CN115409803A - Oil level identification method, system and medium for transformer in power distribution room - Google Patents
Oil level identification method, system and medium for transformer in power distribution room Download PDFInfo
- Publication number
- CN115409803A CN115409803A CN202211053186.8A CN202211053186A CN115409803A CN 115409803 A CN115409803 A CN 115409803A CN 202211053186 A CN202211053186 A CN 202211053186A CN 115409803 A CN115409803 A CN 115409803A
- Authority
- CN
- China
- Prior art keywords
- oil level
- image
- pointer
- transformer
- distribution room
- 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
- 238000000034 method Methods 0.000 title claims abstract description 68
- 238000009826 distribution Methods 0.000 title claims abstract description 29
- 238000001914 filtration Methods 0.000 claims abstract description 18
- 238000007781 pre-processing Methods 0.000 claims abstract description 18
- 238000012545 processing Methods 0.000 claims description 17
- 238000001514 detection method Methods 0.000 claims description 7
- 230000002159 abnormal effect Effects 0.000 claims description 6
- 230000007797 corrosion Effects 0.000 claims description 6
- 238000005260 corrosion Methods 0.000 claims description 6
- 238000003860 storage Methods 0.000 claims description 5
- 230000000877 morphologic effect Effects 0.000 claims description 4
- 238000000605 extraction Methods 0.000 claims description 3
- 238000012549 training Methods 0.000 claims description 3
- 238000005286 illumination Methods 0.000 abstract description 11
- 230000008569 process Effects 0.000 abstract description 5
- 230000005856 abnormality Effects 0.000 abstract description 2
- 230000009466 transformation Effects 0.000 abstract 1
- 238000004364 calculation method Methods 0.000 description 8
- 238000010586 diagram Methods 0.000 description 4
- 230000006870 function Effects 0.000 description 3
- 238000012544 monitoring process Methods 0.000 description 3
- 230000008859 change Effects 0.000 description 2
- 238000006243 chemical reaction Methods 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 230000011218 segmentation Effects 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000005540 biological transmission Effects 0.000 description 1
- 238000005282 brightening Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000003384 imaging method Methods 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 238000007689 inspection Methods 0.000 description 1
- 238000012423 maintenance Methods 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/0002—Inspection of images, e.g. flaw detection
- G06T7/0004—Industrial image inspection
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T5/00—Image enhancement or restoration
- G06T5/73—Deblurring; Sharpening
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/12—Edge-based segmentation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/13—Edge detection
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/60—Analysis of geometric attributes
- G06T7/62—Analysis of geometric attributes of area, perimeter, diameter or volume
-
- 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
- G06V10/25—Determination of region of interest [ROI] or a volume of interest [VOI]
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/20—Special algorithmic details
- G06T2207/20024—Filtering details
- G06T2207/20032—Median filtering
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/20—Special algorithmic details
- G06T2207/20081—Training; Learning
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/20—Special algorithmic details
- G06T2207/20092—Interactive image processing based on input by user
- G06T2207/20104—Interactive definition of region of interest [ROI]
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Geometry (AREA)
- Quality & Reliability (AREA)
- Multimedia (AREA)
- Image Analysis (AREA)
Abstract
The invention discloses a method, a system and a medium for distinguishing the oil level of a transformer in a power distribution room, wherein the method comprises the steps of obtaining an oil level image of the transformer and preprocessing the image; positioning a pointer of the transformer oil meter to obtain the axis position and the pointer position of the oil level gauge; identifying a starting point and an end point of the oil level gauge to obtain an oil level main reading; and judging the result, and if the abnormality is identified, outputting alarm information. The method performs homomorphic filtering and median filtering on the oil level image to eliminate the influence of illumination and noise on the image acquisition process, and improves the adaptability to different illuminations; then, hough transformation is carried out on the axis center position of the oil level image to detect the pointer, so that global edge points are prevented from being searched, the calculated amount and the time for program operation are greatly reduced, and the searching accuracy is improved; finally, the oil level reading can be read according to the angle method. The method has the advantages of high accuracy, short program running time, adaptability to various illumination conditions and high robustness.
Description
Technical Field
The invention belongs to the technical field of transformer oil level detection, and particularly relates to a method, a system and a medium for distinguishing an oil level of a transformer in a power distribution room.
Background
Whether the transformer equipment can normally operate or not as important equipment in the power system plays a crucial role in the safety of the whole power system. The problems of low efficiency and poor accuracy of manual collection of the transformer oil level in inspection are solved, and the situations of false detection, missing detection and the like are easy to occur. Meanwhile, the transformer information provided by maintenance has hysteresis, is not used for management, and the oil level state of the transformer cannot be known and mastered clearly. The transformer oil level state is identified through the monitoring video in the power distribution room, whether the oil level meets the safe operation requirement is judged, when the oil level reaches a red warning position, early warning is timely discovered and made, a worker is reminded to check and get rid of problems through the technology, hidden dangers are restrained at the earliest stage, the potential safety hazards caused by too low oil level are reduced, and the safety production of the power distribution room is guaranteed.
In the prior art, for the identification of the transformer oil level image monitored by a video, the method mainly depends on target color characteristics and geometric characteristics, and the methods are often influenced by factors such as brightness change, complex background and the like, so that the method has poor generalization capability.
Disclosure of Invention
The invention mainly aims to overcome the defects of the prior art, and provides a method, a system and a medium for distinguishing the oil level of a transformer in a power distribution room, wherein an oil level monitoring video stream is accessed. And detecting an oil level meter, identifying an oil level identifier, and alarming the oil level from the oil level to a warning level to realize automatic monitoring and alarming all day long.
In order to achieve the purpose, the invention adopts the following technical scheme:
in one aspect of the present invention, there is provided a method for identifying an oil level of a transformer in a distribution room, comprising the steps of:
acquiring an oil level image of the transformer, and preprocessing the image;
positioning a pointer of the transformer oil meter to obtain the axis position and the pointer position of the oil level gauge;
identifying a starting point and an end point of the oil level indicator to obtain an oil level main reading;
and judging the result, and if the abnormality is identified, outputting alarm information.
As a preferred technical scheme, the acquiring an oil level image of a transformer and performing image preprocessing specifically includes:
acquiring an oil level image of the transformer, including acquiring an image of a shooting device and intercepting the image from a video; the image is intercepted from the video, the frame selection interval of the video is set according to the scene, the requirement and the comprehensive consideration of the performance, and the intercepted single-frame image is converted into the image in the JPG format which can be processed by the oil surface state recognition model;
image preprocessing, including homomorphic filtering, median filtering, binarization and morphological processing; and the data preprocessing operation is used for removing part of noise interference in the picture and enabling the training picture to be consistent with the picture to be predicted.
As a preferred technical solution, the homomorphic filtering specifically includes:
the oil level gauge image is subjected to a filtering process by a gaussian homomorphic filter H (μ, v) as follows:
wherein D (μ, v) is the distance of the point (μ, v) from the center of the frequency rectangle
The distance of (d); d o To cut off the frequency, the invention takes D o Median value of D (μ, v); the parameter c is used for controlling sharpening of a homomorphic filter, and the invention enables c =2; parameter gamma H And gamma L Controlling the variation range of H (mu, v), H (mu, v) epsilon [ gamma ] L ，γ H ]The present invention makes gamma H ＝2.0，γ L ＝0.3。
As a preferred technical solution, the binarization and image morphology processing specifically includes:
and distinguishing the pointer from other areas in the image by using a threshold value method, wherein the formula is as follows:
wherein f (x, y) is a binarized image processed by a thresholding method, and T is a set threshold;
and processing the binary image processed by the threshold method by using corrosion operation in morphology.
As a preferred technical scheme, the identification of the axis position of the oil level indicator adopts a method of traversing pointer pixels, and specifically comprises the following steps:
carrying out contour extraction operation on the image subjected to the image morphology processing: two pointers interrupted into two sections by the axis are framed by two minimum external rectangle frames, and the mass center coordinates (x) of the two rectangles are respectively obtained 1 ，y 1 )，(x 2 ，y 2 )；
Connecting the coordinates of the centers of mass of the two rectangles, and solving the slope k and intercept b of a straight line, wherein the equation of the straight line is y = k x + b;
traversing the pixels on the straight line, judging according to RGB, and obtaining the first non-black point (X) on the straight line 1 ，Y 1 ) And the last non-black dot (X) 2 ，Y 2 ) Finally, the center point is obtained to obtain the coordinates (X, Y) of the axis as follows:
X＝(X 1 +X 2 )/2
Y＝(Y 1 +Y 2 )/2。
as a preferred technical solution, the identification of the pointer position of the oil level indicator adopts an improved hough transform, specifically:
judging the position of the pointer approximately according to the first-order work of detecting the axis of the pointer;
utilizing a minimum rectangular frame to frame out the ROI where the pointer is located;
performing Hough transform within the ROI to detect a pointer straight line;
judging the detection result according to the characteristic that the pointer is the longest, and only outputting and displaying the longest straight line to obtain the coordinates of two end points of the straight line, namely the detected pointer;
judging the distance between two end points of the pointer and the axle center, and the end with larger distance is the coordinate (P) of the pointer head 1 ，P 2 )。
As a preferred technical solution, the positions of the starting point and the ending point of the oil level indicator are identified by a color division and outline area method, which specifically includes the following steps:
converting the picture from RGB to HSV color space;
according to the red H value, making a red mask, and performing beta _ and operation on the mask and the original image to obtain an image of the region of interest;
carrying out corrosion operation on the obtained interested region image, removing non-interested interference points, and then carrying out expansion operation to enlarge the edge of the red region;
marking the rectangles of the starting point and the end point by using a contour method, and obtaining the centroids of the rectangles, wherein the coordinates of the two centroids correspond to the position coordinates (K) of the starting point and the end point 1 ，D 1 )，(K 2 ，D 2 )。
As a preferred technical scheme, the main reading of the identification oil level is performed by adopting an angle method, which specifically comprises the following steps:
according to the axis coordinate (X, Y) and the pointer head coordinate (P) 1 ，P 2 ) Starting point coordinate (K) 1 ，D 1 ) And end point coordinates (K) 2 ，D 2 ) And calculating an angle alpha between scales 0 and 10 by a vector included angle method, and then:
the dial angle with scales is 360-alpha;
the included angle gamma between each scale mark of the oil level indicator is = (360-alpha)/10;
then according to the pointer head coordinate (P) 1 ，P 2 ) And starting point coordinates (K) 1 ，D 1 ) And end point coordinates (K) 2 ，D 2 ) And calculating an included angle beta between the pointer and the scale mark 0, and then the reading theta of the oil level indicator is as follows:
in another aspect of the present invention, a system for distinguishing an oil level of a transformer in a distribution room is provided, wherein the system for distinguishing an oil level of a transformer in a distribution room is applied to the method for distinguishing an oil level of a transformer in a distribution room, and comprises an image acquisition module, a pointer positioning module, an oil level reading acquisition module and an output module;
the image acquisition module is used for acquiring an oil level image of the transformer and carrying out image preprocessing;
the pointer positioning module is used for positioning a pointer of the transformer oil meter to obtain the axis position and the pointer position of the oil level gauge;
the oil level reading acquisition module is used for identifying a starting point and an end point of the oil level gauge to obtain an oil level main reading;
and the output module is used for judging the result and outputting alarm information if the result is identified to be abnormal.
In another aspect of the present invention, there is provided a storage medium storing a program, characterized in that: when the program is executed by the processor, the oil level identification method of the transformer in the power distribution room is realized.
Compared with the prior art, the invention has the following advantages and beneficial effects:
the oil level image is subjected to necessary image preprocessing operations such as homomorphic filtering, median filtering and the like so as to eliminate the influence of illumination and noise on the oil level image in the image acquisition process and improve the adaptability of the algorithm to different illuminations; then, pointer contour searching is carried out on the oil level image after binarization and morphological operation processing, and a pointer is detected by carrying out Hough transform in the region according to the axis position, so that global edge point searching is avoided, the calculation amount and the program running time are greatly reduced, and the searching accuracy is improved; converting the oil level image into an HSV color space, and dividing a starting point and an end point of the oil level according to the red H value because the starting point and the end point of the oil level are red, so as to obtain the coordinate positions of the starting point and the end point of the oil level; finally, the oil level reading can be read according to the angle method. The algorithm has high accuracy, short program running time, high robustness and can adapt to various illumination conditions.
Drawings
FIG. 1 is a flow chart of a method for identifying oil levels of transformers in a power distribution room according to an embodiment of the invention;
fig. 2 (a), 2 (b) and 2 (c) are schematic views illustrating the process of determining the axis of the oil level gauge according to the embodiment of the present invention;
fig. 3 is a schematic diagram of the principle of hough transform according to an embodiment of the present invention;
FIG. 4 is a schematic diagram of improving Hough transform for pointer location identification according to an embodiment of the present invention
FIG. 5 is a schematic diagram of an embodiment of a system for identifying transformer oil level in a power distribution room according to the present invention;
fig. 6 is a schematic structural diagram of a storage medium according to an embodiment of the present invention.
Detailed Description
In order to make the technical solutions better understood by those skilled in the art, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application. It is to be understood that the embodiments described are only a few embodiments of the present application and not all embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present application.
Examples
As shown in fig. 1, the present embodiment provides a method for identifying an oil level of a transformer in a distribution room, including the following steps:
s1, obtaining an oil level image of the transformer.
Preferably, the acquiring the transformer oil level image comprises acquiring an image of a shooting device and intercepting the image from a video; the method comprises the steps of intercepting an image from a video, setting a video frame selection interval according to the scene, the requirement and the comprehensive consideration of the performance, and converting the intercepted single-frame image into an image in a JPG format which can be processed by an oil table state recognition model;
s2, image preprocessing: the method comprises the steps of image filtering processing, image enhancement processing and image preprocessing on a picture after format conversion, so that partial noise interference in the picture is removed, and the consistency of data preprocessing operation of a training picture and a picture to be predicted is ensured;
preferably, the oil level indicator image collected by the camera is susceptible to illumination, and a large amount of noise is carried in the shooting and transmission processes. For accurate reading of a subsequent oil level indicator, preprocessing of the image is required, including homomorphic filtering, median filtering, binarization, morphological processing and the like.
S2.1 homomorphic filtering processing
Aiming at the fact that the collected oil level gauge image is susceptible to illumination, a homomorphic filter is adopted to enhance the oil level gauge image under weak illumination. The homomorphic filter is a frequency domain filtering algorithm based on the image illumination reflection imaging model principle, and can change the image gray scale range and enhance the image contrast. The invention adopts the Gaussian homomorphic filter H (mu, v) to carry out filtering processing on the oil level gauge image, enhances the brightness of the oil level gauge image under weak illumination and improves the contrast of the image. The calculation formula of the high-pass filter H (mu, v) is as follows:
where D (μ, v) is the distance of the point (μ, v) from the center of the frequency rectangle
The distance of (a); d o To cut off the frequency, the invention takes D o Median value of D (μ, v); the parameter c is used to control the sharpening of the homomorphic filter, and the invention makes c =2; parameter gamma H And gamma L Controlling the variation range of H (mu, v), H (mu, v) epsilon [ gamma ] L ，γ H ]The present invention makes gamma H ＝2.0，γ L ＝0.3。
The result is processed by using the Gaussian homomorphic filter, so that the effect of brightening the dark oil level indicator image in the camera acquisition process can be achieved, and meanwhile, the contrast of the pointer area and the non-pointer area of the oil level indicator image is enhanced.
S2.2, binarization and image morphology processing
Since the pointer of the oil level gauge image is mostly black, and other areas of the oil level gauge are mostly gray or white, the pointer can be distinguished from the other areas using the threshold method. The binarized image f (x, y) after processing by the thresholding method is expressed as:
wherein f (x, y) is a binarized image processed by a thresholding method, and T is a set threshold;
in the binarized picture, some tiny interference points still exist in the image, and the method effectively removes the interference points by using corrosion operation in morphology.
S3, positioning an oil meter pointer: and identifying the axis position of the oil level indicator by adopting a method of traversing pointer pixels, and identifying the position of the pointer by utilizing improved Hough transform.
S3.1, determining the axle center of the oil level gauge
As shown in fig. 2 (a), 2 (b) and 2 (c), for the identification of the axis of the oil level gauge, the present invention identifies the axis of the oil level gauge by traversing the pixels of the pointer. It can be seen that the color of the axis of the pointer is still different from that of the black pointer. Two segments of pointer interrupted by the axis can be found in the binarized image, the contour extraction operation is firstly carried out on the morphologically operated image according to the characteristic, two segments of interrupted pointers can be framed by two minimum circumscribed rectangle frames, then the mass centers of the two rectangles are respectively obtained, and the coordinates (x) of two mass points are obtained 1 ，y 1 )，(x 2 ，y 2 ). Connecting the two points can obtain a straight line on the pointer, and the slope k and the intercept b of the straight line can be obtained according to the two points, so that a straight line equation can be obtained:
y = k X + b, the pixels on the line are traversed, and the judgment can be made according to RGB to find the first non-black point (X) on the line 1 ，Y 1 ) And the last non-black dot (X) 2 ，Y 2 ) And recorded. And finally, calculating the middle point of the coordinates of the two points to obtain the coordinates (X, Y) of the axis.
X＝(X 1 +X 2 )/2
Y＝(Y 1 +Y 2 )/2
S3.2, pointer position identification
(1) Principle of Hough transform
The hough transform works based on dotted duality, as shown in fig. 3, where a straight line y = k x + b in a rectangular coordinate system is transformed into a parameter space straight line equation b = -k x + y, and considering the infinite slope possibility, the straight line is finally transformed into a polar coordinate space curve equation ρ = x cos θ + y sin θ, and θ ∈ (0, 180) is set to traverse the image with θ step size of 1, as shown in fig. 3.
(2) Hough transform improvement
And (4) identifying the position of the pointer, and adopting global Hough transform by a large number of scholars. The method firstly converts the coordinates into a polar coordinate system, and a straight line under the polar coordinate system can be expressed as follows: ρ = x cos θ + y sin θ
Where ρ represents the distance from the origin to the straight line; θ represents the inclination angle of the straight line perpendicular.
Each point on the edge of the image is then calculated by substituting it into the polar coordinate system conversion formula and the result is taken into the accumulator a (ρ, θ). Because points on the same straight line have the same rho and theta, every time a pair of (rho, theta) is calculated, A (rho, theta) = A (rho, theta) +1, and finally the maximum value in the accumulator is taken to obtain the straight line detection result. The method needs to calculate all the edge points, has large calculation amount, is easy to detect a plurality of nonexistent straight lines, and has great influence on the subsequent pointer position determination and the accuracy of oil level gauge reading.
As shown in fig. 4, the present invention adopts an improved hough transform, firstly, according to the first-order work of detecting the axis of the pointer, the position where the pointer is located is determined, and the ROI where the pointer is located is framed by the minimum rectangular frame; and then Hough transform is carried out in the small range to detect the straight line of the pointer, so that the calculation amount can be greatly reduced. And secondly, judging the detection result according to the characteristic that the pointer is the longest, and outputting and displaying the longest straight line, namely the detected pointer. After the pointer is detected, the coordinates of two end points of the straight line can be obtained, and then the distance L from the head part of the pointer to the axle center is used for obtaining the coordinates of the two end points of the straight line h Is greater than the distance L from the tail of the pointer to the axle center e Condition (L) h ＞L e ) Obtaining the coordinate of the head of the pointer as (P) 1 ，P 2 ). Compared with the traditional Hough transform, the method reduces the search range of the transform by limiting the parameter space, greatly reduces the storage space and the calculation amount, and improves the calculation speed.
S4, oil level main reading: and performing color segmentation according to the red color of the starting point and the ending point of the oil level indicator to obtain coordinates of the starting point and the ending point, and inputting the coordinates into the oil meter state identification model to judge whether the oil meter is detected.
And S4.1, reading the oil level gauge by using an angle method, wherein the determination of the starting point and the end point of the oil level gauge is quite important. Through observation, the starting point and the end point on the instrument scale ring are mostly red rectangular frames, so that the position of the starting point and the end point is obtained through a color segmentation and outline area method, and the specific steps are as follows:
(1) Converting the picture from RGB to HSV color space; the choice of converting the picture into the HSV color space is made because the HSV color space more easily represents a color, and the hue represented by the H value can substantially determine a certain color.
(2) And according to the red H value, making a red mask, and performing beta _ and calculation on the mask and the original image to obtain the region-of-interest image.
(3) Some fine interference points exist in the obtained interested area image, corrosion operation is needed to be carried out, the non-interested interference points are removed, and then expansion operation is carried out to enlarge the edge of the red area.
(4) Marking the rectangles of the starting point and the end point by using a contour method, and obtaining the centroids of the rectangles, wherein the coordinates of the two centroids correspond to the position coordinates (K) of the starting point and the end point 1 ，D 1 )，(K 2 ，D 2 )。
S4.2, oil level gauge reading
The pointer type oil level gauge is read by an angle method, and the axis coordinates (X, Y) and the pointer head coordinate (P) are obtained according to the previous steps 1 ，P 2 ) Starting point coordinate (K) 1 ，D 1 ) And end point coordinates (K) 2 ，D 2 ). Obtaining an angle alpha between the scales 0 and 10 by a vector angle method, wherein the angle of the dial plate with the scales is 360-alpha, the angle between each scale of the oil level gauge is gamma = (360-alpha)/10, and then the head coordinate (P) of the pointer is obtained according to calculation 1 ，P 2 ) And starting point coordinates (K) 1 ，D 1 ) And end point coordinates (K) 2 ，D 2 ) And the included angle beta between the pointer and the scale line 0 is calculated, the reading theta of the oil level indicator can be calculated according to the following formula:
and S5, judging a result and outputting alarm information. And according to the oil table state of the identification picture, if the oil table state is normal, directly returning to the normal state, and continuing to identify. If the abnormal condition is identified, alarm information is output.
As shown in fig. 5, in another embodiment of the present application, there is provided a power distribution room transformer oil level identification system, which includes an image acquisition module, a pointer positioning module, an oil level reading acquisition module, and an output module;
the image acquisition module is used for acquiring an oil level image of the transformer and carrying out image preprocessing;
the pointer positioning module is used for positioning a pointer of the transformer oil meter to obtain the axis position and the pointer position of the oil level gauge;
the oil level reading acquisition module is used for identifying a starting point and an end point of the oil level gauge to obtain an oil level main reading;
and the output module is used for judging the result and outputting alarm information if the result is identified to be abnormal.
It should be noted that the system provided in the above embodiment is only exemplified by the division of the above functional modules, and in practical applications, the above function distribution may be performed by different functional modules according to needs, that is, the internal structure is divided into different functional modules to perform all or part of the above described functions.
As shown in fig. 6, in another embodiment of the present application, there is further provided a storage medium storing a program, which when executed by a processor, implements the method for distinguishing the oil level of the distribution room transformer of the above embodiment, specifically:
acquiring an oil level image of the transformer, and preprocessing the image;
positioning a pointer of the transformer oil meter to obtain the axis position and the pointer position of an oil level gauge;
identifying a starting point and an end point of the oil level indicator to obtain an oil level main reading;
and judging the result, and if the result is identified to be abnormal, outputting alarm information.
It should be understood that portions of the present application may be implemented in hardware, software, firmware, or a combination thereof. In the above embodiments, various steps or methods may be implemented in software or firmware stored in a memory and executed by a suitable instruction execution system. For example, if implemented in hardware, as in another embodiment, any one or combination of the following technologies, which are well known in the art, may be used: a discrete logic circuit having a logic gate circuit for implementing a logic function on a data signal, an application specific integrated circuit having an appropriate combinational logic gate circuit, a Programmable Gate Array (PGA), a Field Programmable Gate Array (FPGA), or the like.
The above embodiments are preferred embodiments of the present invention, but the present invention is not limited to the above embodiments, and any other changes, modifications, substitutions, combinations, and simplifications which do not depart from the spirit and principle of the present invention should be construed as equivalents thereof, and all such modifications are intended to be included in the scope of the present invention.
Claims (10)
1. The method for distinguishing the oil level of the transformer in the power distribution room is characterized by comprising the following steps of:
acquiring an oil level image of the transformer, and preprocessing the image;
positioning a pointer of the transformer oil meter to obtain the axis position and the pointer position of the oil level gauge;
identifying a starting point and an end point of the oil level indicator to obtain an oil level main reading;
and judging the result, and if the result is identified to be abnormal, outputting alarm information.
2. The method for distinguishing the transformer oil level in the power distribution room according to claim 1, wherein the step of obtaining the transformer oil level image and performing image preprocessing comprises:
acquiring an oil level image of the transformer, including acquiring an image of a shooting device and intercepting the image from a video; intercepting an image from a video, setting a video frame selection interval, and converting the intercepted single-frame image into an image in a JPG format which can be processed by an oil table state recognition model;
image preprocessing, including homomorphic filtering, median filtering, binaryzation and morphological processing; and the data preprocessing operation is used for removing partial noise interference in the picture and enabling the training picture to be consistent with the picture to be predicted.
3. The method for distinguishing the oil level of the transformer in the power distribution room according to claim 2, wherein the homomorphic filtering is specifically as follows:
the oil level gauge image is subjected to filtering processing by a gaussian homomorphic filter H (μ, v) as follows:
where D (μ, v) is the distance of the point (μ, v) from the center of the frequency rectangle
The distance of (d); d o To cut off the frequency, the invention takes D o Median value of D (μ, v); the parameter c is used to control the sharpening of the homomorphic filter, and the invention makes c =2; parameter gamma H And gamma L Control H (variation range of [ mu ] v, [ mu ] v ] E [ gamma ] L ,γ H ]The present invention relates to gamma H ＝2.0，γ L ＝0.3。
4. The method for distinguishing the oil level of the transformer in the power distribution room according to claim 2, wherein the binarization and image morphology processing specifically comprises:
and (3) distinguishing the pointer from other areas in the image by using a threshold value method, wherein the following formula is as follows:
wherein f (x, y) is a binarized image processed by a threshold method, and T is a set threshold;
and processing the binary image processed by the threshold method by using corrosion operation in morphology.
5. The method for distinguishing the oil level of the transformer in the power distribution room according to claim 1, wherein the identification of the axis position of the oil level indicator adopts a method of traversing pointer pixels, and specifically comprises the following steps:
carrying out contour extraction operation on the image subjected to the image morphology processing: by using two pointers whose minimum external rectangle frames are interrupted into two sections by axis center, respectively obtaining mass center coordinates (x) of these two rectangles 1 ，y 1 )，(x 2 ，y 2 )；
Connecting the coordinates of the centers of mass of the two rectangles, and solving the slope k and intercept b of a straight line, wherein the equation of the straight line is y = k x + b;
traversing the pixels on the straight line, judging according to RGB, and obtaining the first non-black point (X) on the straight line 1 ，Y 1 ) And the last non-black dot (X) 2 ，Y 2 ) And finally, calculating the intermediate point to obtain the coordinates (X, Y) of the axis as follows:
X＝(X 1 +X 2 )/2
Y＝(Y 1 +Y 2 )/2。
6. the method for distinguishing the oil level of the transformer in the power distribution room according to claim 1, wherein the identification of the position of the pointer of the oil level indicator adopts improved Hough transform, and specifically comprises the following steps:
judging the position of the pointer approximately according to the first-order work of detecting the axis of the pointer;
using the minimum rectangular frame to frame out the ROI where the pointer is located;
performing Hough transform within the ROI to detect a pointer straight line;
judging the detection result according to the characteristic that the pointer is the longest, and only outputting and displaying the longest straight line to obtain the coordinates of two end points of the straight line, namely the detected pointer;
judging the distance between two end points of the pointer and the axle center, wherein the end with larger distance is the coordinate (P) of the pointer head 1 ，P 2 )。
7. The oil level identification method for the transformer in the power distribution room according to claim 1, wherein the positions of the starting point and the end point of the oil level gauge are identified by a color division and outline area method, and the method comprises the following specific steps:
converting the picture from RGB to HSV color space;
according to the red H value, making a red mask, and performing beta _ and operation on the mask and the original image to obtain an image of the region of interest;
carrying out corrosion operation on the obtained interested region image, removing non-interested interference points, and then carrying out expansion operation to enlarge the edge of the red region;
marking the rectangles of the starting point and the end point by using a contour method, and obtaining the centroids of the rectangles, wherein the coordinates of the two centroids correspond to the position coordinates (K) of the starting point and the end point 1 ，D 1 )，(K 2 ，D 2 )。
8. The method for distinguishing the oil level of the transformer in the power distribution room according to claim 1, wherein the main reading of the identified oil level is performed by adopting an angle method, and the method comprises the following steps:
according to the axis coordinate (X, Y) and the pointer head coordinate (P) 1 ，P 2 ) Starting point coordinate (K) 1 ，D 1 ) And end point coordinates (K) 2 ，D 2 ) And calculating an angle alpha between scales 0 and 10 by a vector included angle method, and then:
the angle of the dial plate with scales is 360-alpha;
the included angle gamma between each scale mark of the oil level indicator is = (360-alpha)/10;
then according to the fingerTip segment coordinate (P) 1 ，P 2 ) And starting point coordinates (K) 1 ，D 1 ) And end point coordinates (K) 2 ，D 2 ) And calculating an included angle beta between the pointer and the scale line 0, and reading theta of the oil level indicator is as follows:
9. the oil level distinguishing system for the transformer of the power distribution room is applied to the oil level distinguishing method for the transformer of the power distribution room, and comprises an image acquiring module, a pointer positioning module, an oil level reading acquiring module and an output module;
the image acquisition module is used for acquiring an oil level image of the transformer and carrying out image preprocessing;
the pointer positioning module is used for positioning a pointer of the transformer oil meter to obtain the axis position and the pointer position of the oil level gauge;
the oil level reading acquisition module is used for identifying a starting point and an end point of the oil level gauge to obtain an oil level main reading;
and the output module is used for judging the result and outputting alarm information if the result is identified to be abnormal.
10. A storage medium storing a program, characterized in that: the program, when executed by a processor, implements the electrical distribution room transformer oil level identification method of any one of claims 1-8.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211053186.8A CN115409803A (en) | 2022-08-31 | 2022-08-31 | Oil level identification method, system and medium for transformer in power distribution room |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211053186.8A CN115409803A (en) | 2022-08-31 | 2022-08-31 | Oil level identification method, system and medium for transformer in power distribution room |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN115409803A true CN115409803A (en) | 2022-11-29 |
Family
ID=84163574
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202211053186.8A Pending CN115409803A (en) | 2022-08-31 | 2022-08-31 | Oil level identification method, system and medium for transformer in power distribution room |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115409803A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| 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 |
-
2022
- 2022-08-31 CN CN202211053186.8A patent/CN115409803A/en active Pending
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| 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 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110580480B (en) | Meter reading identification method based on image processing | |
| WO2022057103A1 (en) | Deep-learning-based method for automatically reading pointer instrument | |
| CN109308447A (en) | The method of equipment operating parameter and operating status is automatically extracted in remote monitoriong of electric power | |
| CN108764234B (en) | Liquid level meter reading identification method based on inspection robot | |
| US9111183B2 (en) | Performing a comparison between two images which are scaled to a common resolution | |
| CN109087363B (en) | HSV color space-based sewage discharge detection method | |
| CN116758045B (en) | Surface defect detection method and system for semiconductor light-emitting diode | |
| CN115049664A (en) | Vision-based ship engine fitting defect detection method | |
| CN111476246B (en) | Robust and efficient intelligent reading method for pointer instrument applied to complex environment | |
| CN111369516A (en) | Transformer bushing heating defect detection method based on infrared image recognition | |
| CN116433666B (en) | Board card line defect online identification method, system, electronic equipment and storage medium | |
| CN109308448A (en) | A method of it prevents from becoming distribution maloperation using image processing techniques | |
| CN106803087A (en) | A kind of car number automatic identification method and system | |
| CN111325164A (en) | Pointer indication number identification method and device and electronic equipment | |
| CN111639568A (en) | Navigation patrol method for intelligent robot of transformer substation | |
| CN113657339A (en) | Instrument pointer counting and reading method and medium based on machine vision | |
| CN115409803A (en) | Oil level identification method, system and medium for transformer in power distribution room | |
| CN114066862A (en) | Indicator identification method and system based on color gamut and contour characteristics | |
| CN112132131A (en) | Measuring cylinder liquid level identification method and device | |
| CN112036391B (en) | Electronic inspection method and system based on shale gas field production station | |
| CN112990064A (en) | Dial pointer identification method based on color segmentation and probability model | |
| CN117911353A (en) | Method for detecting surface defects of circular crystal oscillator wafer | |
| CN116612461A (en) | Target detection-based pointer instrument whole-process automatic reading method | |
| CN114266719B (en) | Hough transform-based product detection method | |
| CN114898347A (en) | Machine vision identification method for pointer instrument |
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 |