CN117607642B - Rail transit data analysis method and system - Google Patents
Rail transit data analysis method and system Download PDFInfo
- Publication number
- CN117607642B CN117607642B CN202410089925.1A CN202410089925A CN117607642B CN 117607642 B CN117607642 B CN 117607642B CN 202410089925 A CN202410089925 A CN 202410089925A CN 117607642 B CN117607642 B CN 117607642B
- Authority
- CN
- China
- Prior art keywords
- images
- pixel
- partial discharge
- gray
- neural network
- 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.)
- Active
Links
- 238000000034 method Methods 0.000 title claims abstract description 41
- 238000007405 data analysis Methods 0.000 title claims abstract description 22
- 238000013527 convolutional neural network Methods 0.000 claims abstract description 72
- 230000008859 change Effects 0.000 claims abstract description 69
- 239000011159 matrix material Substances 0.000 claims abstract description 45
- 238000012549 training Methods 0.000 claims abstract description 27
- 230000003287 optical effect Effects 0.000 claims abstract description 15
- 238000010276 construction Methods 0.000 claims abstract description 7
- 238000004364 calculation method Methods 0.000 claims description 13
- 238000012423 maintenance Methods 0.000 claims description 12
- 230000006870 function Effects 0.000 claims description 9
- 238000013528 artificial neural network Methods 0.000 claims description 8
- 238000012937 correction Methods 0.000 claims description 3
- 238000013473 artificial intelligence Methods 0.000 abstract description 8
- 238000001514 detection method Methods 0.000 description 7
- 230000008569 process Effects 0.000 description 7
- 238000009413 insulation Methods 0.000 description 6
- 230000008901 benefit Effects 0.000 description 5
- 238000007789 sealing Methods 0.000 description 5
- 230000015556 catabolic process Effects 0.000 description 4
- 238000013135 deep learning Methods 0.000 description 4
- 238000005516 engineering process Methods 0.000 description 4
- 230000009286 beneficial effect Effects 0.000 description 3
- 238000010586 diagram Methods 0.000 description 3
- 229910018503 SF6 Inorganic materials 0.000 description 2
- 238000004458 analytical method Methods 0.000 description 2
- 238000006731 degradation reaction Methods 0.000 description 2
- 230000000694 effects Effects 0.000 description 2
- 230000002349 favourable effect Effects 0.000 description 2
- 238000004020 luminiscence type Methods 0.000 description 2
- 239000002184 metal Substances 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- SFZCNBIFKDRMGX-UHFFFAOYSA-N sulfur hexafluoride Chemical compound FS(F)(F)(F)(F)F SFZCNBIFKDRMGX-UHFFFAOYSA-N 0.000 description 2
- 241001270131 Agaricus moelleri Species 0.000 description 1
- 238000007792 addition Methods 0.000 description 1
- 230000004075 alteration Effects 0.000 description 1
- 238000000354 decomposition reaction Methods 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 238000005215 recombination Methods 0.000 description 1
- 230000006798 recombination Effects 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 230000035945 sensitivity Effects 0.000 description 1
- 229960000909 sulfur hexafluoride Drugs 0.000 description 1
- 230000001960 triggered effect Effects 0.000 description 1
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01R—MEASURING ELECTRIC VARIABLES; MEASURING MAGNETIC VARIABLES
- G01R31/00—Arrangements for testing electric properties; Arrangements for locating electric faults; Arrangements for electrical testing characterised by what is being tested not provided for elsewhere
- G01R31/12—Testing dielectric strength or breakdown voltage ; Testing or monitoring effectiveness or level of insulation, e.g. of a cable or of an apparatus, for example using partial discharge measurements; Electrostatic testing
- G01R31/1218—Testing dielectric strength or breakdown voltage ; Testing or monitoring effectiveness or level of insulation, e.g. of a cable or of an apparatus, for example using partial discharge measurements; Electrostatic testing using optical methods; using charged particle, e.g. electron, beams or X-rays
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01R—MEASURING ELECTRIC VARIABLES; MEASURING MAGNETIC VARIABLES
- G01R31/00—Arrangements for testing electric properties; Arrangements for locating electric faults; Arrangements for electrical testing characterised by what is being tested not provided for elsewhere
- G01R31/12—Testing dielectric strength or breakdown voltage ; Testing or monitoring effectiveness or level of insulation, e.g. of a cable or of an apparatus, for example using partial discharge measurements; Electrostatic testing
- G01R31/1227—Testing dielectric strength or breakdown voltage ; Testing or monitoring effectiveness or level of insulation, e.g. of a cable or of an apparatus, for example using partial discharge measurements; Electrostatic testing of components, parts or materials
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01R—MEASURING ELECTRIC VARIABLES; MEASURING MAGNETIC VARIABLES
- G01R31/00—Arrangements for testing electric properties; Arrangements for locating electric faults; Arrangements for electrical testing characterised by what is being tested not provided for elsewhere
- G01R31/12—Testing dielectric strength or breakdown voltage ; Testing or monitoring effectiveness or level of insulation, e.g. of a cable or of an apparatus, for example using partial discharge measurements; Electrostatic testing
- G01R31/1227—Testing dielectric strength or breakdown voltage ; Testing or monitoring effectiveness or level of insulation, e.g. of a cable or of an apparatus, for example using partial discharge measurements; Electrostatic testing of components, parts or materials
- G01R31/1254—Testing dielectric strength or breakdown voltage ; Testing or monitoring effectiveness or level of insulation, e.g. of a cable or of an apparatus, for example using partial discharge measurements; Electrostatic testing of components, parts or materials of gas-insulated power appliances or vacuum gaps
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/0464—Convolutional networks [CNN, ConvNet]
-
- 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
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
- G06V10/764—Arrangements for image or video recognition or understanding using pattern recognition or machine learning using classification, e.g. of video objects
- G06V10/765—Arrangements for image or video recognition or understanding using pattern recognition or machine learning using classification, e.g. of video objects using rules for classification or partitioning the feature space
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
- G06V10/82—Arrangements for image or video recognition or understanding using pattern recognition or machine learning using neural networks
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Evolutionary Computation (AREA)
- Software Systems (AREA)
- Health & Medical Sciences (AREA)
- Artificial Intelligence (AREA)
- Multimedia (AREA)
- Computing Systems (AREA)
- General Health & Medical Sciences (AREA)
- Medical Informatics (AREA)
- Databases & Information Systems (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Life Sciences & Earth Sciences (AREA)
- Biomedical Technology (AREA)
- Biophysics (AREA)
- Computational Linguistics (AREA)
- Data Mining & Analysis (AREA)
- Molecular Biology (AREA)
- General Engineering & Computer Science (AREA)
- Mathematical Physics (AREA)
- Image Analysis (AREA)
Abstract
The application belongs to the field of artificial intelligence, and particularly discloses a rail transit data analysis method and system, which aim to solve the problem that partial discharge of a GIS cabinet is difficult to find in the prior art. The invention comprises the following steps: acquiring continuous images of a plurality of groups of historical time intervals; taking the gray value of each pixel point from the continuous images, and calculating a gray change curve; construction of the firstiA pixel point gray scale change trend matrix of the group sample image; training a convolutional neural network by using pixel point gray scale change trend matrixes of a plurality of groups of sample images; installing optical sensors in a target GIS cabinet, and collecting n images in the target GIS cabinet at n continuous time nodes in a time interval T; taking the gray value of each pixel point in n images and calculating a gray change curve; and constructing a pixel gray level change trend matrix of n images, inputting a convolutional neural network, and judging whether the partial discharge phenomenon occurs in the target GIS cabinet in a time interval T. The invention can timely and accurately discover the partial discharge phenomenon in the GIS cabinet.
Description
Technical Field
The invention relates to the technical field of artificial intelligence, and more particularly relates to a rail transit data analysis method and system.
Background
The prior rail transit mostly adopts SF6 (sulfur hexafluoride) gas with excellent insulation and arc extinguishing performance as insulation and arc extinguishing medium, and takes a GIS (gas insulated enclosed metal switch) cabinet with all high-voltage electrical appliance elements sealed in a grounded metal cylinder as a main power distribution device of the rail transit.
However, the operation and maintenance research of the GIS cabinet is still in a primary stage, partial discharge detection is carried out on each component in the GIS cabinet in daily maintenance work, the partial discharge phenomenon is greatly influenced by the outside, and the related experience of the detection personnel is insufficient, so that the partial discharge phenomenon in the GIS cabinet is difficult to accurately find in time.
Disclosure of Invention
In order to solve the technical problems, namely the problem that the partial discharge of the GIS cabinet is difficult to find in the prior art, the application provides a rail transit data analysis method and a rail transit data analysis system for timely and accurately finding the partial discharge phenomenon in the GIS cabinet by means of an artificial intelligence technology.
In a first aspect, the present invention provides a method for analyzing rail traffic data, including:
acquiring a plurality of groups of sample images, wherein an ith group of sample images in the plurality of groups of sample images is in a historical time intervalContinuously recorded +.>A sheet of image, said->The GIS cabinet records +.>Continuous>Partial discharge phenomena occurring at each time node, and marking the partial discharge phenomena;
from the succession of the ith set of sample imagesTaking the gray value of each pixel point in the image;
based on the continuityPosition-corresponding continuation in a sheet of images>Gray value of each pixel point calculates gray change curve +.>Wherein->The position of any pixel point is represented, and t represents time;
constructing a pixel point gray scale change trend matrix of the ith group of sample imagesWherein a is the number of pixels of each image in the ith group of sample images in the horizontal direction, and b is the number of pixels of each image in the ith group of sample images in the vertical direction;
training a convolutional neural network by using the pixel point gray level change trend matrixes corresponding to the plurality of groups of sample images and the partial discharge phenomena marked in the plurality of groups of sample images;
installing an optical sensor in a target GIS cabinet for controlling rail transit, and acquiring n images in the target GIS cabinet through the optical sensor at n continuous time nodes in a time interval T;
taking the gray value of each pixel point in the n images;
calculating gray level change curves of the n continuous pixel points based on gray level values of the n continuous pixel points corresponding to positions in the n continuous images；
Constructing a pixel point gray scale change trend matrix of the n images；
And inputting the pixel point gray level change trend matrix of the n images into the convolutional neural network, and judging whether the partial discharge phenomenon occurs in the target GIS cabinet in the time interval T according to the output result of the convolutional neural network.
Optionally, in the aforementioned rail transit data analysis method, the gray level change curve is calculated based on gray level values of n consecutive pixel points corresponding to positions in the n consecutive imagesBefore, still include:
taking any one of the n images as an image to be corrected, selecting p areas with the pixel number higher than a preset number and the pixel gray values higher than a preset first gray value from the image to be corrected according to the gray value of each pixel in the image to be corrected, and calculating the central positions of the p areas;
taking any pixel which is positioned outside the p areas in the image to be corrected as a pixel to be corrected, and calculating the gray value attenuation distance of the pixel to be correctedWherein->For the position of the pixel to be corrected, < >>A center position of a q-th region among the p regions;
correcting the gray value of the pixel to be corrected, wherein the corrected gray valueWherein w isThe preset weight value, X is the maximum distance of the horizontal direction of the image to be corrected, Y is the maximum distance of the vertical direction of the image to be corrected, and +.>And correcting the gray value before correction for the pixel to be corrected.
Optionally, in the aforementioned rail transit data analysis method, the gray level change curve is calculated based on gray level values of n consecutive pixel points corresponding to positions in the n consecutive imagesBefore, still include:
taking any one image of the n images as an image to be filtered, and selecting any pixel with a pixel gray value lower than a preset second gray value from the image to be filtered as a pixel to be filtered;
calculating the significance level of the pixel to be filtered:wherein->To indicate the position of the pixel to be filtered, < >>Representing the locations of neighboring pixels of the pixel to be filtered,representing the gray value of said pixel to be filtered, is->Gray values of neighboring pixels representing said pixel to be filtered +.>Is an exponential function based on a natural constant e, < ->The scale coefficient is preset;
and setting the gray value of the pixel to be filtered to 0 when the significance level of the pixel to be filtered is lower than a preset level.
Optionally, in the aforementioned rail transit data analysis method, training a convolutional neural network using a pixel point gray scale change trend matrix corresponding to the plurality of groups of sample images and the partial discharge phenomenon marked in the plurality of groups of sample images includes:
setting a loss function for the convolutional neural networkWherein C is the number of training times completed by the convolutional neural network, < >>Indicating whether the convolutional neural network is marked with the partial discharge phenomenon at the kth training time,/->Indicating whether the partial discharge phenomenon exists in a sample image used by the convolutional neural network in the kth training.
Optionally, in the aforementioned rail transit data analysis method, training a convolutional neural network using a pixel point gray scale change trend matrix corresponding to the plurality of groups of sample images and the partial discharge phenomenon marked in the plurality of groups of sample images includes:
setting an initial learning rate for the convolutional neural network, wherein the initial learning rate represents an adjustment step length s of any parameter in the convolutional neural network after the convolutional neural network is trained for the first time;
after the convolutional neural network is trained for c times, calculating the current adjustment step length of any parameter in the convolutional neural networkWherein d is a preset parameter adjustment amplitude base.
Optionally, the aforementioned rail transit data analysis method further includes, before n images in the target GIS cabinet are collected by the light sensor at n consecutive time nodes in the time interval T:
installing an acoustic wave sensor in the target GIS cabinet, and detecting an acoustic wave signal in the target GIS cabinet through the acoustic wave sensor;
calculating the power difference between the currently detected sound wave signal and the previously detected sound wave signal;
when the power difference exceeds a preset power, performing: and acquiring n images in the target GIS cabinet through the continuous n time nodes of the light sensor in the time interval T.
Optionally, the aforementioned rail transit data analysis method further includes, before n images in the target GIS cabinet are collected by the light sensor at n consecutive time nodes in the time interval T:
inquiring the duration time of a plurality of historical partial discharge phenomena of the target GIS cabinet;
minimum duration of partial discharge according to the multiple historiesSetting a time interval for acquiring images in the target GIS cabinet>Said time interval->Not exceed->。
Optionally, the aforementioned method for analyzing rail traffic data further includes:
when the convolution neural network judges that the partial discharge phenomenon occurs in the target GIS cabinet in the time interval T, the position of the convolution neural network for marking the partial discharge phenomenon in the n images is obtained;
according to the positions of the partial discharge phenomena marked in the n images by the convolutional neural network, judging the positions of the partial discharge phenomena in the target GIS cabinet;
and generating operation and maintenance information according to the position of the partial discharge phenomenon in the target GIS cabinet so as to instruct operation and maintenance personnel to check the position of the partial discharge phenomenon in the target GIS cabinet.
In a second aspect, the present invention provides a rail transit data analysis system, comprising:
the sample image acquisition module acquires a plurality of groups of sample images, wherein an ith group of sample images in the plurality of groups of sample images are in a historical time intervalContinuously recorded +.>A sheet of image, saidThe GIS cabinet records +.>Continuous>Partial discharge phenomena occurring at each time node, and marking the partial discharge phenomena;
a sample gray level calculation module for calculating the successive sample images from the ith groupTaking the gray value of each pixel point in the image;
a sample curve calculation module based on the successionPosition-corresponding continuation in a sheet of images>Gray value of each pixel point is calculatedGray level change curve->Wherein->The position of any pixel point is represented, and t represents time;
the sample matrix construction module is used for constructing a pixel point gray level change trend matrix of the ith group of sample imagesWherein a is the number of pixels of each image in the ith group of sample images in the horizontal direction, and b is the number of pixels of each image in the ith group of sample images in the vertical direction;
the training module is used for training a convolutional neural network by using the pixel point gray level change trend matrix corresponding to the plurality of groups of sample images and the partial discharge phenomenon marked in the plurality of groups of sample images;
the image acquisition module is used for installing an optical sensor in a target GIS cabinet for controlling rail transit, and acquiring n images in the target GIS cabinet through n continuous time nodes of the optical sensor in a time interval T;
the gray level calculation module is used for taking the gray level value of each pixel point in the n images;
a curve calculation module for calculating gray level change curve based on gray level values of continuous n pixel points corresponding to positions in the continuous n images；
Matrix construction module for constructing pixel point gray scale variation trend matrix of n images；
And the judging module inputs the pixel point gray level change trend matrix of the n images into the convolutional neural network, and judges whether the partial discharge phenomenon occurs in the time interval T of the target GIS cabinet according to the result output by the convolutional neural network.
The technical scheme provided by the invention has at least one or more of the following beneficial effects:
(1) The difficulty of using optical detection in GIS cabinet partial discharge phenomenon detection in the prior art is that the duration time of the partial discharge phenomenon is short, the scale is small, and the full and obvious partial discharge image is difficult to detect in a closed GIS cabinet by the optical detection technology, so that the partial discharge image is difficult to find timely and accurately.
(2) It should be understood by those skilled in the art that the pixel gray scale change trend represents the relative relation of the pixel gray scale values of the image, the duration time of the partial discharge phenomenon is short, the scale is small, and the influence of the GIS cabinet environment sealing on the acquired image gray scale is counteracted and eliminated in the process of calculating the relative relation, namely, the obtained pixel gray scale change trend matrix can accurately reflect whether the GIS cabinet has the partial discharge phenomenon or not, and at the moment, the pixel gray scale change trend matrix can be analyzed by using more convolutional neural networks in the artificial intelligence field, so that whether the GIS cabinet has the partial discharge phenomenon or not can be timely and accurately judged.
Drawings
The foregoing and other objects, features and advantages of the present application will become more apparent from the following more particular description of embodiments of the present application, as illustrated in the accompanying drawings. The accompanying drawings are included to provide a further understanding of embodiments of the application and are incorporated in and constitute a part of this specification, illustrate the application and not constitute a limitation to the application. In the drawings, like reference numerals generally refer to like parts or steps.
FIG. 1 is a flow chart of a method of analyzing rail transit data according to an embodiment of the present application;
FIG. 2 is a first partial flow chart of a rail transit data analysis method according to an embodiment of the present application;
FIG. 3 is a second partial flow chart of a rail transit data analysis method according to an embodiment of the present application;
FIG. 4 is a third partial flow chart of a rail transit data analysis method according to an embodiment of the present application;
FIG. 5 is a fourth partial flow chart of a rail transit data analysis method according to an embodiment of the present application;
fig. 6 is a block diagram of a rail transit data analysis system according to an embodiment of the present application.
Detailed Description
Some embodiments of the invention are described below with reference to the accompanying drawings. It should be understood by those skilled in the art that these embodiments are merely for explaining the technical principles of the present invention, and are not intended to limit the scope of the present invention.
As shown in fig. 1, in one embodiment of the present invention, there is provided a rail transit data analysis method, including:
step S110, obtaining a plurality of groups of sample images, wherein the ith group of sample images in the plurality of groups of sample images are in a historical time intervalContinuously recorded +.>A sheet of image, said->The GIS cabinet records +.>Continuous>Partial discharge occurrence at each time nodeLike, and annotate the partial discharge phenomenon.
It will be appreciated by those skilled in the art that partial discharge is an important cause of insulation breakdown of high voltage electrical equipment, and is also an important indicator of insulation degradation. In the prior art, the problem of detecting the partial discharge phenomenon by using the optical detection technology in the GIS cabinet is that the duration time of the partial discharge phenomenon is shorter, the scale is smaller, the environment of the GIS cabinet is closed, the light is poorer, thereby the partial discharge image is difficult to be found timely and accurately,
step S120, from the succession of the ith group of sample imagesAnd taking the gray value of each pixel point in the image.
Step S130, based on the successionPosition-corresponding continuation in a sheet of images>Gray value of each pixel point calculates gray change curve +.>Wherein->The position of any one pixel point is indicated, and t indicates time.
In the present embodiment, the type of the gradation change curve is not limited, and may be, for example, a parabolic curve or an elliptic curve.
Step S140, constructing a pixel gray scale variation trend matrix of the ith group of sample imagesWherein a is the number of pixels of each image in the ith group of sample images in the horizontal direction, and b is the number of pixels of each image in the ith group of sample images in the vertical direction.
In this embodiment, the partial discharge image is not directly identified according to the image collected in the GIS cabinet, but the pixel gray level change trend matrix is calculated by continuously capturing the image of the GIS cabinet and according to the pixel gray level values at the corresponding positions, that is, the pixel gray level change trend in the continuous image is analyzed, and it should be understood by those skilled in the art that the pixel gray level change trend represents the relative relationship of the pixel gray level values between the images, and that the partial discharge phenomenon has a short duration and a small scale, and the interference of the GIS cabinet environment sealing on the image gray level is offset and eliminated in the process of calculating the relative relationship, that is, the pixel gray level change trend matrix calculated in this embodiment can accurately reflect whether the partial discharge phenomenon exists in the GIS cabinet.
And step S150, training the convolutional neural network by using the pixel point gray level change trend matrixes corresponding to the plurality of groups of sample images and the partial discharge phenomena marked in the plurality of groups of sample images.
In this embodiment, convolutional neural networks (Convolutional Neural Networks, CNN) are a type of feedforward neural network (Feedforward Neural Networks) that includes convolutional computation and has a deep structure, and are one of representative algorithms of artificial intelligence deep learning (deep learning).
Step S160, installing a light sensor in a target GIS cabinet for controlling rail transit, and acquiring n images in the target GIS cabinet through n continuous time nodes of the light sensor in a time interval T.
In this embodiment, the type of the light sensor is not limited, and may be, for example, a micro camera.
Further, the duration of the multiple historical partial discharge phenomena occurring in the GIS cabinet can be queried in advance, and the minimum duration of the multiple historical partial discharge phenomena is determinedSetting a time interval for acquiring images in the target GIS cabinet>Said time interval->Not exceed->。
In this embodiment, the time interval of the current acquired image is reasonably set based on the duration of the historical partial discharge phenomenon, which is favorable for ensuring that the continuously shot image can cover at least one complete partial discharge process, and has important value for subsequent identification.
Step S170, taking the gray value of each pixel point in the n images.
Step S180 of calculating a gray scale variation curve based on gray scale values of n consecutive pixel points corresponding to positions in the n consecutive images。
Step S190, constructing a pixel point gray scale change trend matrix of the n images。
Step S1100, inputting the pixel point gray scale change trend matrix of the n images into a convolutional neural network, and judging whether the partial discharge phenomenon occurs in the target GIS cabinet in the time interval T according to the output result of the convolutional neural network.
Further, the method may further include:
(1) When the partial discharge phenomenon of the target GIS cabinet in the time interval T is judged through the convolutional neural network, the positions of the convolutional neural network for marking the partial discharge phenomenon in n images are obtained.
(2) And judging the position of the partial discharge phenomenon in the target GIS cabinet according to the positions of the partial discharge phenomenon marked in the n images by the convolutional neural network.
In this embodiment, the partial discharge phenomenon occurring in the actual environment is located by using the standard position of the convolutional neural network in the image.
(3) And generating operation and maintenance information according to the position of the partial discharge phenomenon in the target GIS cabinet so as to indicate the operation and maintenance personnel to check the position of the partial discharge phenomenon in the target GIS cabinet.
In this embodiment, the positioning of the partial discharge phenomenon is utilized to generate reasonable and reliable operation and maintenance information, which is beneficial to timely disposing the partial discharge problem.
According to the technical scheme of the embodiment, the pixel gray level change trend matrix in the continuously acquired image is calculated by continuously shooting the images of the GIS cabinet and according to the pixel gray level values of the corresponding positions, namely, the pixel gray level change trend in the continuous image is analyzed, the pixel gray level change trend shows the relative relation of the pixel gray level values of the image, the partial discharge phenomenon is short in duration and small in scale, and the influence of the GIS cabinet environment sealing on the acquired image gray level is counteracted and eliminated in the process of calculating the relative relation, namely, whether the partial discharge phenomenon exists in the GIS cabinet can be accurately reflected by the pixel gray level change trend matrix, and at the moment, the partial discharge phenomenon can be timely and accurately judged by analyzing the pixel gray level change trend matrix by using more convolutional neural networks in the artificial intelligence field.
As shown in fig. 2, in an embodiment of the present invention, compared to the foregoing embodiment, the method for analyzing rail traffic data in the present embodiment further includes, before step S130:
step S210, taking any one of the n images as an image to be corrected, selecting p areas with the pixel number higher than the preset number and the pixel gray value higher than the preset first gray value from the image to be corrected according to the gray value of each pixel in the image to be corrected, and calculating the central positions of the p areas.
It will be appreciated by those skilled in the art that the partial discharge phenomenon produces a luminescence phenomenon in a partial region, and other regions are affected by the luminescence region to increase the light intensity. In this embodiment, p regions in which the light emission phenomenon occurs are identified and extracted from the captured image based on the characteristic of high luminance of the light emission region.
Step S220, using any pixel outside the p areas in the image to be corrected as the pixel to be correctedCalculating gray value attenuation distance of pixel to be correctedWherein->For the position of the pixel to be corrected, < >>Is the center position of the q-th region of the p regions.
In this embodiment, the farther the distance between the other regions and the light-emitting region is, the smaller the light intensity increasing level is, and the gray value attenuation distance calculated according to the above formula is calculated, so as to reasonably estimate the influence distance of the light-emitting region on the non-light-emitting region by calculating the minimum distance between the gray value attenuation distance and the center positions of the p light-emitting regions.
Step S230, correcting the gray value of the pixel to be corrected, the corrected gray valueWherein w is a preset weight value, X is the maximum distance in the horizontal direction of the image to be corrected, Y is the maximum distance in the vertical direction of the image to be corrected, +.>The gray value before correction for the pixel to be corrected.
According to the technical scheme of the embodiment, the spatial size reflected by the shot image is considered based on the maximum distance of the horizontal direction and the vertical direction of the image in the formula, the pixel gray of the image is corrected by combining the attenuation distance of the combined gray value, the pixel gray in the area less influenced by the light-emitting area is reduced, the change of the pixel gray in the image is increased, a more remarkable gray change curve is calculated, and the subsequent analysis and identification are facilitated.
As shown in fig. 3, in an embodiment of the present invention, compared to the foregoing embodiment, the method for analyzing rail traffic data in the present embodiment further includes, before step S130:
step S310, taking any one of the n images as an image to be filtered, and selecting any pixel with a pixel gray value lower than a preset second gray value from the image to be filtered as a pixel to be filtered.
In this embodiment, the pixels with lower gray values are often located in the non-light-emitting area less affected by the light-emitting area, and the gray scales of the pixels in the non-light-emitting area in the captured image are also different due to the influence of external factors, but actually the brightness of the non-light-emitting area is very close.
Step S320, calculating the saliency level of the pixel to be filtered:wherein->To indicate the position of the pixel to be filtered, < >>Representing the locations of neighboring pixels of the pixel to be filtered,representing the gray value of said pixel to be filtered, is->Gray values of neighboring pixels representing said pixel to be filtered +.>Is an exponential function based on a natural constant e, < ->Is a preset scale factor.
In this embodiment, the saliency level of the pixel to be filtered is calculated based on the gray value difference between the pixel to be filtered and other adjacent pixels, and the saliency level actually reflects the influence of the light-emitting area on the pixel.
In step S330, when the saliency level of the pixel to be filtered is lower than the preset level, the gray value of the pixel to be filtered is set to 0.
According to the technical scheme of the embodiment, when the significance level of the pixel to be filtered is low, that is, the gray scale of the pixel to be filtered and the adjacent pixels thereof is less affected by the light emitting area, all pixels in the area where the pixel to be filtered is located should be at the same low level, so that the gray scale value of the pixel to be filtered is set to 0.
In one embodiment of the present invention, compared to the foregoing embodiment, the method for analyzing rail traffic data in this embodiment, step S150 includes:
setting a loss function for the convolutional neural networkWherein C is the number of training times completed by the convolutional neural network, < >>Indicating whether the convolutional neural network is marked with the partial discharge phenomenon at the kth training time,/->Indicating whether the partial discharge phenomenon exists in a sample image used by the convolutional neural network in the kth training.
In this embodiment, an innovative loss function for recognition and analysis of partial discharge phenomenon is provided, where the loss function is characterized by lightweight calculation, and meanwhile, the loss function is more focused on the partial discharge phenomenon that is not successfully marked in training but actually exists along with the increase of training times, so that the sensitivity of the convolutional neural network after training to the partial discharge phenomenon is improved.
As shown in fig. 4, in an embodiment of the present invention, compared to the aforementioned embodiment, a method for analyzing rail transit data is provided, and step S150 includes:
step S410, an initial learning rate is set for the convolutional neural network, wherein the initial learning rate represents an adjustment step length S of any parameter in the convolutional neural network after the convolutional neural network is trained for the first time.
Step S420, after the convolutional neural network is trained c times, calculating the current adjustment step length of any parameter in the convolutional neural networkWherein d is a preset parameter adjustment amplitude base.
According to the technical scheme of the embodiment, an innovative step length adjusting formula is designed, so that the step length is adjusted to be stable after being increased along with the increase of the training times, and the improvement of the training efficiency and stability of the convolutional neural network is facilitated.
As shown in fig. 5, in an embodiment of the present invention, compared to the foregoing embodiment, the method for analyzing rail traffic data in the present embodiment further includes, before step S160:
and S510, installing an acoustic wave sensor in the target GIS cabinet, and detecting an acoustic wave signal in the target GIS cabinet through the acoustic wave sensor.
In step S520, a power difference between the currently detected acoustic wave signal and the previously detected acoustic wave signal is calculated.
It will be appreciated by those skilled in the art that when a partial discharge occurs in a target GIS cabinet, the acoustic signal inside it will change significantly.
Step S530, when the power difference exceeds the preset power, step S160 is performed.
According to the technical scheme of the embodiment, the image acquisition of the internal environment of the GIS cabinet is triggered by utilizing the change of the acoustic wave signals generated by the partial discharge phenomenon, so that continuous shooting of the interior of the GIS cabinet for a long time is avoided, and the resource consumption and the equipment loss are reduced.
As shown in fig. 6, in one embodiment of the present invention, there is provided a rail transit data analysis system including:
the sample image acquisition module 610 acquires multiple sets of sample images, whichIn which the ith group of sample images in the plurality of groups of sample images is in a historical time intervalContinuously recorded +.>A sheet of image, said->The GIS cabinet records +.>Continuous>And marking the partial discharge phenomenon of each time node.
It will be appreciated by those skilled in the art that partial discharge is an important cause of insulation breakdown of high voltage electrical equipment, and is also an important indicator of insulation degradation. In the prior art, the problem of detecting the partial discharge phenomenon by using the optical detection technology in the GIS cabinet is that the duration time of the partial discharge phenomenon is shorter, the scale is smaller, the environment of the GIS cabinet is closed, the light is poorer, thereby the partial discharge image is difficult to be found timely and accurately,
sample gray level calculation module 620, from the succession of the i-th set of sample imagesAnd taking the gray value of each pixel point in the image.
A sample curve calculation module 630 based on the successionPosition-corresponding continuation in a sheet of images>Gray value of each pixel point calculates gray change curve +.>Wherein->The position of any one pixel point is indicated, and t indicates time.
In the present embodiment, the type of the gradation change curve is not limited, and may be, for example, a parabolic curve or an elliptic curve.
Sample matrix construction module 640 for constructing a pixel gray scale variation trend matrix of the ith group of sample imagesWherein a is the number of pixels of each image in the ith group of sample images in the horizontal direction, and b is the number of pixels of each image in the ith group of sample images in the vertical direction.
In this embodiment, the partial discharge image is not directly identified according to the image collected in the GIS cabinet, but the pixel gray level change trend matrix is calculated by continuously capturing the image of the GIS cabinet and according to the pixel gray level values at the corresponding positions, that is, the pixel gray level change trend in the continuous image is analyzed, and it should be understood by those skilled in the art that the pixel gray level change trend represents the relative relationship of the pixel gray level values between the images, and that the partial discharge phenomenon has a short duration and a small scale, and the interference of the GIS cabinet environment sealing on the image gray level is offset and eliminated in the process of calculating the relative relationship, that is, the pixel gray level change trend matrix calculated in this embodiment can accurately reflect whether the partial discharge phenomenon exists in the GIS cabinet.
The training module 650 trains the convolutional neural network by using the pixel gray scale change trend matrix corresponding to the plurality of groups of sample images and the partial discharge phenomena marked in the plurality of groups of sample images.
In this embodiment, convolutional neural networks (Convolutional Neural Networks, CNN) are a type of feedforward neural network (Feedforward Neural Networks) that includes convolutional computation and has a deep structure, and are one of representative algorithms of artificial intelligence deep learning (deep learning).
The image acquisition module 660 is used for installing an optical sensor in a target GIS cabinet for controlling rail transit, and acquiring n images in the target GIS cabinet through n continuous time nodes of the optical sensor in a time interval T.
In this embodiment, the type of the light sensor is not limited, and may be, for example, a micro camera.
Further, the duration of the multiple historical partial discharge phenomena occurring in the GIS cabinet can be queried in advance, and the minimum duration of the multiple historical partial discharge phenomena can be determinedSetting a time interval for acquiring images in the target GIS cabinet>Said time interval->Not exceed->。
In this embodiment, the time interval of the current acquired image is reasonably set based on the duration of the historical partial discharge phenomenon, which is favorable for ensuring that the continuously shot image can cover at least one complete partial discharge process, and has important value for subsequent identification.
The gray level calculation module 670 takes the gray level value of each pixel point in the n images.
A curve calculation module 680 for calculating a gray scale variation curve based on gray scale values of n consecutive pixel points corresponding to positions in the n consecutive images。
Matrix construction module 690 constructs a pixel gray scale variation trend matrix of the n images。
The judging module 6100 inputs the pixel point gray scale change trend matrix of the n images into the convolutional neural network, and judges whether the partial discharge phenomenon occurs in the target GIS cabinet in the time interval T according to the output result of the convolutional neural network.
Further, the method may further include:
(1) When the partial discharge phenomenon of the target GIS cabinet in the time interval T is judged through the convolutional neural network, the positions of the convolutional neural network for marking the partial discharge phenomenon in n images are obtained.
(2) And judging the position of the partial discharge phenomenon in the GIS cabinet according to the positions of the partial discharge phenomenon marked in the n images by the convolutional neural network.
In this embodiment, the partial discharge phenomenon occurring in the actual environment is located by using the standard position of the convolutional neural network in the image.
(3) And generating operation and maintenance information according to the position of the partial discharge phenomenon in the GIS cabinet so as to instruct operation and maintenance personnel to check the position of the partial discharge phenomenon in the GIS cabinet.
In this embodiment, the positioning of the partial discharge phenomenon is utilized to generate reasonable and reliable operation and maintenance information, which is beneficial to timely disposing the partial discharge problem.
According to the technical scheme of the embodiment, the pixel gray level change trend matrix in the continuously acquired image is calculated by continuously shooting the images of the GIS cabinet and according to the pixel gray level values of the corresponding positions, namely, the pixel gray level change trend in the continuous image is analyzed, the pixel gray level change trend shows the relative relation of the pixel gray level values of the image, the partial discharge phenomenon is short in duration and small in scale, and the influence of the GIS cabinet environment sealing on the acquired image gray level is counteracted and eliminated in the process of calculating the relative relation, namely, whether the partial discharge phenomenon exists in the GIS cabinet can be accurately reflected by the pixel gray level change trend matrix, and at the moment, the partial discharge phenomenon can be timely and accurately judged by analyzing the pixel gray level change trend matrix by using more convolutional neural networks in the artificial intelligence field.
The basic principles of the present application have been described above in connection with specific embodiments, however, it should be noted that the advantages, benefits, effects, etc. mentioned in the present application are merely examples and not limiting, and these advantages, benefits, effects, etc. are not to be considered as necessarily possessed by the various embodiments of the present application. Furthermore, the specific details disclosed herein are for purposes of illustration and understanding only, and are not intended to be limiting, as the application is not intended to be limited to the details disclosed herein as such.
The block diagrams of the devices, apparatuses, devices, systems referred to in this application are only illustrative examples and are not intended to require or imply that the connections, arrangements, configurations must be made in the manner shown in the block diagrams. As will be appreciated by one of skill in the art, the devices, apparatuses, devices, systems may be connected, arranged, configured in any manner. Words such as "including," "comprising," "having," and the like are words of openness and mean "including but not limited to," and are used interchangeably therewith. The terms "or" and "as used herein refer to and are used interchangeably with the term" and/or "unless the context clearly indicates otherwise. The term "such as" as used herein refers to, and is used interchangeably with, the phrase "such as, but not limited to.
It is also noted that in the apparatus, devices and methods of the present application, the components or steps may be disassembled and/or assembled. Such decomposition and/or recombination should be considered as equivalent to the present application.
The previous description of the disclosed aspects is provided to enable any person skilled in the art to make or use the present application. Various modifications to these aspects will be readily apparent to those skilled in the art, and the generic principles defined herein may be applied to other aspects without departing from the scope of the application. Thus, the present application is not intended to be limited to the aspects shown herein but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.
The foregoing description has been presented for purposes of illustration and description. Furthermore, this description is not intended to limit the embodiments of the application to the form disclosed herein. Although a number of example aspects and embodiments have been discussed above, a person of ordinary skill in the art will recognize certain variations, modifications, alterations, additions, and subcombinations thereof.
Claims (9)
1. A method of analyzing rail transit data, comprising:
acquiring a plurality of groups of sample images, wherein an ith group of sample images in the plurality of groups of sample images is in a historical time intervalContinuously recorded +.>A sheet of image, said->The GIS cabinet records +.>Continuous>Partial discharge phenomena occurring at each time node, and marking the partial discharge phenomena;
from the succession of the ith set of sample imagesTaking the gray value of each pixel point in the image;
based on the continuityPosition-corresponding continuation in a sheet of images>Gray value of each pixel point calculates gray change curve +.>Wherein->The position of any pixel point is represented, and t represents time;
constructing a pixel point gray scale change trend matrix of the ith group of sample imagesWherein a is the number of pixels of each image in the ith group of sample images in the horizontal direction, and b is the number of pixels of each image in the ith group of sample images in the vertical direction;
training a convolutional neural network by using the pixel point gray level change trend matrixes corresponding to the plurality of groups of sample images and the partial discharge phenomena marked in the plurality of groups of sample images;
installing an optical sensor in a target GIS cabinet for controlling rail transit, and acquiring n images in the target GIS cabinet through the optical sensor at n continuous time nodes in a time interval T;
taking the gray value of each pixel point in the n images;
calculating gray level change curves of the n continuous pixel points based on gray level values of the n continuous pixel points corresponding to positions in the n continuous images；
Constructing a pixel point gray scale change trend matrix of the n images；
And inputting the pixel point gray level change trend matrix of the n images into the convolutional neural network, and judging whether the partial discharge phenomenon occurs in the target GIS cabinet in the time interval T according to the output result of the convolutional neural network.
2. The method according to claim 1, wherein a gradation change curve is calculated based on gradation values of consecutive n pixel points corresponding to positions in consecutive n imagesBefore, still include:
taking any one of the n images as an image to be corrected, selecting p areas with the pixel number higher than a preset number and the pixel gray values higher than a preset first gray value from the image to be corrected according to the gray value of each pixel in the image to be corrected, and calculating the central positions of the p areas;
taking any pixel which is positioned outside the p areas in the image to be corrected as a pixel to be corrected, and calculating the gray value attenuation distance of the pixel to be correctedWherein->For the position of the pixel to be corrected, < >>A center position of a q-th region among the p regions;
correcting the gray value of the pixel to be corrected, wherein the corrected gray valueWherein w is a preset weight value, X is the maximum distance in the horizontal direction of the image to be corrected, Y is the maximum distance in the vertical direction of the image to be corrected, and->And correcting the gray value before correction for the pixel to be corrected.
3. According to claim 1Wherein the gray scale change curve is calculated based on gray scale values of n consecutive pixel points corresponding to positions in the n consecutive imagesBefore, still include:
taking any one image of the n images as an image to be filtered, and selecting any pixel with a pixel gray value lower than a preset second gray value from the image to be filtered as a pixel to be filtered;
calculating the significance level of the pixel to be filtered:wherein->To indicate the position of the pixel to be filtered, < >>Representing the position of the neighboring pixels of said pixel to be filtered,/or->Representing the gray value of said pixel to be filtered, is->Gray values of neighboring pixels representing said pixel to be filtered +.>Is an exponential function based on a natural constant e, < ->The scale coefficient is preset;
and setting the gray value of the pixel to be filtered to 0 when the significance level of the pixel to be filtered is lower than a preset level.
4. The method according to claim 1, wherein training a convolutional neural network using the pixel point gray scale change trend matrix corresponding to the plurality of sets of sample images and the partial discharge phenomenon noted in the plurality of sets of sample images, comprises:
setting a loss function for the convolutional neural networkWherein C is the number of training times completed by the convolutional neural network, < >>Indicating whether the convolutional neural network is marked with the partial discharge phenomenon at the kth training time,/->Indicating whether the partial discharge phenomenon exists in a sample image used by the convolutional neural network in the kth training.
5. The method according to claim 1, wherein training a convolutional neural network using the pixel point gray scale change trend matrix corresponding to the plurality of sets of sample images and the partial discharge phenomenon noted in the plurality of sets of sample images, comprises:
setting an initial learning rate for the convolutional neural network, wherein the initial learning rate represents an adjustment step length s of any parameter in the convolutional neural network after the convolutional neural network is trained for the first time;
after the convolutional neural network is trained for c times, calculating the current adjustment step length of any parameter in the convolutional neural networkWherein d is a preset parameter adjustment amplitude base.
6. The method of analyzing rail transit data according to claim 1, further comprising, before n images in the target GIS cabinet are acquired by the light sensor at n consecutive time nodes within a time interval T:
installing an acoustic wave sensor in the target GIS cabinet, and detecting an acoustic wave signal in the target GIS cabinet through the acoustic wave sensor;
calculating the power difference between the currently detected sound wave signal and the previously detected sound wave signal;
when the power difference exceeds a preset power, performing: and acquiring n images in the target GIS cabinet through the continuous n time nodes of the light sensor in the time interval T.
7. The method of analyzing rail transit data according to claim 1, further comprising, before n images in the target GIS cabinet are acquired by the light sensor at n consecutive time nodes within a time interval T:
inquiring the duration time of a plurality of historical partial discharge phenomena of the target GIS cabinet;
minimum duration of partial discharge according to the multiple historiesSetting a time interval for acquiring images in the target GIS cabinet>Said time interval->Not exceed->。
8. The rail transit data analysis method as claimed in claim 1, further comprising:
when the convolution neural network judges that the partial discharge phenomenon occurs in the target GIS cabinet in the time interval T, the position of the convolution neural network for marking the partial discharge phenomenon in the n images is obtained;
according to the positions of the partial discharge phenomena marked in the n images by the convolutional neural network, judging the positions of the partial discharge phenomena in the target GIS cabinet;
and generating operation and maintenance information according to the position of the partial discharge phenomenon in the target GIS cabinet so as to instruct operation and maintenance personnel to check the position of the partial discharge phenomenon in the target GIS cabinet.
9. A rail transit data analysis system, comprising:
the sample image acquisition module acquires a plurality of groups of sample images, wherein an ith group of sample images in the plurality of groups of sample images are in a historical time intervalContinuously recorded +.>A sheet of image, said->The GIS cabinet records +.>Continuous>Partial discharge phenomena occurring at each time node, and marking the partial discharge phenomena;
a sample gray level calculation module for calculating the successive sample images from the ith groupTaking the gray of each pixel point in the imageA degree value;
a sample curve calculation module based on the successionPosition-corresponding continuation in a sheet of images>Gray value of each pixel point calculates gray change curve +.>Wherein->The position of any pixel point is represented, and t represents time;
the sample matrix construction module is used for constructing a pixel point gray level change trend matrix of the ith group of sample imagesWherein a is the number of pixels of each image in the ith group of sample images in the horizontal direction, and b is the number of pixels of each image in the ith group of sample images in the vertical direction;
the training module is used for training a convolutional neural network by using the pixel point gray level change trend matrix corresponding to the plurality of groups of sample images and the partial discharge phenomenon marked in the plurality of groups of sample images;
the image acquisition module is used for installing an optical sensor in a target GIS cabinet for controlling rail transit, and acquiring n images in the target GIS cabinet through n continuous time nodes of the optical sensor in a time interval T;
the gray level calculation module is used for taking the gray level value of each pixel point in the n images;
a curve calculation module for calculating gray level change curve based on gray level values of continuous n pixel points corresponding to positions in the continuous n images；
Matrix construction module for constructing pixel point gray scale variation trend matrix of n images；
And the judging module inputs the pixel point gray level change trend matrix of the n images into the convolutional neural network, and judges whether the partial discharge phenomenon occurs in the time interval T of the target GIS cabinet according to the result output by the convolutional neural network.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410089925.1A CN117607642B (en) | 2024-01-23 | 2024-01-23 | Rail transit data analysis method and system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410089925.1A CN117607642B (en) | 2024-01-23 | 2024-01-23 | Rail transit data analysis method and system |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN117607642A CN117607642A (en) | 2024-02-27 |
| CN117607642B true CN117607642B (en) | 2024-03-26 |
Family
ID=89958261
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202410089925.1A Active CN117607642B (en) | 2024-01-23 | 2024-01-23 | Rail transit data analysis method and system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117607642B (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN118090808A (en) * | 2024-04-28 | 2024-05-28 | 陕西润泽博泽科技有限公司 | Oil seal thermal stability detection method based on data processing |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110633690A (en) * | 2019-09-24 | 2019-12-31 | 北京邮电大学 | Vehicle feature identification method and system based on bridge monitoring |
| CN111242046A (en) * | 2020-01-15 | 2020-06-05 | 江苏北斗星通汽车电子有限公司 | Ground traffic sign identification method based on image retrieval |
| CN112364844A (en) * | 2021-01-12 | 2021-02-12 | 北京三维天地科技股份有限公司 | Data acquisition method and system based on computer vision technology |
| CN116168028A (en) * | 2023-04-25 | 2023-05-26 | 中铁电气化局集团有限公司 | High-speed rail original image processing method and system based on edge filtering under low visibility |
| WO2023213332A1 (en) * | 2022-06-27 | 2023-11-09 | 上海格鲁布科技有限公司 | Separation and identification method for multi-source hybrid ultra-high frequency partial discharge diagram |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| DE102021124670B3 (en) * | 2021-09-23 | 2023-01-26 | Audi Aktiengesellschaft | Testing device for locating a partial discharge in or on an electrical component and method for locating the partial discharge |
-
2024
- 2024-01-23 CN CN202410089925.1A patent/CN117607642B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110633690A (en) * | 2019-09-24 | 2019-12-31 | 北京邮电大学 | Vehicle feature identification method and system based on bridge monitoring |
| CN111242046A (en) * | 2020-01-15 | 2020-06-05 | 江苏北斗星通汽车电子有限公司 | Ground traffic sign identification method based on image retrieval |
| CN112364844A (en) * | 2021-01-12 | 2021-02-12 | 北京三维天地科技股份有限公司 | Data acquisition method and system based on computer vision technology |
| WO2023213332A1 (en) * | 2022-06-27 | 2023-11-09 | 上海格鲁布科技有限公司 | Separation and identification method for multi-source hybrid ultra-high frequency partial discharge diagram |
| CN116168028A (en) * | 2023-04-25 | 2023-05-26 | 中铁电气化局集团有限公司 | High-speed rail original image processing method and system based on edge filtering under low visibility |
Also Published As
| Publication number | Publication date |
|---|---|
| CN117607642A (en) | 2024-02-27 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN117607642B (en) | Rail transit data analysis method and system | |
| CN107655159B (en) | Air conditioner energy-saving control method and system based on crowd density sensing model | |
| CN102891960B (en) | Method and camera for determining an image adjustment parameter | |
| CN100433832C (en) | Image processing apparatus and image processing method | |
| CN113887412B (en) | Detection method, detection terminal, monitoring system and storage medium for pollution emission | |
| CN113436184B (en) | Power equipment image defect discriminating method and system based on improved twin network | |
| CN112067616B (en) | Real-time detection device and method for grain quality | |
| CN111401246A (en) | Smoke concentration detection method, device, equipment and storage medium | |
| KR102391853B1 (en) | System and Method for Processing Image Informaion | |
| CN113065484A (en) | Insulator contamination state assessment method based on ultraviolet spectrum | |
| CN113674280A (en) | Method for measuring temperature of hearth of power station boiler | |
| CN100590645C (en) | Hit point calculation method and apparatus based on topology grid learning | |
| KR20100118368A (en) | Fire detecting method using hidden markov models in video surveillance and monitoring system | |
| CN116612438A (en) | Steam boiler combustion state real-time monitoring system based on thermal imaging | |
| CN112668754A (en) | Power equipment defect diagnosis method based on multi-source characteristic information fusion | |
| CN115912183B (en) | Ecological measure inspection method and system for high-voltage transmission line and readable storage medium | |
| CN109919936B (en) | Method, device and equipment for analyzing running state of composite insulator | |
| CN201155944Y (en) | Ball firing system | |
| CN112085096A (en) | Method for detecting local abnormal heating of object based on transfer learning | |
| CN114241792B (en) | Traffic flow detection method and system | |
| CN113963013B (en) | Markless power transmission tower displacement vibration identification method based on computer vision | |
| CN112255141B (en) | Thermal imaging gas monitoring system | |
| CN111145219A (en) | Efficient video moving target detection method based on Codebook principle | |
| KR101565829B1 (en) | Apparatus and method for object tracking in multiple non-overlapping cameras | |
| CN113702786B (en) | K-means-based multi-parameter suspension insulator insulation state evaluation method |
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 |