An image recognition-based SF6 meter pressure monitoring method and system

Abstract

The application relates to the technical field of meter detection, and discloses an SF6 meter pressure monitoring method and system based on image recognition. The method matches feature points of an SF6 meter image to be recognized and a preset meter template image by using a SURF algorithm, determines the positions of pointers in the SF6 meter image and the distribution of the dial scale, and thus determines the meter reading of the SF6 meter image. The meter reading of each SF6 meter image and the change of the environment temperature of the SF6 meter acquired in advance are subjected to curve fitting, the relative deviation of a characteristic curve of the SF6 pressure change with the temperature to a preset characteristic curve of the standard SF6 pressure change with the temperature is calculated, the change rate of the relative deviation is compared with a preset rate threshold, the change trend of the SF6 pressure in a preset period is determined according to the comparison result, and thus the SF6 pressure change can be monitored in time and effectively, and the SF6 pressure monitoring precision is improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of meter detection, in particular to a SF6 meter pressure monitoring method and system based on image recognition. BACKGROUND

[0002] SF6 pressure indication is a key indicator in substation operation and maintenance, which reflects the insulation and reliability of important equipment such as circuit breakers and disconnectors in the substation. The main monitoring method of SF6 pressure indication currently applied in the substation is: 1) manual patrol, substation operation and maintenance personnel periodically patrol each SF6 meter in the substation, observe with naked eyes, manually count the readings of SF6 pressure meters, and preliminarily judge the health degree of the equipment in combination with historical data, weather data, equipment historical defects, etc. The advantage is that manual patrol of the meter ensures the reliability of the judgment result, but the disadvantage is that there are few personnel, many devices, heavy tasks and low efficiency. 2) background monitoring, a density relay is installed in the SF6 pressure meter, when the SF6 pressure is lower than the normal value, the relay acts, and the background sends an SF6 pressure low alarm signal. However, the installation position is outdoor and is greatly affected by the environment, the pressure relay has a misoperation condition, it is difficult to find the change trend of the health degree of the equipment, and there are only normal and alarm information. When the SF6 pressure is lower than the rated value and higher than the alarm value, the SF6 pressure monitoring is in a vacuum period, it is difficult to monitor the SF6 pressure in time and effectively, and the SF6 pressure monitoring precision is also low. SUMMARY

[0003] The present application provides a SF6 meter pressure monitoring method and system based on image recognition, which solves the technical problems that it is difficult to monitor the SF6 pressure in time and effectively, and the SF6 pressure monitoring precision is also low.

[0004] Therefore, the first aspect of the present application provides a SF6 meter pressure monitoring method based on image recognition, comprising the following steps:

[0005] An SF6 meter image set is obtained, which is composed of a plurality of to-be-recognized SF6 meter images with time sequence;

[0006] SURF algorithm is used for feature detection of each to-be-recognized SF6 meter image, and feature point matching is performed on the to-be-recognized SF6 meter image and a preset meter template image to determine the position of the pointer in the SF6 meter image and the distribution of the dial scale;

[0007] The meter indication of the SF6 meter image is determined according to the position of the pointer in the SF6 meter image and the distribution of the dial scale;

[0008] fitting the meter indication of each SF6 meter image and the pre-acquired environmental temperature variation of the SF6 meter, to obtain a characteristic curve of SF6 pressure variation with temperature;

[0009] calculating a relative deviation of the characteristic curve of SF6 pressure variation with temperature from a preset standard characteristic curve of SF6 pressure variation with temperature;

[0010] calculating a variation rate of the relative deviation according to the relative deviation and the meter indication of the SF6 meter image in the preset period and the standard SF6 pressure;

[0011] comparing the variation rate of the relative deviation with a preset rate threshold, and determining a variation trend of the SF6 pressure in the preset period according to a comparison result.

[0012] Preferably, the step of acquiring the SF6 meter image set specifically comprises:

[0013] The UAV performs a flight patrol task according to a preset flight route, and takes a photograph of the SF6 meter at a preset photographing point to obtain a plurality of SF6 meter images to be identified, and forms the SF6 meter image set.

[0014] Preferably, the step of performing feature detection on each of the SF6 meter images to be identified, and performing feature point matching on the SF6 meter image to be identified and a preset meter template image by using a SURF algorithm to determine the position of the pointer and the distribution of the dial scale specifically comprises:

[0015] performing gray scale processing on each of the SF6 meter images to be identified to obtain SF6 meter gray scale images;

[0016] performing multi-scale detection on the SF6 meter gray scale images by using a SURF algorithm to obtain key feature points, and describing the key feature points by using feature vectors of local image regions;

[0017] comparing the Euclidean distance between the feature vectors of the key feature points corresponding to the SF6 meter gray scale images and the feature vectors of the key feature points of the preset meter template image to determine whether the SF6 meter gray scale images have pointers;

[0018] If it is determined that the SF6 meter gray scale images have pointers, then mapping the coordinates of the pre-labeled pointers and the coordinates of the scales on the preset meter template image to each of the SF6 meter gray scale images by using a perspective matrix transformation, so as to determine the position of the pointer and the distribution of the dial scale in the SF6 meter image.

[0019] Preferably, the step of determining the meter indication of the SF6 meter image according to the position of the pointer and the distribution of the dial scale in the SF6 meter image specifically comprises:

[0020] K-means binaryzation is performed on the SF6 meter image to obtain a binary image, and a pointer rotation center of a preset meter template image is mapped on the binary image to obtain a position of the pointer rotation center of the SF6 meter image;

[0021] The rotation angle of the pointer of the SF6 meter image is determined according to the relative position of the pointer rotation center and the pointer;

[0022] The meter reading of the SF6 meter image is determined according to the rotation angle of the pointer and the distribution of the dial scale by using the angle method.

[0023] Preferably, the step of calculating the relative deviation of the characteristic curve of the SF6 pressure changing with temperature from the preset standard characteristic curve of the SF6 pressure changing with temperature specifically comprises:

[0024] The relative deviation of the characteristic curve of the SF6 pressure changing with temperature from the preset standard characteristic curve of the SF6 pressure changing with temperature is calculated by using the least square method as follows:

[0025] d j % = aj + b

[0026] In the formula, j represents a sampling point, d j represents the relative deviation of the jth sampling point, a represents a slope, and b represents a constant.

[0027]

[0028] In the formula, p j represents the SF6 pressure of the jth sampling point, p j ' represents the standard SF6 pressure of the jth sampling point.

[0029] Preferably, the step of calculating the change rate of the relative deviation according to the relative deviation and the meter reading of the SF6 meter image and the standard SF6 pressure in a preset period specifically comprises:

[0030] The change rate of the relative deviation is calculated according to the relative deviation and the meter reading of the SF6 meter image and the standard SF6 pressure in a preset period by the following formula:

[0031]

[0032] In the formula, n represents the number of sampling points in the preset period.

[0033] Preferably, the step of comparing the change rate of the relative deviation with a preset rate threshold and determining the change trend of the SF6 pressure in the preset period according to the comparison result specifically comprises:

[0034] The rate of change of the relative deviation is compared with zero, if the rate of change of the relative deviation is less than zero, it is determined that the SF6 pressure continuously decreases in the preset period, and a warning signal is sent; if the rate of change of the relative deviation is equal to zero, it is determined that the SF6 pressure remains stable in the preset period; if the rate of change of the relative deviation is greater than zero, it is determined that the SF6 pressure continuously increases in the preset period.

[0035] In a second aspect, the present application provides an SF6 meter pressure monitoring system based on image recognition, comprising:

[0036] An image acquisition module is configured to acquire an SF6 meter image set, wherein the SF6 meter image set is composed of a plurality of to-be-recognized SF6 meter images with time sequence;

[0037] A feature matching module is configured to perform feature detection on each of the to-be-recognized SF6 meter images by using a SURF algorithm, and perform feature point matching on the to-be-recognized SF6 meter images and a preset meter template image, so as to determine the position of a pointer and the distribution of a dial scale in the SF6 meter image;

[0038] A meter reading recognition module is configured to determine the meter reading of the SF6 meter image according to the position of the pointer and the distribution of the dial scale in the SF6 meter image;

[0039] A curve fitting module is configured to perform curve fitting on the meter reading of each SF6 meter image and the pre-acquired environmental temperature variation of the SF6 meter, so as to obtain a characteristic curve of the SF6 pressure variation with temperature;

[0040] A deviation calculation module is configured to calculate the relative deviation between the characteristic curve of the SF6 pressure variation with temperature and a preset standard characteristic curve of the SF6 pressure variation with temperature;

[0041] A deviation rate calculation module is configured to calculate the rate of change of the relative deviation according to the relative deviation and the meter reading of the SF6 meter image in a preset period and the standard SF6 pressure;

[0042] A pressure variation determination module is configured to compare the rate of change of the relative deviation with a preset rate threshold, and determine the variation trend of the SF6 pressure in the preset period according to the comparison result.

[0043] In a third aspect, the present application provides an electronic device, comprising: a memory configured to store executable instructions; and a processor configured to communicate with the memory to execute the executable instructions so as to complete the steps of the above method.

[0044]

[0045] ​Fourthly, the present invention provides a computer storage medium for storing computer-readable instructions, which, when executed, perform the steps of the above-described method.

[0046] As can be seen from the above technical solutions, the present invention has the following advantages:

[0047] This method acquires a set of SF6 meter images and uses the SURF algorithm to match feature points between the SF6 meter image to be identified and a preset meter template image. This determines the position of the pointer and the scale distribution on the dial in the SF6 meter image, thereby determining the meter reading. Curve fitting is then performed on the meter reading of each SF6 meter image and the pre-acquired ambient temperature change data to obtain the characteristic curve of SF6 pressure versus temperature. The relative deviation between this characteristic curve and the preset standard SF6 pressure characteristic curve is calculated, and the rate of change of the relative deviation is compared with a preset rate threshold. Based on the comparison results, the trend of SF6 pressure change within a preset period is determined. This allows for timely and effective monitoring of SF6 pressure changes based on the trend, improving the accuracy of SF6 pressure monitoring and filling the gaps in existing SF6 pressure monitoring. Attached Figure Description

[0048] Figure 1 A flowchart of an SF6 meter pressure monitoring method based on image recognition provided in an embodiment of the present invention;

[0049] Figure 2 This is a schematic diagram of an SF6 meter pressure monitoring system based on image recognition, provided as an embodiment of the present invention. Detailed Implementation

[0050] To enable those skilled in the art to better understand the present invention, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0051] For easier understanding, please refer to Figure 1 The present invention provides a method for monitoring SF6 meter pressure based on image recognition, comprising the following steps:

[0052] S1. Obtain the SF6 meter image set, which consists of multiple SF6 meter images to be identified that have a temporal sequence.

[0053] It is understandable that SF6 meter images can be continuously sampled within a certain period to obtain a set of SF6 meter images consisting of multiple SF6 meter images to be identified, which are in a time sequence.

[0054] In one example, the steps to obtain an SF6 meter image set specifically include: using a drone to perform a flight inspection mission along a preset route, taking pictures of the SF6 meters at preset photo points, obtaining multiple images of the SF6 meters to be identified, and forming an SF6 meter image set.

[0055] The preset photo-taking points are located on the preset flight path. At the same time, the drone can upload the collected SF6 meter images to the backend database for storage.

[0056] S2. Use the SURF algorithm to perform feature detection on each SF6 meter image to be identified, and perform feature point matching between the SF6 meter image to be identified and the preset meter template image to determine the position of the pointer and the scale distribution of the dial in the SF6 meter image.

[0057] S3. Determine the meter reading of the SF6 meter image based on the position of the pointer and the scale distribution on the dial.

[0058] S4. Perform curve fitting on the meter readings of each SF6 meter image and the pre-acquired ambient temperature changes of the SF6 meter to obtain the characteristic curve of SF6 pressure changing with temperature.

[0059] Specifically, when acquiring each SF6 meter image, the ambient temperature and meter number of the corresponding SF6 meter can be obtained and imported into the historical dataset:

[0060]

[0061] In the formula, [P i ,T i ] represents the historical dataset of [SF6 pressure, temperature] for SF6 meter numbered i, p 1～n t represents the set of all SF6 pressure monitoring data for SF6 meter i. 1～n p′ represents the set of ambient temperature data corresponding to all SF6 pressures of SF6 meter i. 1～n This represents the characteristic curve of the standard SF6 pressure of SF6 meter i changing with temperature at temperatures from t1 to n.

[0062] Among them, the characteristic curve of SF6 pressure changing with temperature shows the normal SF6 pressure reading of the equipment at different temperatures. The characteristic curve of standard SF6 pressure changing with temperature can be provided by the manufacturer or obtained by fitting based on empirical values.

[0063] In practical applications, the substation operation procedures also specify a rated pressure P for SF6. k Low pressure alarm value P a and low pressure lockout value P b And P k >P b >Pb.

[0064] The existing SF6 monitoring and alarm logic is as follows: when P b <P i <P a At that time, the background system issued a "low SF6 pressure alarm" signal; when the SF6 pressure continued to drop to P i <P b At that time, the background system issued a "SF6 pressure low lockout" signal; however, when P k >P i >P a At that time, the existing monitoring system did not have a corresponding status signal set up, therefore, it was a monitoring vacuum period, and there was no corresponding status signal in the background. However, at this time, the actual SF6 pressure was already lower than the rated pressure P. k Furthermore, there is a possibility of a further decline. Therefore, this embodiment monitors the SF6 pressure change trend within this vacuum monitoring range to provide early warning of SF6 pressure changes.

[0065] S5. Calculate the relative deviation between the characteristic curve of SF6 pressure changing with temperature and the preset characteristic curve of SF6 pressure changing with temperature.

[0066] In one example, the relative deviation between the characteristic curve of SF6 pressure versus temperature calculated using the least squares method and the preset standard characteristic curve of SF6 pressure versus temperature is:

[0067] d j % = aj + b

[0068] In the formula, j represents the sampling point, and d j Let a represent the relative deviation of the j-th sampling point, a represent the slope, and b represent a constant.

[0069]

[0070] In the formula, p j p represents the SF6 pressure at the j-th sampling point. j ′ represents the standard SF6 pressure at the j-th sampling point.

[0071] S6. Calculate the rate of change of the relative deviation based on the relative deviation and the meter readings of the SF6 meter image within the preset period and the standard SF6 pressure.

[0072] In one example, the rate of change of the relative deviation is calculated using the following formula, based on the relative deviation and the meter readings of the SF6 meter image within a preset period, compared to the standard SF6 pressure:

[0073]

[0074] In the formula, n represents the number of sampling points within the preset period.

[0075] S7. Compare the rate of change of the relative deviation with the preset rate threshold, and determine the trend of SF6 pressure change within the preset period based on the comparison result.

[0076] In one feasible approach, the rate of change of the relative deviation is compared with zero. If the rate of change of the relative deviation is less than zero, it is determined that the SF6 pressure continues to decrease within a preset period, and an early warning signal is issued. If the rate of change of the relative deviation is equal to zero, it is determined that the SF6 pressure remains stable within the preset period. If the rate of change of the relative deviation is greater than zero, it is determined that the SF6 pressure continues to rise within the preset period.

[0077] It should be noted that this method acquires a set of SF6 meter images, uses the SURF algorithm to match feature points between the SF6 meter image to be identified and a preset meter template image, determines the position of the pointer and the scale distribution on the dial in the SF6 meter image, and thus determines the meter reading of the SF6 meter image. Curve fitting is then performed on the meter reading of each SF6 meter image and the pre-acquired ambient temperature change of the SF6 meter to obtain the characteristic curve of SF6 pressure changing with temperature. The relative deviation between the characteristic curve of SF6 pressure changing with temperature and the preset standard characteristic curve of SF6 pressure changing with temperature is calculated, and the rate of change of the relative deviation is compared with a preset rate threshold. Based on the comparison result, the trend of SF6 pressure change within a preset period is determined. Therefore, timely and effective monitoring of SF6 pressure changes can be carried out based on the trend of SF6 pressure change, improving the accuracy of SF6 pressure monitoring and filling the gaps in existing SF6 pressure monitoring.

[0078] In one specific embodiment, step S2 specifically includes:

[0079] 201. Perform grayscale processing on each SF6 meter image to be identified to obtain an SF6 meter grayscale image.

[0080] This method can convert the red, green, and blue channel information in each SF6 meter image to be identified into a single gray channel information, reducing the amount of data and increasing computational efficiency.

[0081] The grayscale processing procedure is as follows:

[0082] G(x,y)=0.299R(x,y)+0.578G(x,y)+0.114B(x,y)

[0083] In the formula, R(x,y), G(x,y), and B(x,y) are the red, green, and blue channel values ​​corresponding to the image coordinates (x,y), respectively, while G(x,y) is the channel value after grayscale processing. Based on the meter's range and scale, find the key scale coordinates and their corresponding coordinate value set [X,Y,1].

[0084]

[0085] 202. Use the SURF algorithm to perform multi-scale detection on the grayscale image of SF6 meter to obtain key feature points, and use the feature vectors of local image regions to describe the key feature points.

[0086] Specifically, the SURF algorithm can be used with the difference of Gaussians function to detect key feature points at different scales in the grayscale image of SF6 meters. During the detection process, the SF6 meter image can be blurred and smoothed at different scales, and key feature points in the SF6 meter image can be calculated.

[0087] 203. Compare the Euclidean distance between the feature vectors of the key feature points corresponding to the grayscale image of the SF6 meter and the feature vectors of the key feature points of the preset meter template image to determine whether the grayscale image of the SF6 meter has a pointer.

[0088] It should be noted that in the feature description stage, the SURF algorithm uses feature vectors of local image regions to describe key points. The feature vector value of the template image feature points is TIFV(i), and the feature vector value of the target image feature points is OIFV(i).

[0089]

[0090] In the formula, (x i ,y i ) and (x i ′,y i ′) are the coordinates of feature points in the SF6 meter template image and the SF6 meter image to be identified, respectively.

[0091] The SURF algorithm performs feature matching by comparing the Euclidean distance Ed between feature vectors. Typically, feature vectors of feature points are 64-dimensional or 128-dimensional. This invention selects 64 dimensions. The formula for calculating the Euclidean distance Ed(i) between TIFV and OIFV is as follows:

[0092]

[0093] Determine whether the Euclidean distance Ed(i) is less than a preset distance threshold. If the Euclidean distance Ed(i) is less than the preset distance threshold, then determine that the grayscale image of the SF6 meter has a pointer. If the Euclidean distance Ed(i) is not less than the preset distance threshold, then determine that the grayscale image of the SF6 meter does not have a pointer.

[0094] 204. If it is determined that there is a pointer in the grayscale image of the SF6 meter, the coordinates of the pointer and the scale on the preset meter template image are mapped to each grayscale image of the SF6 meter through perspective matrix transformation, thereby determining the position of the pointer and the distribution of the dial scale in the SF6 meter image.

[0095] In this process, the key scale coordinates marked in the template image can be mapped to each target image to be matched through perspective matrix transformation, thus completing the matching of key scale values ​​of the target images, as shown in the following formula:

[0096]

[0097] In the formula, M is the perspective matrix. These are the position coordinates mapped to the grayscale image of the SF6 meter.

[0098] In one specific embodiment, step S3 specifically includes:

[0099] 301. Perform K-means binarization on the SF6 meter image to obtain a binary image. Map the pointer rotation center of the preset meter template image onto the binary image to obtain the position of the pointer rotation center of the SF6 meter image.

[0100] 302. Determine the rotation angle of the pointer in the SF6 meter image based on the pointer rotation center and the relative position of the pointer.

[0101] 303. Using the angle method, determine the meter reading of the SF6 meter image based on the rotation angle of the pointer and the scale distribution of the dial.

[0102] The above is a detailed description of a real-time example of an SF6 meter pressure monitoring method based on image recognition provided by the present invention. The following is a detailed description of an embodiment of an SF6 meter pressure monitoring system based on image recognition provided by the present invention.

[0103] For easier understanding, please refer to Figure 2 The present invention also provides an SF6 meter pressure monitoring system based on image recognition, comprising:

[0104] The image acquisition module 100 is used to acquire an SF6 meter image set, which consists of multiple SF6 meter images to be identified that have a temporal sequence.

[0105] The feature matching module 200 is used to perform feature detection on each SF6 meter image to be identified using the SURF algorithm, and to perform feature point matching between the SF6 meter image to be identified and the preset meter template image to determine the position of the pointer and the scale distribution of the dial in the SF6 meter image.

[0106] The meter reading recognition module 300 is used to determine the meter reading of the SF6 meter image based on the position of the pointer and the scale distribution of the dial in the SF6 meter image.

[0107] The curve fitting module 400 is used to perform curve fitting on the meter readings of each SF6 meter image and the pre-acquired ambient temperature changes of the SF6 meter to obtain the characteristic curve of SF6 pressure changing with temperature.

[0108] The deviation calculation module 500 is used to calculate the relative deviation between the characteristic curve of SF6 pressure changing with temperature and the preset standard characteristic curve of SF6 pressure changing with temperature.

[0109] The deviation rate calculation module 600 is used to calculate the rate of change of the relative deviation based on the relative deviation and the meter reading of the SF6 meter image within a preset period and the standard SF6 pressure.

[0110] The pressure change determination module 700 is used to compare the rate of change of the relative deviation with a preset rate threshold, and determine the trend of SF6 pressure change within a preset period based on the comparison result.

[0111] The present invention also provides an electronic device, comprising: a memory for storing executable instructions; and a processor for communicating with the memory to execute the executable instructions to perform the steps of the above method.

[0112] The present invention also provides a computer storage medium for storing computer-readable instructions that, when executed, perform the steps of the above-described method.

[0113] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the specific working processes of the systems, electronic devices, and computer storage media described above can be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here.

[0114] In the embodiments provided by this invention, it should be understood that the disclosed systems, electronic devices, computer storage media, and methods can be implemented in other ways. For example, the device embodiments described above are merely illustrative; for instance, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be an indirect coupling or communication connection between devices or units through some interfaces, and may be electrical, mechanical, or other forms.

[0115] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0116] Furthermore, the functional units in the various embodiments of the present invention can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit.

[0117] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions for executing all or part of the steps of the methods described in the various embodiments of the present invention through a computer device (which may be a personal computer, server, or network device, etc.). The aforementioned storage medium includes: USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, optical disks, and other media capable of storing program code.

[0118] The above embodiments are only used to illustrate the technical solutions of the present invention, and are not intended to limit it. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for SF6 gage pressure monitoring based on image recognition, characterized in that, The method comprises the following steps: obtaining an SF6 meter image set composed of multiple SF6 meter images to be recognized with time sequence; detecting features of each SF6 meter image to be recognized by using a SURF algorithm, and matching feature points of the SF6 meter image to be recognized with a preset meter template image to determine the position of a pointer in the SF6 meter image and the distribution of a dial scale, specifically comprising: performing gray scale processing on each SF6 meter image to be recognized to obtain an SF6 meter gray scale image; detecting multiple scales of the SF6 meter gray scale image by using the SURF algorithm to obtain key feature points, and describing the key feature points by using feature vectors of local image regions; comparing the Euclidean distance between the feature vectors of the key feature points corresponding to the SF6 meter gray scale image and the feature vectors of the key feature points of the preset meter template image to determine whether the SF6 meter gray scale image has a pointer; if it is determined that the SF6 meter gray scale image has a pointer, mapping the coordinates of the pre-labeled pointer and the coordinates of the scale on the preset meter template image to each SF6 meter gray scale image by perspective matrix transformation to determine the position of the pointer in the SF6 meter image and the distribution of the dial scale; determining the meter reading of the SF6 meter image according to the position of the pointer in the SF6 meter image and the distribution of the dial scale, specifically comprising: performing K-means binary processing on the SF6 meter image to obtain a binary image, mapping the pointer rotation center of the preset meter template image on the binary image to obtain the position of the pointer rotation center of the SF6 meter image; determining the rotation angle of the pointer of the SF6 meter image according to the relative position of the pointer rotation center and the pointer; determining the meter reading of the SF6 meter image according to the rotation angle of the pointer and the distribution of the dial scale by using an angle method; performing curve fitting on the meter reading of each SF6 meter image and the change of the environmental temperature of the SF6 meter obtained in advance to obtain a characteristic curve of SF6 pressure change with temperature; calculating the relative deviation of the characteristic curve of SF6 pressure change with temperature from a preset standard characteristic curve of SF6 pressure change with temperature; determining the change rate of the relative deviation of the SF6 meter image in a preset period from the standard SF6 pressure calculation relative deviation according to the relative deviation and the meter reading of the SF6 meter image in the preset period; comparing the change rate of the relative deviation with a preset rate threshold, and determining the change trend of the SF6 pressure in the preset period according to the comparison result.

2. The image recognition based SF6 gage pressure monitoring method of claim 1, wherein, The step of obtaining the SF6 meter image set specifically comprises: performing a flight patrol task according to a preset flight route by using a drone, and taking pictures of SF6 meters at preset photographing points to obtain multiple SF6 meter images to be recognized and form the SF6 meter image set.

3. The image recognition based SF6 gage pressure monitoring method of claim 1, wherein, The step of calculating the relative deviation of the characteristic curve of SF6 pressure change with temperature from the preset standard characteristic curve of SF6 pressure change with temperature specifically comprises: The least square method is used to calculate a relative deviation of the characteristic curve of the SF6 pressure changing with temperature from a preset standard characteristic curve of the SF6 pressure changing with temperature, and the relative deviation is calculated according to the relative deviation and the meter indication of the SF6 meter image and the change rate of the relative deviation of the standard SF6 pressure within a preset period. d j % = a j + b where j represents a sampling point, d j represents the relative deviation of the jth sampling point, a represents the slope, and b represents the constant, wherein, where p j represents the SF6 pressure at the jth sampling point, p j represents the standard SF6 pressure at the jth sampling point.

4. The image recognition based SF6 meter pressure monitoring method of claim 3, wherein, The step of calculating the change rate of the relative deviation according to the relative deviation and the meter indication of the SF6 meter image and the change rate of the relative deviation of the standard SF6 pressure within a preset period specifically comprises: The step of calculating the change rate of the relative deviation according to the relative deviation and the meter indication of the SF6 meter image and the change rate of the relative deviation of the standard SF6 pressure within a preset period specifically comprises: In the formula, n represents the number of sampling points within the preset period.

5. The image recognition based SF6 meter pressure monitoring method of claim 4, wherein, The step of comparing the change rate of the relative deviation with a preset rate threshold and determining the change trend of the SF6 pressure within the preset period according to a comparison result specifically comprises: The change rate of the relative deviation is compared with zero, if the change rate of the relative deviation is less than zero, it is determined that the SF6 pressure within the preset period continuously decreases, and a warning signal is sent out; if the change rate of the relative deviation is equal to zero, it is determined that the SF6 pressure within the preset period remains stable; if the change rate of the relative deviation is greater than zero, it is determined that the SF6 pressure within the preset period continuously increases.

6. An image recognition based SF6 meter pressure monitoring system characterized in that, Comprise: An image acquisition module is configured to acquire a set of SF6 meter images, which are composed of a plurality of to-be-identified SF6 meter images with time sequence; A feature matching module is configured to perform feature detection on each of the to-be-identified SF6 meter images by using a SURF algorithm, and perform feature point matching on the to-be-identified SF6 meter images and a preset meter template image, so as to determine the position of a pointer and the distribution of a dial scale in the SF6 meter images; The step of performing feature detection on each of the to-be-identified SF6 meter images by using a SURF algorithm, and performing feature point matching on the to-be-identified SF6 meter images and a preset meter template image, so as to determine the position of a pointer and the distribution of a dial scale in the SF6 meter images specifically comprises: The step of performing feature detection on each of the to-be-identified SF6 meter images by using a SURF algorithm, and performing feature point matching on the to-be-identified SF6 meter images and a preset meter template image, so as to determine the position of a pointer and the distribution of a dial scale in the SF6 meter images specifically comprises: The step of performing feature detection on each of the to-be-identified SF6 meter images by using a SURF algorithm, and performing feature point matching on the to-be-identified SF6 meter images and a preset meter template image, so as to determine the position of a pointer and the distribution of a dial scale in the SF6 meter images specifically comprises: The step of performing feature detection on each of the to-be-identified SF6 meter images by using a SURF algorithm, and performing feature point matching on the to-be-identified SF6 meter images and a preset meter template image, so as to determine the position of a pointer and the distribution of a dial scale in the SF6 meter images specifically comprises: The step of performing feature detection on each of the to-be-identified SF6 meter images by using a SURF algorithm, and performing feature point matching on the to-be-identified SF6 meter images and a preset meter template image, so as to determine the position of a pointer and the distribution of a dial scale in the SF6 meter images specifically comprises: The step of determining the meter indication of the SF6 meter image according to the position of the pointer and the distribution of the dial scale in the SF6 meter image specifically comprises: The step of determining the meter indication of the SF6 meter image according to the position of the pointer and the distribution of the dial scale in the SF6 meter image specifically comprises: The SF6 meter image is subjected to K-means binarization to obtain a binary image, and a pointer rotation center of a preset meter template image is mapped on the binary image to obtain a position of the pointer rotation center of the SF6 meter image; A rotation angle of the pointer of the SF6 meter image is determined according to the relative position of the pointer rotation center and the pointer; An angle method is used to determine the meter reading of the SF6 meter image according to the rotation angle of the pointer and a scale distribution of the dial; A curve fitting module is configured to perform curve fitting on the meter reading of each SF6 meter image and a pre-acquired temperature variation of the SF6 meter to obtain a characteristic curve of SF6 pressure variation with temperature; A deviation calculation module is configured to calculate a relative deviation of the characteristic curve of SF6 pressure variation with temperature from a preset standard characteristic curve of SF6 pressure variation with temperature; A deviation rate calculation module is configured to calculate a variation rate of the relative deviation according to the relative deviation and the meter reading of the SF6 meter image in a preset period and a standard SF6 pressure; A pressure variation determination module is configured to compare the variation rate of the relative deviation with a preset rate threshold, and determine a variation trend of the SF6 pressure in the preset period according to a comparison result.

7. An electronic device, comprising: The method comprises: a memory configured to store executable instructions; and a processor configured to communicate with the memory to execute the executable instructions to complete the steps of the method of any one of claims 1 to 5.

8. A computer storage medium storing computer-readable instructions, characterized in that, The instructions, when executed, perform the steps of the method of any one of claims 1 to 5.

Images