecting station | |
| CN107590498A (en) | A kind of self-adapted car instrument detecting method based on Character segmentation level di- grader | |
| CN112257676A (en) | Pointer instrument reading method and system and inspection robot | |
| CN109544497A (en) | Image interfusion method and electronic equipment for transmission line faultlocating | |
| CN109636784A (en) | Saliency object detection method based on maximum neighborhood and super-pixel segmentation | |
| CN104077577A (en) | Trademark detection method based on convolutional neural network | |
| CN109858480A (en) | Digital instrument identification method | |
| CN110276285A (en) | A kind of shipping depth gauge intelligent identification Method in uncontrolled scene video | |
| CN111126253A (en) | Knife switch state detection method based on image recognition | |
| CN114549993B (en) | Method, system and device for grading line segment image in experiment and readable storage medium | |
| CN111598098A (en) | Water gauge water line detection and effectiveness identification method based on full convolution neural network | |
| CN109255279A (en) | A kind of method and system of road traffic sign detection identification | |
| CN115359239A (en) | Wind power blade defect detection and positioning method and device, storage medium and electronic equipment | |
| CN113592839B (en) | Distribution network line typical defect diagnosis method and system based on improved fast RCNN | |
| CN116188756A (en) | Instrument angle correction and indication recognition method based on deep learning | |
| CN112734729A (en) | Water gauge water level line image detection method and device suitable for night light supplement condition and storage medium | |
| CN113947563A (en) | Cable process quality dynamic defect detection method based on deep learning | |
| CN111291818B (en) | Non-uniform class sample equalization method for cloud mask | |
| CN116188755A (en) | Instrument angle correction and reading recognition device based on deep learning | |
| CN115019294A (en) | Pointer instrument reading identification method and system |
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 |