CN114067212A - Intelligent detection method and device based on temperature vision and electronic equipment - Google Patents
Intelligent detection method and device based on temperature vision and electronic equipment Download PDFInfo
- Publication number
- CN114067212A CN114067212A CN202111413561.0A CN202111413561A CN114067212A CN 114067212 A CN114067212 A CN 114067212A CN 202111413561 A CN202111413561 A CN 202111413561A CN 114067212 A CN114067212 A CN 114067212A
- Authority
- CN
- China
- Prior art keywords
- target
- power equipment
- image
- training
- temperature
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/214—Generating training patterns; Bootstrap methods, e.g. bagging or boosting
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
-
- 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
- G06N3/045—Combinations of networks
-
- 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
- G06N3/08—Learning methods
Abstract
The invention provides an intelligent detection method and device based on temperature vision and electronic equipment, and relates to the technical field of computer vision. Wherein the method comprises the following steps: acquiring a temperature visual image of target power equipment; inputting the temperature visual image into a target detection model generated by pre-training so as to acquire type information of the target power equipment; determining a target area image and a sub-classification model corresponding to the target electric power equipment according to the type information and the position information of the target electric power equipment; inputting the target area image into the sub-classification model to obtain an image to be detected containing each part frame marked in the target area image; and inputting the image to be detected into a fault detection model generated by pre-training so as to determine the fault type and the fault position of the target power equipment. Therefore, the positions of the power equipment can be positioned through the target detection network based on the temperature visual information, and the positions of the components of the power equipment can be positioned through the characteristic point network, so that the positions of the components of the power equipment can be accurately positioned and corresponding temperature information can be acquired.
Description
Technical Field
The invention relates to the technical field of computer vision, in particular to an intelligent detection method and device based on temperature vision and electronic equipment.
Background
At present, a method for detecting electric equipment mainly uses an infrared image as a research basis to research the problem of abnormal heating of the electric equipment. The infrared picture comprises various pseudo colors, the shooting environment is complex, the interference of the shot equipment is serious, the training data types are few, and the like, so that the existing method has the defects of low detection accuracy, poor model generalization capability and the like.
Therefore, how to accurately and reliably detect the power equipment is a problem which needs to be solved at present.
Disclosure of Invention
The present invention is directed to solving, at least to some extent, one of the technical problems in the related art.
The embodiment of the first aspect of the invention provides an intelligent detection method based on temperature vision, which comprises the following steps:
acquiring a temperature visual image of target power equipment;
inputting the temperature visual image into a target detection model generated by pre-training so as to acquire type information of the target power equipment;
determining a target area image and a sub-classification model corresponding to the target electric power equipment according to the type information and the position information of the target electric power equipment;
inputting the target area image into the sub-classification model to obtain an image to be detected containing each part frame marked in the target area image;
and inputting the image to be detected into a fault detection model generated by pre-training so as to determine the fault type and the fault position of the target power equipment.
Optionally, before the inputting the training data set into a target detection model generated by pre-training to obtain type information and location information of the target electrical device, the method further includes:
acquiring a pre-training data set of a target electric power device, wherein the pre-training data set is a temperature visual data set of the target electric power device;
labeling the temperature visual data set according to a specified mode, wherein the specified mode at least comprises equipment frame labeling and characteristic point labeling;
and training the initial detection model by using the marked temperature visual data set to obtain a target detection model.
Optionally, before determining the target area image and the sub-classification model corresponding to the target electrical device according to the type information and the location information of the target electrical device, the method further includes:
cutting each temperature visual image in a pre-training data set of target power equipment to obtain a training data set containing a labeled area;
classifying the training data set according to the type of the target power equipment to obtain a plurality of training data subsets;
and training a plurality of corresponding initial sub-classification models by using the plurality of training data subsets to obtain a plurality of trained sub-classification models.
Optionally, the determining, according to the type information and the location information of the target electrical device, a target area image and a sub-classification model corresponding to the target electrical device includes:
according to the position information of the target power equipment, cutting a temperature visual image of the target power equipment to obtain a target area image;
and determining a sub-classification model corresponding to the current target electric power equipment based on the preset corresponding relation between the type information of the target electric power equipment and the sub-classification model.
Optionally, before the inputting the image to be detected to the fault detection model generated by pre-training, the method further includes:
acquiring two classification temperature data sets divided according to fault equipment and non-fault equipment;
and training an initial fault diagnosis model by using the two classified temperature data sets to generate a fault detection model.
Optionally, the method further includes:
inputting the image to be detected into a fault detection model generated by training to obtain a label corresponding to the image to be detected, wherein the label comprises a fault type label and a fault position label;
inputting the image to be detected into an initial fault diagnosis model to be trained so as to obtain a prediction label corresponding to the image to be detected;
and respectively correcting the initial fault diagnosis model and the fault detection model according to the difference between the prediction label and the labeling label.
The embodiment of the second aspect of the invention provides an intelligent detection device based on temperature vision, which comprises:
the first acquisition module is used for acquiring a temperature visual image of the target power equipment;
the second acquisition module is used for inputting the temperature visual image into a target detection model generated by pre-training so as to acquire the type information of the target power equipment;
the first determining module is used for determining a target area image and a sub-classification model corresponding to the target electric power equipment according to the type information and the position information of the target electric power equipment;
a third obtaining module, configured to input the target area image into the sub-classification model, so as to obtain an image to be detected that includes each part frame marked in the target area image;
and the second determining module is used for inputting the image to be detected into a fault detection model generated by pre-training so as to determine the fault type and the fault position of the target power equipment.
Optionally, the second obtaining module is further configured to:
acquiring a pre-training data set of a target electric power device, wherein the pre-training data set is a temperature visual data set of the target electric power device;
labeling the temperature visual data set according to a specified mode, wherein the specified mode at least comprises equipment frame labeling and characteristic point labeling;
and training the initial detection model by using the marked temperature visual data set to obtain a target detection model.
Optionally, the first determining module is further configured to:
cutting each temperature visual image in a pre-training data set of target power equipment to obtain a training data set containing a labeled area;
classifying the training data set according to the type of the target power equipment to obtain a plurality of training data subsets;
and training a plurality of corresponding initial sub-classification models by using the plurality of training data subsets to obtain a plurality of trained sub-classification models.
Optionally, the first determining module is specifically configured to:
according to the position information of the target power equipment, cutting a temperature visual image of the target power equipment to obtain a target area image;
and determining a sub-classification model corresponding to the current target electric power equipment based on the preset corresponding relation between the type information of the target electric power equipment and the sub-classification model.
Optionally, the second determining module is further configured to:
acquiring two classification temperature data sets divided according to fault equipment and non-fault equipment;
and training an initial fault diagnosis model by using the two classified temperature data sets to generate a fault detection model.
Optionally, the apparatus further comprises:
the third acquisition module is used for inputting the image to be detected into a fault detection model generated by training so as to acquire a label corresponding to the image to be detected, wherein the label comprises a fault type label and a fault position label;
the fourth acquisition module is used for inputting the image to be detected into the initial fault diagnosis model to be trained so as to acquire a prediction label corresponding to the image to be detected;
and the correction module is used for respectively correcting the initial fault diagnosis model and the fault detection model according to the difference between the prediction label and the labeling label.
A third embodiment of the present invention provides a non-transitory computer-readable storage medium storing a computer program, which when executed by a processor implements the intelligent detection method based on temperature vision as set forth in the first embodiment of the present invention.
A fourth aspect of the present invention provides a computer program product, which when executed by an instruction processor performs the method for intelligent detection based on temperature vision according to the first aspect of the present invention.
In the embodiment of the invention, a temperature visual image of target power equipment is firstly acquired, then the temperature visual image is input into a target detection model generated by pre-training to acquire type information of the target power equipment, then a target area image and a sub-classification model corresponding to the target power equipment are determined according to the type information and the position information of the target power equipment, then the target area image is input into the sub-classification model to acquire an image to be detected including each component frame marked in the target area image, and finally the image to be detected is input into a fault detection model generated by pre-training to determine the fault type and the fault position of the target power equipment. Therefore, the target detection model can be trained by using the temperature visual data to acquire the type and the position of the equipment in the temperature visual image, the component position of the power equipment in the image can be locked more accurately, and the fault of the power equipment can be diagnosed intelligently. That is, based on the temperature visual information, the position of the power equipment can be located through the target detection network, and the positions of the components of the power equipment can be located through the characteristic point network, so that the positions of the components of the power equipment can be accurately located and corresponding temperature information can be acquired.
Additional aspects and advantages of the invention will be set forth in part in the description which follows and, in part, will be obvious from the description, or may be learned by practice of the invention.
Drawings
Fig. 1 is a schematic flow chart of an intelligent detection method based on temperature vision according to an embodiment of the present invention;
fig. 2 is a schematic structural diagram of an intelligent detection apparatus based on temperature vision according to an embodiment of the present invention;
FIG. 3 illustrates a block diagram of an exemplary electronic device suitable for use in implementing embodiments of the present invention.
Detailed Description
Reference will now be made in detail to embodiments of the present invention, examples of which are illustrated in the accompanying drawings, wherein like or similar reference numerals refer to the same or similar elements or elements having the same or similar function throughout. The embodiments described below with reference to the drawings are illustrative and intended to be illustrative of the invention and are not to be construed as limiting the invention.
The following describes an intelligent detection method, an intelligent detection device and electronic equipment based on temperature vision according to embodiments of the invention with reference to the accompanying drawings. The intelligent detection method based on the temperature vision in the embodiment of the invention can be executed by the intelligent detection device based on the temperature vision in the embodiment of the invention, and the device can be configured in electronic equipment.
For convenience of description, the intelligent detection device based on temperature vision may be referred to as "the device".
Fig. 1 is a schematic flow chart of an intelligent detection method based on temperature vision according to an embodiment of the present invention. As shown in fig. 1, the intelligent detection method based on temperature vision may include the following steps:
Specifically, the thermal infrared imager may be used to collect information of the target power device to obtain a temperature visual image of the target power device.
Optionally, a pre-training data set of the target electrical device may be obtained first, where the pre-training data set is a temperature visual data set of the target electrical device.
The target power equipment may be equipment to be diagnosed currently, and may include various types of power equipment, for example, a GIS air chamber, a neutral point voltage transformer, a JP cabinet, a bushing, a low-voltage pile head, a damping resistor box, a junction filter, a high-voltage side bushing, a lighting box, a metal wire, a wire inlet bin, a gathering box, a pump machine, an insulator, a gathering box, a GIS bushing, a combined transformer, a transfer box, a lightning arrester, a coupling transformer, a junction box, a transformer, a storage battery, a mechanism box, a switch cabinet, a filter, a power box, a wall bushing, a lightning rod, a terminal cabinet, a magnetic reactor, a busbar, a capacitor box, a station transformer, a bus, a resistor box, a low-voltage side pile head, a support insulator, a metering box, a reactor, an on-column vacuum switch, an intelligent control cabinet, a cable terminal, a coupling capacitor, an intelligent control cabinet, a power cable, an air switch, an intelligent component cabinet, A current transformer, a compensation device, an intelligent terminal cabinet, a capacitor, a fuse, a screen cabinet, an oil tank, a vertical insulator, an overhaul power supply box, an arc extinction device, a lead, an overhaul power line box, a voltage transformer, a radiator, a capacitor box, a connecting tube, a wire clamp, a control box, a resistor, an arc extinction coil, a convergence control cabinet, a drop-out fuse, a suspension insulator, a distribution box, a connecting bridge, an oil conservator, a terminal box, a flexible connection, a low-voltage resistance, a damping box, a circuit breaker, an inflation sleeve, a regulating cabinet, a discharge gap, a power box, a discharge coil, an on-column circuit breaker, an electric control box, a voltage divider, an on-column isolating switch, a component cabinet, a cable terminal tail tube, a pile head, a switch box, a high-voltage fuse, a wave damper, a wind control box, a high-voltage sleeve, a tube bus, a low-voltage switch box, an isolating switch, an alternating current filter, a low-voltage side switch box, a converter valve, a grounding transformer, an overhaul box and a cold control box, and are not limited herein.
Specifically, the thermal infrared imager may be used to collect information of the target power device to obtain an image in an infrared general data file storage format of the target power device, that is, an infrared thermograph, and then extract a temperature visual image from the image as a temperature visual data set, that is, a pre-training data set.
It should be noted that the infrared temperature sensor may reflect an infrared radiation energy distribution diagram of the target to be detected received by the detector onto a photosensitive element of the infrared detector, so as to obtain a file in an infrared general data file storage format, i.e., an infrared thermography, which is obtained by processing the file by the processor.
The infrared thermography image may include the following data formats, for example: file header, calibration data, temperature measurement parameters, temperature data, imaging parameters, analytical data, voice and user-defined data, file footer, and the like, without limitation.
The file header can be a file version, a resolution of an image and a photographing time of the image, the calibration data can be stored in a flash, the calibration data and script data are written in the calibration process, the temperature measurement parameters can be parameters which influence temperature measurement such as storage radiance, ambient temperature, relative humidity and distance, the imaging parameters can be imaging range and isothermal analysis data, and the file tail can be offset addresses (used for rapid positioning) of 4 data sections such as the file header, the temperature measurement parameters, the temperature data and the sound data, and data such as file identification, and the limitation is not carried out.
For example, in the corresponding imaging, the temperature data portion may be a float type two-dimensional matrix having the same length and width as the resolution of the infrared thermography, and the ambient temperature is imaged in a numerical form, that is, temperature vision. In some cases, the value of an element in the matrix may correspond to the actual temperature value of a region in the environment, for example, if the temperature value of the upper left corner of the matrix is 37.0, the actual temperature of the upper left corner region in the environment is 37.0 ℃.
And 102, inputting the temperature visual image into a target detection model generated by pre-training so as to acquire the type information and the position information of the target power equipment.
Before inputting the training data set into the target detection model, the training data set that is originally input needs to be normalized to reduce the influence on the result due to the large difference between the data.
The target detection model can be a target detection network such as R-CNN, Fast R-CNN, YOLO, SSD, RetinaNet, RefineDet and the like. It should be noted that, in step 102, the pre-training data set is labeled according to a specified manner to generate a training data set, where the specified manner at least includes device box labeling and feature point labeling.
Optionally, after the pre-training data set is obtained, the temperature visual data set is labeled according to a specified mode, where the specified mode at least includes device frame labeling and feature point labeling, and then the labeled temperature visual data set is used to train the initial detection model to obtain the target detection model.
The specified manner may be a predefined manner, such as device frame labeling, that is, labeling the target frame according to the region of interest, and in addition, labeling the feature points in the power device, which is not limited herein.
It should be noted that the initial detection model can be trained through the labeled temperature visual image to obtain the target detection model.
And 103, determining the target area image and the corresponding sub-classification model of the target electric power equipment according to the type information and the position information of the target electric power equipment.
As a possible implementation manner, each temperature visual image in the pre-training data set of the target power device may be first clipped to obtain a training data set including the labeled region, then the training data set is classified according to the type of the target power device to obtain a plurality of training data subsets, and then a plurality of corresponding initial sub-classification models are trained by using the plurality of training data subsets to obtain a plurality of trained sub-classification models.
The background part except the labeling frame in the temperature visual image can be removed through cutting, and only the interested area is left. The training data subsets corresponding to the power devices of the respective types can be obtained by classifying the training data sets, and are not limited herein.
The initial sub-classification model may be a network model to be trained currently, and the sub-classification model may be a network model after training.
Through the sub-classification model, the temperature visual image after cutting, namely the target area image, can be identified, so that the frames of a plurality of component positions in the target area image, namely the image frames, can be obtained.
Optionally, the temperature visual image of the target power device may be cut according to the position information of the target power device to obtain a target area image, and then the sub-classification model corresponding to the current target power device is determined based on a preset corresponding relationship between the type information of the target power device and the sub-classification model.
It should be noted that, in order to perform diagnosis on the electrical equipment more accurately, the present invention may detect the position of a component in each electrical equipment based on the temperature visual image, and then perform fault diagnosis on the electrical equipment. In order to more accurately acquire the positions of all parts in the power equipment, the detection of the parts of the power equipment is performed step by step.
For a plurality of kinds of electric power apparatuses, since there are differences in component composition structure, appearance shape, and the like of each kind of electric power apparatus, first, each electric power apparatus is classified into a plurality of subclasses. And then, cutting the labeling frame region of the temperature visual target frame labeling data set of the various power equipment by using a convolutional neural network, and then training the obtained temperature visual data set of the power equipment, namely the data with the background part removed, so that a plurality of classification models can be obtained to classify different appearance equipment under the same type of equipment.
It can be understood that, because the temperature visual data obtains the device position information and the type information through the target detection model, the temperature visual data of the area where the device is located is cut out according to the device position information, and then the corresponding classification model can be called according to the type information for classification.
Specifically, each of the power devices may be divided more finely according to the appearance, that is, divided into small categories, so as to obtain a plurality of small category models.
And 104, inputting the target area image into the sub-classification model to obtain an image to be detected containing each part frame in the marked target area image.
It should be noted that, for each small-classified power device, a sub-classification model is provided correspondingly, for example, based on convolutional neural networks such as LeNet, AlexNet, cafnenet, ZFNet, vggtet, ResNet, NiN, google LeNet, densneet, sennet, BAM, and MobileNet, the sub-classification model is modified into a feature point extraction network, and the network outputs feature point coordinates of each component of the power device and trains a temperature visual data set labeled with the feature points to obtain a feature point extraction model. The device can detect the image to be detected of the power equipment according to the sub-classification model, output the position of the characteristic point of the power equipment, and finally convert the characteristic point into a component frame for framing the positions of all the components of the power equipment and diagnosing the subsequent faults.
And 105, inputting the image to be detected into a fault detection model generated by pre-training so as to determine the fault type and the fault position of the target power equipment.
Optionally, two classification temperature data sets divided according to the faulty device and the non-faulty device may be obtained first, and then the initial fault diagnosis model is trained by using the two classification temperature data sets to generate the fault detection model.
Specifically, for a given power device, the power device may be divided into two classification temperature visual data sets according to a faulty device and a non-faulty device in advance, and then the two classification temperature visual data sets are trained by using initial fault diagnosis models such as LeNet, AlexNet, vggent, ResNet, NiN, google LeNet, densnet, sennet, BAM, MobileNet, and the like, to obtain a fault diagnosis model. The fault diagnosis model can detect the output temperature visual data and judge whether the data is faulty or not, so that intelligent diagnosis of temperature vision is realized.
Optionally, the server may further input the image to be detected into a fault detection model generated by training to obtain an annotation label corresponding to the image to be detected, where the annotation label includes a fault type label and a fault location label, then input the image to be detected into the initial fault diagnosis model to be trained to obtain a prediction label corresponding to the image to be detected, and finally correct the initial fault diagnosis model and the fault detection model respectively according to a difference between the prediction label and the annotation label.
Specifically, the device may compare the prediction tag with the labeling tag to determine a difference between the prediction tag and the labeling tag, for example, a correction gradient may be determined in a gradient descent manner, a random gradient descent manner, and the like, so as to correct the initial fault diagnosis model and the fault detection model generated by training, respectively.
Optionally, the parameter of the initial fault diagnosis model may be corrected by using a regression network through the cross loss of the fault detection model. Therefore, the fault detection model can output the fault corresponding to the current image to be detected more accurately and reliably.
The present invention can accurately identify and locate 106 types of electric power equipment based on the temperature visual image data, where the identified mAP for the 106 types of electric power equipment is 98%, the identified mAP for the 106 small categories of electric power equipment is 99%, the location IoU for the characteristic points of the 106 types of electric power equipment components is 95%, and the fault diagnosis acc for the 106 types of electric power equipment is 97%.
In the embodiment of the invention, a temperature visual image of target power equipment is firstly acquired, then the temperature visual image is input into a target detection model generated by pre-training to acquire type information of the target power equipment, then a target area image and a sub-classification model corresponding to the target power equipment are determined according to the type information and the position information of the target power equipment, then the target area image is input into the sub-classification model to acquire an image to be detected including each component frame marked in the target area image, and finally the image to be detected is input into a fault detection model generated by pre-training to determine the fault type and the fault position of the target power equipment. Therefore, the target detection model can be trained by using the temperature visual data to acquire the type and the position of the equipment in the temperature visual image, the component position of the power equipment in the image can be locked more accurately, and the fault of the power equipment can be diagnosed intelligently. Based on the temperature visual information, the position of the power equipment is positioned through a target detection network, and the positions of the components of the power equipment are positioned through a feature point network, so that the positions of the components of the power equipment are accurately positioned and corresponding temperature information is obtained.
In order to realize the embodiment, the invention further provides an intelligent detection device based on temperature vision.
Fig. 2 is a schematic structural diagram of an intelligent detection device based on temperature vision according to an embodiment of the present invention.
As shown in fig. 2, the intelligent detection device 200 based on temperature vision may include: a first obtaining module 210, a second obtaining module 220, a first determining module 230, a third obtaining module 240, and a second determining module 250.
The first acquisition module is used for acquiring a temperature visual image of the target power equipment;
the second acquisition module is used for inputting the temperature visual image into a target detection model generated by pre-training so as to acquire the type information of the target power equipment;
the first determining module is used for determining a target area image and a sub-classification model corresponding to the target electric power equipment according to the type information and the position information of the target electric power equipment;
a third obtaining module, configured to input the target area image into the sub-classification model, so as to obtain an image to be detected that includes each part frame marked in the target area image;
and the second determining module is used for inputting the image to be detected into a fault detection model generated by pre-training so as to determine the fault type and the fault position of the target power equipment.
Optionally, the second obtaining module is further configured to:
acquiring a pre-training data set of a target electric power device, wherein the pre-training data set is a temperature visual data set of the target electric power device;
labeling the temperature visual data set according to a specified mode, wherein the specified mode at least comprises equipment frame labeling and characteristic point labeling;
and training the initial detection model by using the marked temperature visual data set to obtain a target detection model.
Optionally, the first determining module is further configured to:
cutting each temperature visual image in a pre-training data set of target power equipment to obtain a training data set containing a labeled area;
classifying the training data set according to the type of the target power equipment to obtain a plurality of training data subsets;
and training a plurality of corresponding initial sub-classification models by using the plurality of training data subsets to obtain a plurality of trained sub-classification models.
Optionally, the first determining module is specifically configured to:
according to the position information of the target power equipment, cutting a temperature visual image of the target power equipment to obtain a target area image;
and determining a sub-classification model corresponding to the current target electric power equipment based on the preset corresponding relation between the type information of the target electric power equipment and the sub-classification model.
Optionally, the second determining module is further configured to:
acquiring two classification temperature data sets divided according to fault equipment and non-fault equipment;
and training an initial fault diagnosis model by using the two classified temperature data sets to generate a fault detection model.
Optionally, the apparatus further comprises:
the third acquisition module is used for inputting the image to be detected into a fault detection model generated by training so as to acquire a label corresponding to the image to be detected, wherein the label comprises a fault type label and a fault position label;
the fourth acquisition module is used for inputting the image to be detected into the initial fault diagnosis model to be trained so as to acquire a prediction label corresponding to the image to be detected;
and the correction module is used for respectively correcting the initial fault diagnosis model and the fault detection model according to the difference between the prediction label and the labeling label.
In the embodiment of the invention, a temperature visual image of target power equipment is firstly acquired, then the temperature visual image is input into a target detection model generated by pre-training to acquire type information of the target power equipment, then a target area image and a sub-classification model corresponding to the target power equipment are determined according to the type information and the position information of the target power equipment, then the target area image is input into the sub-classification model to acquire an image to be detected including each component frame marked in the target area image, and finally the image to be detected is input into a fault detection model generated by pre-training to determine the fault type and the fault position of the target power equipment. Therefore, the target detection model can be trained by using the temperature visual data to acquire the type and the position of the equipment in the temperature visual image, the component position of the power equipment in the image can be locked more accurately, and the fault of the power equipment can be diagnosed intelligently. Based on the temperature visual information, the position of the power equipment is positioned through a target detection network, and the positions of the components of the power equipment are positioned through a feature point network, so that the positions of the components of the power equipment are accurately positioned and corresponding temperature information is obtained.
In order to implement the above embodiments, the present invention further provides an electronic device, including: the intelligent detection method based on temperature vision is realized when the processor executes the program.
In order to achieve the above embodiments, the present invention further proposes a non-transitory computer readable storage medium storing a computer program, which when executed by a processor implements the intelligent detection method based on temperature vision as proposed by the foregoing embodiments of the present invention.
In order to implement the foregoing embodiments, the present invention further provides a computer program product, which when executed by an instruction processor in the computer program product, performs the intelligent detection method based on temperature vision as proposed in the foregoing embodiments of the present invention.
FIG. 3 illustrates a block diagram of an exemplary electronic device suitable for use in implementing embodiments of the present invention. The electronic device 12 shown in fig. 3 is only an example and should not bring any limitations to the function and scope of use of the embodiments of the present invention.
As shown in FIG. 3, electronic device 12 is embodied in the form of a general purpose computing device. The components of electronic device 12 may include, but are not limited to: one or more processors or processing units 16, a system memory 28, and a bus 18 that couples various system components including the system memory 28 and the processing unit 16.
A program/utility 40 having a set (at least one) of program modules 42 may be stored, for example, in memory 28, such program modules 42 including, but not limited to, an operating system, one or more application programs, other program modules, and program data, each of which examples or some combination thereof may comprise an implementation of a network environment. Program modules 42 generally carry out the functions and/or methodologies of the described embodiments of the invention.
The processing unit 16 executes various functional applications and data processing, for example, implementing the methods mentioned in the foregoing embodiments, by executing programs stored in the system memory 28.
In the embodiment of the invention, a temperature visual image of target power equipment is firstly acquired, then the temperature visual image is input into a target detection model generated by pre-training to acquire type information of the target power equipment, then a target area image and a sub-classification model corresponding to the target power equipment are determined according to the type information and the position information of the target power equipment, then the target area image is input into the sub-classification model to acquire an image to be detected including each component frame marked in the target area image, and finally the image to be detected is input into a fault detection model generated by pre-training to determine the fault type and the fault position of the target power equipment. Therefore, the target detection model can be trained by using the temperature visual data to acquire the type and the position of the equipment in the temperature visual image, the component position of the power equipment in the image can be locked more accurately, and the fault of the power equipment can be diagnosed intelligently. Based on the temperature visual information, the position of the power equipment is positioned through a target detection network, and the positions of the components of the power equipment are positioned through a feature point network, so that the positions of the components of the power equipment are accurately positioned and corresponding temperature information is obtained.
In the description herein, references to the description of the term "one embodiment," "some embodiments," "an example," "a specific example," or "some examples," etc., mean that a particular feature, structure, material, or characteristic described in connection with the embodiment or example is included in at least one embodiment or example of the invention. In this specification, the schematic representations of the terms used above are not necessarily intended to refer to the same embodiment or example. Furthermore, the particular features, structures, materials, or characteristics described may be combined in any suitable manner in any one or more embodiments or examples. Furthermore, various embodiments or examples and features of different embodiments or examples described in this specification can be combined and combined by one skilled in the art without contradiction.
Furthermore, the terms "first", "second" and "first" are used for descriptive purposes only and are not to be construed as indicating or implying relative importance or implicitly indicating the number of technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include at least one such feature. In the description of the present invention, "a plurality" means at least two, e.g., two, three, etc., unless specifically limited otherwise.
Any process or method descriptions in flow charts or otherwise described herein may be understood as representing modules, segments, or portions of code which include one or more executable instructions for implementing steps of a custom logic function or process, and alternate implementations are included within the scope of the preferred embodiment of the present invention in which functions may be executed out of order from that shown or discussed, including substantially concurrently or in reverse order, depending on the functionality involved, as would be understood by those reasonably skilled in the art of the present invention.
The logic and/or steps represented in the flowcharts or otherwise described herein, e.g., an ordered listing of executable instructions that can be considered to implement logical functions, can be embodied in any computer-readable medium for use by or in connection with an instruction execution system, apparatus, or device, such as a computer-based system, processor-containing system, or other system that can fetch the instructions from the instruction execution system, apparatus, or device and execute the instructions. For the purposes of this description, a "computer-readable medium" can be any means that can contain, store, communicate, propagate, or transport the program for use by or in connection with the instruction execution system, apparatus, or device. More specific examples (a non-exhaustive list) of the computer-readable medium would include the following: an electrical connection (electronic device) having one or more wires, a portable computer diskette (magnetic device), a Random Access Memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber device, and a portable compact disc read-only memory (CDROM). Additionally, the computer-readable medium could even be paper or another suitable medium upon which the program is printed, as the program can be electronically captured, via for instance optical scanning of the paper or other medium, then compiled, interpreted or otherwise processed in a suitable manner if necessary, and then stored in a computer memory.
It should be understood that portions of the present invention may be implemented in hardware, software, firmware, or a combination thereof. In the above embodiments, the various steps or methods may be implemented in software or firmware stored in memory and executed by a suitable instruction execution system. If implemented in hardware, as in another embodiment, any one or combination of the following techniques, which are known in the art, may be used: a discrete logic circuit having a logic gate circuit for implementing a logic function on a data signal, an application specific integrated circuit having an appropriate combinational logic gate circuit, a Programmable Gate Array (PGA), a Field Programmable Gate Array (FPGA), or the like.
It will be understood by those skilled in the art that all or part of the steps carried by the method for implementing the above embodiments may be implemented by hardware related to instructions of a program, which may be stored in a computer readable storage medium, and when the program is executed, the program includes one or a combination of the steps of the method embodiments.
In addition, functional units in the embodiments of the present invention may be integrated into one processing module, or each unit may exist alone physically, or two or more units are integrated into one module. The integrated module can be realized in a hardware mode, and can also be realized in a software functional module mode. The integrated module, if implemented in the form of a software functional module and sold or used as a stand-alone product, may also be stored in a computer readable storage medium.
The storage medium mentioned above may be a read-only memory, a magnetic or optical disk, etc. Although embodiments of the present invention have been shown and described above, it is understood that the above embodiments are exemplary and should not be construed as limiting the present invention, and that variations, modifications, substitutions and alterations can be made to the above embodiments by those of ordinary skill in the art within the scope of the present invention.
Claims (10)
1. An intelligent detection method based on temperature vision is characterized by comprising the following steps:
acquiring a temperature visual image of target power equipment;
inputting the temperature visual image into a target detection model generated by pre-training so as to acquire type information of the target power equipment;
determining a target area image and a sub-classification model corresponding to the target electric power equipment according to the type information and the position information of the target electric power equipment;
inputting the target area image into the sub-classification model to obtain an image to be detected containing each part frame marked in the target area image;
and inputting the image to be detected into a fault detection model generated by pre-training so as to determine the fault type and the fault position of the target power equipment.
2. The method according to claim 1, further comprising, before the inputting the training data set into a target detection model generated by pre-training to obtain type information and location information of a target power device:
acquiring a pre-training data set of a target electric power device, wherein the pre-training data set is a temperature visual data set of the target electric power device;
labeling the temperature visual data set according to a specified mode, wherein the specified mode at least comprises equipment frame labeling and characteristic point labeling;
and training the initial detection model by using the marked temperature visual data set to obtain a target detection model.
3. The method according to claim 1, before determining a target area image and a corresponding sub-classification model of the target electrical device according to the type information and the location information of the target electrical device, further comprising:
cutting each temperature visual image in a pre-training data set of target power equipment to obtain a training data set containing a labeled area;
classifying the training data set according to the type of the target power equipment to obtain a plurality of training data subsets;
and training a plurality of corresponding initial sub-classification models by using the plurality of training data subsets to obtain a plurality of trained sub-classification models.
4. The method according to claim 1, wherein the determining a target area image and a corresponding sub-classification model of the target electrical device according to the type information and the location information of the target electrical device comprises:
according to the position information of the target power equipment, cutting a temperature visual image of the target power equipment to obtain a target area image;
and determining a sub-classification model corresponding to the current target electric power equipment based on the preset corresponding relation between the type information of the target electric power equipment and the sub-classification model.
5. The method according to claim 1, wherein before inputting the image to be detected to a fault detection model generated by pre-training, the method further comprises:
acquiring two classification temperature data sets divided according to fault equipment and non-fault equipment;
and training an initial fault diagnosis model by using the two classified temperature data sets to generate a fault detection model.
6. The method of claim 1, further comprising:
inputting the image to be detected into a fault detection model generated by training to obtain a label corresponding to the image to be detected, wherein the label comprises a fault type label and a fault position label;
inputting the image to be detected into an initial fault diagnosis model to be trained so as to obtain a prediction label corresponding to the image to be detected;
and respectively correcting the initial fault diagnosis model and the fault detection model according to the difference between the prediction label and the labeling label.
7. An intelligent detection device based on temperature vision, comprising:
the first acquisition module is used for acquiring a temperature visual image of the target power equipment;
the second acquisition module is used for inputting the temperature visual image into a target detection model generated by pre-training so as to acquire the type information of the target power equipment;
the first determining module is used for determining a target area image and a sub-classification model corresponding to the target electric power equipment according to the type information and the position information of the target electric power equipment;
a third obtaining module, configured to input the target area image into the sub-classification model, so as to obtain an image to be detected that includes each part frame marked in the target area image;
and the second determining module is used for inputting the image to be detected into a fault detection model generated by pre-training so as to determine the fault type and the fault position of the target power equipment.
8. The apparatus of claim 7, wherein the second obtaining module is further configured to:
acquiring a pre-training data set of a target electric power device, wherein the pre-training data set is a temperature visual data set of the target electric power device;
labeling the temperature visual data set according to a specified mode, wherein the specified mode at least comprises equipment frame labeling and characteristic point labeling;
and training the initial detection model by using the marked temperature visual data set to obtain a target detection model.
9. The apparatus of claim 7, wherein the first determining module is further configured to:
cutting each temperature visual image in a pre-training data set of target power equipment to obtain a training data set containing a labeled area;
classifying the training data set according to the type of the target power equipment to obtain a plurality of training data subsets;
and training a plurality of corresponding initial sub-classification models by using the plurality of training data subsets to obtain a plurality of trained sub-classification models.
10. The apparatus of claim 7, wherein the first determining module is specifically configured to:
according to the position information of the target power equipment, cutting a temperature visual image of the target power equipment to obtain a target area image;
and determining a sub-classification model corresponding to the current target electric power equipment based on the preset corresponding relation between the type information of the target electric power equipment and the sub-classification model.
Priority Applications (2)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202111413561.0A CN114067212A (en) | 2021-11-25 | 2021-11-25 | Intelligent detection method and device based on temperature vision and electronic equipment |
PCT/CN2021/139367 WO2023092759A1 (en) | 2021-11-25 | 2021-12-17 | Temperature vision-based intelligent detection method and apparatus, and electronic device |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202111413561.0A CN114067212A (en) | 2021-11-25 | 2021-11-25 | Intelligent detection method and device based on temperature vision and electronic equipment |
Publications (1)
Publication Number | Publication Date |
---|---|
CN114067212A true CN114067212A (en) | 2022-02-18 |
Family
ID=80276302
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202111413561.0A Pending CN114067212A (en) | 2021-11-25 | 2021-11-25 | Intelligent detection method and device based on temperature vision and electronic equipment |
Country Status (2)
Country | Link |
---|---|
CN (1) | CN114067212A (en) |
WO (1) | WO2023092759A1 (en) |
Cited By (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN114511120A (en) * | 2022-04-21 | 2022-05-17 | 浙江天铂云科光电股份有限公司 | Power equipment fault diagnosis method based on temperature vision electric wire |
CN115115590A (en) * | 2022-06-23 | 2022-09-27 | 华南理工大学 | Composite insulator overheating defect detection method based on rotary RetinaNet |
CN115661054A (en) * | 2022-10-14 | 2023-01-31 | 蓝思系统集成有限公司 | Method and device for detecting seal, electronic device and storage medium |
Families Citing this family (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN116704266B (en) * | 2023-07-28 | 2023-10-31 | 国网浙江省电力有限公司信息通信分公司 | Power equipment fault detection method, device, equipment and storage medium |
CN117129088A (en) * | 2023-08-08 | 2023-11-28 | 上海根根信息技术有限公司 | Chip temperature testing method and system |
CN117409261B (en) * | 2023-12-14 | 2024-02-20 | 成都数之联科技股份有限公司 | Element angle classification method and system based on classification model |
CN117629422A (en) * | 2024-01-25 | 2024-03-01 | 徐州日托新能源科技有限公司 | Temperature measurement method and system for photovoltaic module |
Family Cites Families (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US9858525B2 (en) * | 2015-10-14 | 2018-01-02 | Microsoft Technology Licensing, Llc | System for training networks for semantic segmentation |
CN107563412A (en) * | 2017-08-09 | 2018-01-09 | 浙江大学 | A kind of infrared image power equipment real-time detection method based on deep learning |
CN109101906A (en) * | 2018-07-27 | 2018-12-28 | 中国南方电网有限责任公司超高压输电公司贵阳局 | A kind of converting station electric power equipment infrared image exception real-time detection method and device |
CN110598736B (en) * | 2019-08-06 | 2022-12-20 | 西安理工大学 | Power equipment infrared image fault positioning, identifying and predicting method |
CN111179249A (en) * | 2019-12-30 | 2020-05-19 | 南京南瑞信息通信科技有限公司 | Power equipment detection method and device based on deep convolutional neural network |
CN111986172A (en) * | 2020-08-18 | 2020-11-24 | 华北电力科学研究院有限责任公司 | Infrared image fault detection method and device for power equipment |
CN112233073A (en) * | 2020-09-30 | 2021-01-15 | 国网山西省电力公司大同供电公司 | Real-time detection method for infrared thermal imaging abnormity of power transformation equipment |
-
2021
- 2021-11-25 CN CN202111413561.0A patent/CN114067212A/en active Pending
- 2021-12-17 WO PCT/CN2021/139367 patent/WO2023092759A1/en unknown
Cited By (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN114511120A (en) * | 2022-04-21 | 2022-05-17 | 浙江天铂云科光电股份有限公司 | Power equipment fault diagnosis method based on temperature vision electric wire |
CN114511120B (en) * | 2022-04-21 | 2022-07-19 | 浙江天铂云科光电股份有限公司 | Power equipment fault diagnosis method based on temperature vision electric wire |
CN115115590A (en) * | 2022-06-23 | 2022-09-27 | 华南理工大学 | Composite insulator overheating defect detection method based on rotary RetinaNet |
CN115115590B (en) * | 2022-06-23 | 2024-03-08 | 华南理工大学 | Composite insulator overheat defect detection method based on rotary RetinaNet |
CN115661054A (en) * | 2022-10-14 | 2023-01-31 | 蓝思系统集成有限公司 | Method and device for detecting seal, electronic device and storage medium |
Also Published As
Publication number | Publication date |
---|---|
WO2023092759A1 (en) | 2023-06-01 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN114067212A (en) | Intelligent detection method and device based on temperature vision and electronic equipment | |
CN107907775B (en) | Wiring detection method and wiring detection device | |
WO2023092760A1 (en) | Diagnosis method and diagnosis apparatus based on thermal vision | |
CN113590482B (en) | Test method and test system for battery management system BMS | |
CN201653450U (en) | Digital display meter automatic calibration device based on machine vision | |
CN116109635B (en) | Method, device, equipment and medium for detecting surface quality of composite suspension insulator | |
CN113343998A (en) | Reading monitoring system and method for electric power mechanical meter, computer equipment and application | |
CN109061374A (en) | Utilize the Zinc-Oxide Arrester monitoring method and device of electric substation automation system | |
CN111649865A (en) | Wind power and wind direction detection method based on image processing technology | |
Li | Design of infrared anomaly detection for power equipment based on YOLOv3 | |
CN113077447A (en) | Method capable of detecting weak discharge of high-voltage equipment and analyzing defects | |
CN114065642A (en) | Intelligent identification method based on temperature vision | |
CN111044830A (en) | Zinc oxide arrester operating characteristic live monitoring device | |
CN116338392A (en) | Method, device and equipment for identifying lightning discharge defects of glass insulator | |
CN114419037B (en) | Workpiece defect detection method and device | |
CN114494186B (en) | Fault detection method for high-voltage power transmission and transformation line electrical equipment | |
CN107543574B (en) | Automatic detector for high-temperature aging test of airborne sensor and operation method | |
CN114152290B (en) | Method for diagnosing alternating current filter switch of convertor station | |
CN111524108B (en) | Transformer substation equipment detection method and equipment | |
CN111931721A (en) | Method and device for detecting color and number of annual inspection label and electronic equipment | |
KR101743595B1 (en) | Parital discharge diagnosis method and system, and mold transfomer deterioration monitoring system using the same. | |
CN110716099A (en) | Detection system and method for multi-needle glass sealed photoelectron tube | |
CN113139955B (en) | Photovoltaic module fault identification method and system based on double-light image | |
CN117471390A (en) | Calibrating device and method for direct-current ground fault finder | |
CN113449633B (en) | Pointer type meter counting value identification method based on object detection and key point positioning |
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 |