CN113378818B

CN113378818B - Electrical equipment defect determining method and device, electronic equipment and storage medium

Info

Publication number: CN113378818B
Application number: CN202110688346.5A
Authority: CN
Inventors: 全晓方; 郑奇凯; 孙上元; 陈何成; 李顺; 姚日斌; 黄繁朝; 刘彬; 杨海亮; 杨武志
Original assignee: Liuzhou Bureau of Extra High Voltage Power Transmission Co
Current assignee: Liuzhou Bureau of Extra High Voltage Power Transmission Co
Priority date: 2021-06-21
Filing date: 2021-06-21
Publication date: 2024-06-07
Anticipated expiration: 2041-06-21
Also published as: CN113378818A

Abstract

The embodiment of the application relates to the technical field of equipment detection, and discloses a method and a device for determining defects of electrical equipment, electronic equipment and a storage medium, wherein the method comprises the following steps: acquiring a first infrared image of the electrical equipment to be identified; acquiring temperature information of electrical equipment to be identified according to the first infrared image; determining a temperature anomaly region in the first infrared image according to the temperature information; extracting the characteristics of the temperature anomaly region through the trained defect determination model, generating a plurality of first masks corresponding to the temperature anomaly region according to the characteristics, and determining first mask coefficients of defective components in the temperature anomaly region; generating a second mask of the temperature anomaly region according to the plurality of first masks and the first mask coefficients through the defect determination model; and determining the image position of the defective part of the electrical equipment to be identified in the first infrared image according to the second mask through the defect determination model. The defect part can be accurately positioned, and the identification accuracy of the defect part of the electrical equipment can be improved.

Description

Electrical equipment defect determining method and device, electronic equipment and storage medium

Technical Field

The present application relates to the field of device detection technologies, and in particular, to a method and apparatus for determining defects of an electrical device, an electronic device, and a storage medium.

Background

The electric equipment in the electric power system runs in a high-voltage environment for a long time, and the electric equipment inevitably has defects of abnormal temperature, structural damage and the like under the actions of long-term voltage, heating, mechanical stress and environmental factors, so that the timely discovery and elimination of the defects of the electric equipment are beneficial to improving the safety of the electric equipment and the electric power system.

At present, an infrared image is generally applied to an electric power system to judge whether the temperature of electric equipment is abnormal, but the infrared image can only reflect the abnormal temperature area of the electric equipment, and the identification accuracy of defective parts which cause the abnormal temperature in the electric equipment is low.

Disclosure of Invention

The embodiment of the application discloses a method and a device for determining defects of electrical equipment, electronic equipment and a storage medium, which can accurately identify defective parts of the electrical equipment and improve the identification accuracy of the defective parts in the electrical equipment.

The first aspect of the embodiment of the application discloses a method for determining defects of electrical equipment, which comprises the following steps:

Acquiring a first infrared image of the electrical equipment to be identified;

acquiring temperature information of the electrical equipment to be identified according to the first infrared image;

Determining a temperature anomaly region in the first infrared image according to the temperature information;

extracting the characteristics of the temperature anomaly region through a trained defect determination model, generating a plurality of first masks corresponding to the temperature anomaly region according to the characteristics, and determining first mask coefficients of defective components in the temperature anomaly region;

Generating a second mask of the temperature anomaly region according to the plurality of first masks and the first mask coefficients through the defect determination model;

and determining the image position of the defective part of the electrical equipment to be identified in the first infrared image according to the second mask through the defect determination model.

In an optional implementation manner, in the first aspect of the embodiment of the present application, the defect determining model is obtained by training a training image set, where the training image set includes an infrared image corresponding to when each component part in the electrical device is used as a defective part, and the infrared image is labeled with information when each component part in the electrical device is used as a defective part.

As an optional implementation manner, in the first aspect of the embodiment of the present application, after the determining, by the defect determining model, an image position of the defect part of the electrical device to be identified in the first infrared image according to the second mask, the method further includes:

intercepting a first infrared sub-image corresponding to the defect part from the first infrared image according to the image position;

acquiring temperature information of the defective part according to the first infrared sub-image;

Respectively comparing the temperature information of the first infrared sub-image and the temperature information of the defect part with the second infrared image of each component part of the electrical equipment to be identified and the temperature information of each component part in a database, wherein the second infrared image is an infrared image of each component part of the electrical equipment to be identified in a normal state;

and determining the component information of the defective component according to the comparison result.

As an alternative implementation, in the first aspect of the embodiment of the present application,

The defect determination model comprises a depth residual network with a plurality of convolution modules, and the extracting the characteristics of the temperature anomaly region through the trained defect determination model comprises the following steps:

Processing the temperature abnormal region through a plurality of convolution modules of the depth residual error network to obtain a plurality of first feature images with different output sizes;

processing the first feature map with the smallest output size through a convolution layer to obtain a second feature map;

Performing convolution and downsampling operations on the second feature map to obtain a deep network feature map;

Performing convolution processing on the first feature map with the second small output size to obtain a third feature map;

And amplifying the second characteristic diagram, and summing the amplified second characteristic diagram and the third characteristic diagram to obtain a shallow network characteristic diagram.

The defect determination model further comprises a mask generation network and a coefficient generation network, wherein the mask generation network and the coefficient generation network are parallel networks;

the generating a plurality of first masks corresponding to the temperature anomaly region according to the characteristics, and determining first mask coefficients of defective components in the temperature anomaly region, including:

Inputting the shallow network feature map to the mask generation network, and generating a plurality of first masks corresponding to the temperature abnormal region based on the shallow network feature map through the mask generation network;

And the second characteristic diagram, the deep network characteristic diagram and the shallow network characteristic diagram are input to the coefficient generation network together, and the coefficient generation network determines a first mask coefficient of the defect component in the temperature anomaly region based on the second characteristic diagram, the deep network characteristic diagram and the shallow network characteristic diagram.

As an optional implementation manner, in the first aspect of the embodiment of the present application, the determining, by the defect determining model, an image position of the defect part of the electrical device to be identified in the first infrared image according to the second mask includes:

Dividing the second mask of the temperature anomaly region through the defect determination model to obtain a third mask corresponding to each defective component in the temperature anomaly region;

And performing image binarization processing on the third mask through the defect determination model, and determining the image position of the defect part in the first infrared image according to the binarized third mask.

The determining a first mask coefficient of the defective component in the temperature anomaly region includes:

In the defect determination model, generating a plurality of prediction frames of the defect part in the temperature anomaly region according to the characteristics, and determining classification confidence coefficients corresponding to the prediction frames one by one and second mask coefficients corresponding to the prediction frames one by one;

And screening the plurality of prediction frames according to the classification confidence degrees corresponding to the prediction frames, determining a target prediction frame uniquely corresponding to the defect part, and determining a second mask coefficient corresponding to the target prediction frame as a first mask coefficient.

In this embodiment, an image corresponding to a temperature anomaly region in a first infrared image is determined according to temperature information acquired by a first infrared image of an electrical device to be identified, after features of the corresponding image are extracted by a trained defect determination model, a plurality of first masks with the same size as the corresponding image are generated according to the features, first mask coefficients of defective parts in the temperature anomaly region are determined, a second mask capable of indicating the defective parts in the temperature anomaly region is generated according to the plurality of first masks and the first mask coefficients, and the image position of the defective parts in the first infrared image can be accurately determined according to the second mask, so that the defective parts can be accurately positioned, and the identification accuracy of defective parts in the electrical device can be effectively improved.

A second aspect of an embodiment of the present application discloses an electrical equipment defect determining apparatus, including:

the image acquisition module is used for acquiring a first infrared image of the electrical equipment to be identified, which is acquired by the infrared camera;

the temperature acquisition module is used for acquiring temperature information of the electrical equipment to be identified according to the first infrared image;

the region intercepting module is used for determining a temperature abnormal region in the infrared image according to the temperature information;

The mask acquisition module is used for extracting the characteristics of the temperature abnormal region through the trained defect determination model, generating a plurality of first masks corresponding to the temperature abnormal region according to the characteristics, and determining first mask coefficients of defective components in the temperature abnormal region;

A mask determining module, configured to generate, according to the plurality of first masks and the first mask coefficients, a second mask of the temperature anomaly region through the defect determining model;

And the defect determining module is used for determining the image position of the defect part of the electrical equipment to be identified in the first infrared image according to the second mask through the defect determining model.

A third aspect of an embodiment of the present application discloses an electronic device, including: a memory and a processor, the memory storing a computer program executable by the processor to cause the processor to implement the electrical device defect determination method according to the first aspect of the embodiment of the present application.

A fourth aspect of the embodiment of the present application discloses a computer readable storage medium, in which a computer program is stored, the computer program being adapted to be loaded and executed by a processor, so that there is provided the processor to implement a method for determining defects of an electrical device as disclosed in the first aspect of the embodiment of the present application.

Drawings

In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings that are needed in the embodiments will be briefly described below, and it is obvious that the drawings in the following description are only some embodiments of the present application, and other drawings may be obtained according to these drawings without inventive effort for a person skilled in the art.

FIG. 1 is a schematic flow chart of a method for determining defects of an electrical device according to an embodiment of the present application;

FIG. 2 is a schematic diagram of a defect determination model Yolact according to an example of the present application;

FIG. 3 is a schematic diagram of a temperature anomaly region obtained by different region determination methods according to embodiments of the present application;

FIG. 4 is a schematic flow chart of generating a plurality of first masks corresponding to a temperature anomaly region and determining first mask coefficients of a defective component in the temperature anomaly region according to an embodiment of the present application;

FIG. 5 is a schematic diagram of a different size feature map acquisition mode disclosed in an embodiment of the present application;

FIG. 6 is a flow chart of yet another method for determining defects in electrical equipment according to an embodiment of the present application;

FIG. 7 is a block diagram of an electrical equipment defect determining apparatus according to an embodiment of the present application;

FIG. 8 is a block diagram of another electrical device defect determining apparatus disclosed in an embodiment of the present application;

fig. 9 is a schematic structural diagram of an electronic device according to an embodiment of the present application.

Detailed Description

The following description of the embodiments of the present application will be made clearly and completely with reference to the accompanying drawings, in which it is apparent that the embodiments described are only some embodiments of the present application, but not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the application without making any inventive effort, are intended to be within the scope of the application.

It should be noted that the terms "first," "second," "third," and "fourth," etc. in the description and claims of the present application are used for distinguishing between different objects and not for describing a particular sequential order. The terms "comprises," "comprising," and "having," and any variations thereof, are intended to cover a non-exclusive inclusion, such that a process, method, system, article, or apparatus that comprises a list of steps or elements is not necessarily limited to those steps or elements expressly listed or inherent to such process, method, article, or apparatus.

At present, an infrared image is generally applied to an electric power system to judge whether the temperature of the electric equipment is abnormal, but the infrared image only reflects the abnormal temperature region of the electric equipment, and the infrared image is only subjected to characteristic extraction or characteristic comparison operation after being combined with the existing image recognition or image comparison mode due to lower resolution of the infrared image, so that the position of a defective part in the image cannot be determined, and the recognition accuracy of the defective part which causes the abnormal temperature in the electric equipment is lower.

In order to solve the technical problems, the embodiment of the application discloses a method, a device, electronic equipment and a storage medium for determining defects of electrical equipment, which can accurately identify the positions of defect parts of the electrical equipment in infrared images and improve the accuracy of identifying the defects of the electrical equipment. The following detailed description refers to the accompanying drawings.

Referring to fig. 1, fig. 1 is a flowchart of a method for determining defects of an electrical device according to an embodiment of the application. As shown in fig. 1, the electrical device defect determination method may include the following steps.

101. A first infrared image of an electrical device to be identified is acquired.

In the embodiment of the application, the electrical equipment is a generic term of equipment used for ensuring normal operation and transportation of electric power in an electric power system, and the electrical equipment can comprise equipment such as a generator, a transformer, an electric power circuit, a circuit breaker and the like. The infrared image is an image formed by acquiring the intensity of infrared light of an object, is obtained by measuring the heat radiated outwards by the object, and is an image formed by acquiring radiation of a target in an infrared band through an infrared imaging device.

In the embodiment of the application, the first infrared image is an infrared image of an electrical device to be identified, which is acquired through an infrared imaging device, wherein the electrical device to be identified is an electrical device to be identified, which has one or more defective parts, and the infrared imaging device may include an infrared camera, and the like.

In the embodiment of the application, the method for determining the defects of the electrical equipment is suitable for the electronic equipment such as terminal equipment or a server, wherein an operating system of the electronic equipment can include, but is not limited to, an Android operating system, an IOS operating system, a Symbian operating system, a Black Berry operating system, a Windows Phone8 operating system and the like.

In an embodiment of the present application, an electronic device may be provided with a User Interface (UI), an Interface module, and a processor (central processing unit, CPU).

In the embodiment of the application, after the infrared camera imaging equipment collects the first infrared image of the electrical equipment to be identified, the first infrared image is transmitted to the electronic equipment. The transmission mode may be wire transmission or wireless network transmission. The electronic device may support network technologies including, but not limited to: global system for mobile communications (Global System for Mobile Communications, GSM), general Packet Radio Service (GPRS), code Division multiple access (Code Division Multiple Access, CDMA), wideband code Division multiple access (W-CDMA), CDMA2000, IMT single carrier (IMT SINGLE CARRIER), enhanced data rates for GSM Evolution (ENHANCED DATA RATES for GSM Evolution, EDGE), long Term Evolution (Long Term Evolution, LTE), long Term Evolution-advanced (LTE), time-Division Long Term Evolution (Time-Division LTE), high performance Radio local area network (High Performance Radio Local Area Network, hiperLAN), high performance Radio wide area network (HiperWAN), local multipoint distribution Service (Local Multipoint Distribution Service, LMDS), worldwide Interoperability for Microwave Access (WiMAX), zigBee, bluetooth, orthogonal Frequency Division Multiplexing (OFDM), high capacity space Division multiple access (HC-SDMA), universal Mobile Telecommunications System (UMTS), universal mobile telecommunications system Time Division duplex (UMTS-UMTS), evolved high speed packet access (hspa+), time Division synchronous code Division multiple access (TD-SCDMA), evolved data optimization (TDD-DO), digital enhanced telecommunications (DECT), and others.

102. And acquiring temperature information of the electrical equipment to be identified according to the first infrared image.

In some embodiments, the electronic device obtains temperature information of the electrical device to be identified according to the first infrared image, specifically, the first infrared image may be converted into a gray image, and according to a linear relationship between a gray value and a temperature value, the gray value of each pixel may be converted into a corresponding temperature value. The electronic device may further convert each pixel point in the first infrared image into a corresponding temperature value according to a relationship between the RGB value and the temperature value. The present application is not particularly limited in the manner of acquiring temperature information from an infrared image.

103. And determining a temperature abnormal region in the first infrared image according to the temperature information.

In the embodiment of the application, the electronic device can compare the temperature value corresponding to each pixel point in the first infrared image with a preset normal temperature range, and determine one or more image areas formed by the pixel points with the temperature values exceeding the normal temperature range as the temperature abnormal areas.

104. And extracting the characteristics of the temperature anomaly region through the trained defect determination model, generating a plurality of first masks corresponding to the temperature anomaly region according to the characteristics, and determining the first mask coefficients of the defective components in the temperature anomaly region.

In an embodiment of the present application, the trained defect determination model may be a trained example segmentation model, which may include, but is not limited to, an R-CNN (Region with CNN feature) model, a Fast R-CNN model, a Mask R-CNN model, or a Yolact model as shown in FIG. 2, etc.

In some embodiments, the defect determination model is trained by a training image set that includes infrared images of the electrical device when each component is a defective component, and the infrared images of the electrical device when each component is a defective component are labeled with information of the electrical device when each component is a defective component. The electronic device may use the infrared images of the components of the plurality of different electrical devices as the defective components, for example, the infrared images of the electrical devices when the valve of the electrical device is the defective component, to train the defect determination model as a training image set, and the infrared images of the components of the plurality of different electrical devices as the defective components are labeled with information when the components of the plurality of different electrical devices are the defective components, where the information may include a type of defect of the defective component, for example, when the valve of the electrical device is the defective component, the type of defect is labeled as damaged valve or the type of defect is labeled as not closed valve, and so on. The object which can be identified by the defect training model is only the defect part corresponding to each component part in the electrical equipment, and the defect type of the defect part can be determined according to the mark after the defect part is identified. The defect component in the electrical equipment to be identified can be better identified and segmented, the problems of the defect component can be effectively displayed, the accuracy of defect determination is improved, and meanwhile, the efficiency of the defect component repair process is improved.

In some embodiments, the electronic device may treat a corresponding infrared image region of the determined temperature anomaly region in the first infrared image as the second infrared sub-image. For feature extraction of the temperature anomaly region, the second infrared sub-image may be specifically input into a trained defect determination model, and features of the second infrared sub-image, that is, features of the temperature anomaly region, may be extracted by a convolution module included in the defect determination model. The features extracted from the temperature anomaly region can be in the form of a multi-scale feature map. The network for extracting the characteristics in the defect determination model is provided with a plurality of convolution modules, each convolution module comprises a plurality of convolution layers, the convolution kernels of each convolution layer and the characteristic diagrams output by each convolution module are different in size, and detection of objects with different sizes can be realized.

For example, when the trained defect determination model is the trained Yolact example segmentation model, the Yolact example segmentation model uses a depth residual network to extract the multi-scale feature map of the second infrared sub-image to realize feature extraction of the temperature anomaly region. For example, the depth residual network may include 5 convolution modules, each of which includes 3 convolution layers, and the convolution modules are ordered according to the size of the feature map output by the convolution module from large to small, and each of the convolution modules may be named conv1, conv2, conv3, conv4, and conv5. The feature map sizes of the temperature anomaly region sub-images outputted by the 5 convolution modules are different from each other, and may be 112×112, 56×56, 28×28, 14×14, and 7×7, respectively. The number of the convolution layers in each convolution module and the size of the convolution kernel of each convolution layer are not particularly limited in the present application.

In some embodiments, the trained defect determination model includes a mask generation network and a coefficient generation network. After the feature extraction is carried out on the temperature abnormal region through the defect determination model, the extracted features are respectively input into a mask generation network and a coefficient generation network of the defect determination model, and the mask generation network outputs a plurality of first masks with the same size as the second infrared sub-image according to the input features of the temperature abnormal region; the coefficient generation network outputs a series of first mask coefficients predicted for defective components in the temperature anomaly region based on the input characteristics of the temperature anomaly region.

In the embodiment of the application, the first mask is a mask containing each component part of the temperature anomaly region, and the first mask can be used for sharing each component part of the temperature anomaly region contained in the mask, wherein the combination modes of each component part in each first mask are different. For the plurality of first masks, specifically, by inputting the characteristics of the temperature anomaly region into a mask generating network composed of a plurality of convolution layers, one first mask set, that is, a plurality of first masks, is output, wherein the set includes a plurality of first masks with the same size, for example, the output first mask set is denoted as 138×138×32, then 138×138 is denoted as the size of each first mask, and 32 is denoted as the number of first masks. Wherein the mask generation network may be selected as an FCN network.

The first mask coefficient refers to a coefficient of an area where the features of the temperature anomaly area are located, and is used to determine a first mask required for the area where each feature constituting the temperature anomaly area is located. In the embodiment of the application, for the first mask coefficient, the characteristic of the temperature abnormal region is input into the coefficient generation network, and the coefficient generation network comprises a prediction layer. In general, the temperature anomaly region is characterized by a feature map format, and the coefficient generation network has the same number of prediction layers as the feature map, and parameters are shared between the prediction layers.

In some embodiments, in the coefficient generation network, a prediction frame (anchor) is predicted for the feature of the input temperature anomaly region, and a plurality of prediction frames are generated for each pixel point according to different proportions, that is, a range of different shapes is generated for each feature in the sub-image of the temperature anomaly region, for example, 3 prediction frames are generated for each feature. The basic side length of the prediction frame of each feature may be set, for example, 5 features, and then the set basic side lengths of the prediction frames may be 24, 48, 96, 192 and 384, respectively. After generating the prediction frames, the coefficient generation network predicts a first set of mask coefficients for each generated prediction frame, respectively, where the first set of mask coefficients may include a confidence level of each prediction frame corresponding to a different defective component, location information of each prediction frame, and a first mask coefficient of each prediction frame. The number of first mask coefficients for each prediction frame is the same as the number of first masks. The first mask coefficient refers to a coefficient of each prediction frame for determining a first mask required to construct each prediction frame.

105. And generating a second mask of the temperature abnormal region according to the plurality of first masks and the first mask coefficients through the defect determination model.

In the embodiment of the application, the electronic device generates the second mask corresponding to the second infrared sub-image of the temperature anomaly region through the trained defect determination model according to the obtained plurality of first masks and the first mask coefficients, and specifically, the second mask corresponding to the temperature anomaly region can be obtained by linearly combining the first mask coefficients of the first masks respectively corresponding to the features with the corresponding first masks. Wherein the second mask is used to determine defective components in the temperature anomaly region. That is, the first mask is a mask of different combinations of the respective constituent members of the temperature anomaly region. For example, a first mask set with a dimension of 138×138×32 generated by the mask generating network, that is, 32 first masks with temperature anomaly regions of 138×138 are generated, where each first mask includes components of the temperature anomaly region that are different from each other. The first mask coefficients generated by the coefficient generation network are mask coefficients of prediction frames of the defective component in the temperature anomaly region, such as 3×32, that is, 3 prediction frames of the defective component and mask coefficients of each prediction frame relative to 32 first masks. At this time, the linear combination of the plurality of first masks and the first mask coefficients may be matrix multiplication, specifically m=pc ^T, where P is a first mask set, that is, the plurality of first masks generated by the mask generating network, C is a first mask coefficient of each prediction frame, and p=138×138×32, and c=3×32.

106. And determining the image position of the defective part of the electrical equipment to be identified in the first infrared image according to the second mask through the defect determination model.

In some embodiments, the electronic device may perform pixel filtering on the first infrared image according to the obtained second mask including the prediction frame of the defective component in the temperature anomaly area through the trained defect determination model, and display one or more defective components in the temperature anomaly area corresponding to the prediction frame in the second mask according to the position of the prediction frame in the first infrared image, so as to determine the position of the one or more defective components in the electrical device to be identified.

In the embodiment of the application, the image corresponding to the temperature abnormal region in the first infrared image is determined through the temperature information acquired by the first infrared image of the electrical equipment to be identified, after the characteristics of the corresponding image are extracted through the trained defect determination model, a plurality of first masks with the same size as the corresponding image are generated according to the characteristics, the first mask coefficient of the defect part in the temperature abnormal region is determined, the second mask capable of indicating the defect part in the temperature abnormal region is generated according to the plurality of first masks and the first mask coefficient, and the image position of the defect part in the first infrared image can be accurately determined according to the second mask, so that the defect part can be accurately positioned, and the identification accuracy of the defect part in the electrical equipment can be effectively improved.

In some embodiments, the first infrared image is de-noised before the temperature information of the electrical device to be identified is acquired from the first infrared image because the resolution of the first infrared image is poor and the visual effect is blurred. The denoising process specifically includes: the first infrared image is decomposed into three corresponding component images according to R, G, B components, the three component images obtained through decomposition are subjected to image filtering by adopting a filtering function, and the component images after filtering processing are combined into a new first infrared image, namely a synthesized and filtered first infrared image. The filter function may be a two-dimensional median filter function, a square filter function, a mean filter function, a gaussian filter function, a bilateral filter function, or the like.

According to the embodiment of the application, the resolution of the first infrared image can be improved, and the accuracy of the acquired temperature information is further improved.

In some embodiments, for an image area composed of pixels whose temperature values exceed a normal temperature range, pixels whose temperature values exceed the normal temperature range are defined as abnormal pixels. Determining a temperature anomaly region in the first infrared image according to the temperature information, including:

dividing the abnormal pixel points and the pixels adjacent to the abnormal pixel points or within a certain pixel interval into the same abnormal region; the abnormal region and an abnormal region adjacent to the abnormal region or an abnormal region within a certain pixel interval are determined as temperature abnormal regions.

For example, as shown in fig. 3, in a first infrared image with a size of 10×10, each unit cell 1 represents a pixel, and the pixel with the star mark is an abnormal pixel. The temperature anomaly area determined in the first infrared image may be an image area 2 formed by 6 pixels with a star mark, or may be an image area 3 formed by 6 pixels with a star mark and pixels adjacent to the 6 pixels with a star mark.

In some embodiments, after dividing the abnormal pixel point and the pixel points adjacent to the abnormal pixel point or the pixel points within a certain pixel interval into the same abnormal region, the method further includes:

Counting the number of abnormal pixels contained in each abnormal region, comparing the number with a set abnormal pixel threshold, and discharging the abnormal region if the number of abnormal pixels contained in the abnormal region is smaller than the set abnormal pixel threshold. This can avoid dividing the region into abnormal regions due to the abnormal temperature values of a few pixels, further increase the range of the abnormal temperature regions, and increase the operand, thereby reducing the operand and improving the image processing efficiency. In some embodiments

As an alternative embodiment, in the process of determining the image position of the defective component of the electrical device to be identified in the first infrared image according to the second mask by means of the defect determination model in step 106, the following steps may be performed:

Dividing the second mask of the temperature abnormal region through the defect determination model to obtain a third mask corresponding to each defective part in the temperature abnormal region;

In the embodiment of the application, after the electronic device obtains the second mask of the temperature abnormal region, the prediction frames of the defective parts in the second mask are segmented by the defect determination model to obtain the third mask, for example, the prediction frames of the three defective parts in the second mask can be segmented into three third masks respectively comprising one prediction frame by the loop operation in the Yolact model. The electronic device performs image binarization processing on each third mask, that is, according to a set Threshold (for example, 0.5, 0.4, etc., which is not limited herein), the gray value of the pixel point in the prediction frame of each third mask is changed to 0 or 255 through Threshold operation in the Yolact model, so that the display effect of the defective component in the prediction frame can be improved. The image position of each defective component in the temperature anomaly region of the prediction frame can be accurately marked according to each third mask after the binarization processing, and further, the image position can be converted into the position in the electrical equipment to be identified according to the image position in the first infrared image of each defective component. Defective components in the electrical device to be identified can be more clearly seen.

As an alternative embodiment, after performing the process of determining the image position of the defective component of the electrical device to be identified in the first infrared image according to the second mask by means of the defect determination model in step 106, the following steps may be performed:

Respectively comparing the temperature information of the first infrared sub-image and the defective part with the second infrared image of each component part of the electrical equipment to be identified in the database and the temperature information of each component part, wherein the second infrared image is an infrared image of each component part of the electrical equipment to be identified in a normal state;

According to the embodiment of the application, the electronic equipment intercepts a first infrared sub-image corresponding to the defective part from the first infrared image according to the determined image position of the defective part of the electric equipment to be identified in the first infrared image, for example, the sub-image of the defective part of the valve, and determines the infrared image of the region corresponding to the defective part as the first infrared sub-image. For the image area corresponding to the defective component, the determination may be made using the prediction box of the defective component in the second mask.

In the embodiment of the application, the electronic equipment also acquires the temperature information of the first infrared sub-image, namely the temperature information of the defective component, according to the intercepted first infrared sub-image of the defective component. The method for obtaining the temperature information of the defective component according to the first infrared sub-image is the same as the method for obtaining the temperature information of the first infrared image, and will not be described herein. In the embodiment of the application, the electronic equipment compares the first infrared sub-image with the second infrared image of each component part of the electrical equipment to be identified in the database, and compares the temperature information of the defective part with the temperature information of each component part in a normal state, wherein the second infrared image is the infrared image of each component part of the electrical equipment to be identified in the normal state.

For example, the database stores the infrared images of the components of the electrical equipment to be identified in the normal state, namely the second infrared image, including valves, pipelines, transformers, etc. The electronic equipment compares the first infrared sub-image of the defective part with each second infrared image one by one in a traversing mode so as to find the second infrared image corresponding to the first infrared sub-image.

In one embodiment, the image comparison method of the first infrared sub-image and each second infrared image may include: comparing the information of each pixel point in the first infrared sub-image with the information of the corresponding pixel point in each second infrared image; and judging whether the two pixel points belong to the matched pixel points according to the set similarity threshold value. And if the similarity between the information of each pixel point in the first infrared sub-image and the information of the corresponding pixel point in the second infrared image exceeds a similarity threshold value, the two pixel points are considered to be matched pixel points. And judging whether the first infrared sub-image corresponds to the second infrared image or not according to the set pixel point threshold value. And if the number of the matched pixels in the first infrared sub-image and the second infrared image exceeds the pixel threshold value, the first infrared sub-image is considered to correspond to the second infrared image.

In the embodiment of the application, the temperature information of the defective component is compared with the temperature information of each component in a normal state, specifically, the difference value between the temperature information of the defective component and the temperature information of each component in the normal state is calculated, whether the difference value is larger than a set difference value threshold value is judged, and if the difference value is larger than the set difference value threshold value, the component is further determined to be the defective component.

In the embodiment of the application, each second infrared image is bound with information of a corresponding component of the electrical equipment, for example, when one second infrared image is an infrared image in a normal state of the valve, the second infrared image is bound with information of the valve component, and the information can include information such as a component name, a component size, a position of the component in the electrical equipment and the like. The manner in which each second infrared image is bound to information of the component parts of the corresponding electrical apparatus is not particularly limited herein.

According to the embodiment of the application, the electronic equipment acquires the information of the component parts of the electrical equipment bound by the second infrared image according to the comparison result of the first infrared sub-image and the second infrared image, namely according to the second infrared image corresponding to the first infrared sub-image after image comparison, and further determines the part information of the defect part, namely acquires the name, the size, the position and the like of the defect part in the electrical equipment. And the accuracy of the determination of the defective component is further verified according to the comparison result of the temperature information of the defective component and the temperature information of each component in a normal state, so that the defective component can be further determined and the related information of the defective component can be effectively determined, the replacement or maintenance of the subsequent defective component and the improvement of the identification accuracy of the defective component are facilitated, and the safety of the electrical equipment is effectively improved.

In one embodiment, as shown in fig. 4, the step of extracting the features of the temperature anomaly region through the trained defect determination model, generating a plurality of first masks corresponding to the temperature anomaly region according to the features, and determining the first mask coefficients of the defective component in the temperature anomaly region may include the following steps:

401. and processing the temperature abnormal region through a plurality of convolution modules of the depth residual error network to obtain a plurality of first characteristic diagrams with different output sizes.

In the embodiment of the application, the electronic equipment inputs the second infrared sub-image into a network for feature extraction to obtain a plurality of first feature images with different sizes. And taking the corresponding infrared image part of the determined temperature abnormal region in the first infrared image as a second infrared sub-image. The network for feature extraction in the defect determination model is a depth residual network with a plurality of convolution modules. And the depth residual error network performs feature extraction on the second infrared sub-image and then outputs a plurality of first feature images.

402. And processing the first characteristic diagram with the smallest output size through a convolution layer to obtain a second characteristic diagram.

403. And performing convolution and downsampling operation on the second feature map to obtain deep network feature map features.

404. And carrying out convolution processing on the first characteristic diagram with the second small output size to obtain a third characteristic diagram.

405. And amplifying the second characteristic diagram, and summing the amplified second characteristic diagram and the third characteristic diagram to obtain a shallow network characteristic diagram.

In the embodiment of the application, in order to enrich the features contained in the feature map of the temperature anomaly region and generate multi-scale feature representation, the method is more beneficial to determining the defect part in the electrical equipment, and the trained defect determination model can be provided with a network structure for further feature extraction processing of the first feature map. Wherein the network structure is optionally a FPN network.

In the embodiment of the application, after the feature extraction processing is performed on the second infrared sub-image by the depth residual error network in the trained defect determination model, a plurality of first feature images are obtained, and the first feature images are ranked from large to small according to the size. After the first feature images are sequenced, the defect determination model carries out convolution operation on the feature image with the smallest size and the feature image with the second smallest size in the first feature images to obtain a second feature image and a third feature image. And expanding the second characteristic diagram by adopting bilinear interpolation on the obtained second characteristic diagram, and adding the expanded second characteristic diagram with the third characteristic diagram to obtain the deep network characteristic diagram. And additionally, rolling and downsampling the second feature map to obtain a shallow network feature map.

For example, after feature extraction processing is performed on the second infrared sub-image by the depth residual error network in the trained defect determination model, three first feature images are obtained, and the first feature images are respectively ranked as C3-C5 from large to small according to the size. As shown in fig. 5, the trained defect determination model performs convolution operation on the feature map C5 with the smallest size in the first feature map by 1 convolution layer to obtain a second feature map, that is, feature map P5, and then performs bilinear interpolation on the second feature map to double the second feature map, and adds the amplified second feature map to the third feature map C4 with the second smallest output size after the convolution operation to obtain a deep network feature map P4. The second feature map P5 is rolled and downsampled to obtain a shallow network feature map P6. The deep network feature map is obtained through downsampling, has the characteristic that a higher receptive field is abstract, and can better express the information of the first infrared sub-image; the shallow network feature map is enlarged, so that the shallow network feature map has larger size and more detail information.

By implementing the method, the characteristic diagrams with different receptive fields and different sizes can be obtained, the first mask generated according to the characteristic diagrams and the determined first mask coefficient can contain the information of the defect part as much as possible, and the accuracy of defect identification is improved.

In some embodiments, to obtain more different receptive fields and feature maps of different sizes, it may include: carrying out convolution processing on the first feature map with the third small output size to obtain a fourth feature map;

Amplifying the shallow network characteristic diagram, and summing the amplified shallow network characteristic diagram with the fourth characteristic diagram to obtain a second shallow network characteristic diagram;

and performing convolution and downsampling operation on the deep network feature map to obtain a second deep network feature map.

For example, when the trained defect determination model obtains 5 first feature images after feature extraction processing is performed on the second infrared sub-image, the first feature images are respectively ranked as C1-C5 from large to small according to the size. After the defect determination model obtains the second feature map, the third feature map and the fourth feature map through the feature extraction process, the defect determination model can further perform feature extraction processing on the second feature map, the shallow network feature map and the deep network feature map through the trained defect determination model on the basis of the third feature map and the fourth feature map so as to obtain more feature maps with different scales. The specific method is as follows: and (3) the shallow network characteristic diagram P4 is also subjected to bilinear interpolation to double the characteristic diagram, and then added with a fourth characteristic diagram C3 to obtain a second shallow network characteristic diagram P3, wherein the fourth characteristic diagram C3 is obtained by carrying out convolution operation on the first characteristic diagram with the third small output size. The deep network feature map P6 is subjected to the same convolution and down-adopted operation as the first feature map P5 with the second smallest output size, and a second deep network feature map P7 is obtained.

According to the embodiment of the application, more feature images with different sizes can be obtained according to the acquired plurality of first feature images, and the accuracy of defect identification is improved.

406. And generating a plurality of first masks corresponding to the temperature abnormal region according to the second characteristic diagram, the deep network characteristic diagram and the shallow network characteristic diagram, and determining first mask coefficients of defective components in the temperature abnormal region.

In some embodiments, as an optional implementation, the defect determination model further includes a mask generation network and a coefficient generation network, where the mask generation network and the coefficient generation network are parallel networks.

Step 406 may include: inputting the shallow network feature map into a mask generation network, and generating a plurality of first masks corresponding to the temperature abnormal region based on the shallow network feature map through the mask generation network; and the second characteristic diagram, the shallow network characteristic diagram and the deep network characteristic diagram are input into a coefficient generation network together, and a first mask coefficient of a defect part in the temperature abnormal region is determined based on the second characteristic diagram, the shallow network characteristic diagram and the deep network characteristic diagram through the coefficient generation network.

In the embodiment of the application, if the 3 first feature images are further processed in the trained defect determination model to obtain the second feature image, the shallow network feature image and the deep network feature image, the shallow network feature image P4 is input into a mask generation network to generate a plurality of first masks, and the mask generation network can be selected as a full convolution network. The second feature map P5, the shallow network feature map P4, and the deep network feature map P6 are input into a coefficient generation network to generate a first mask coefficient. If in the trained defect determination model, further feature processing is performed on the 5 first feature images to obtain a second feature image, a shallow network feature image, a deep network feature image, a second shallow network feature image P3 and a second deep network feature image P7, the second shallow network feature image P3 is input into a mask generation network to generate a plurality of first masks.

According to the second feature map subjected to convolution operation, the shallow network feature map subjected to image expansion operation, the deep network feature map subjected to downsampling operation and the acquisition process of the feature maps P3 and P7, the feature map with larger size and more detailed information of the temperature anomaly region is input into the mask generation network. The mask generation network may be selected to be a full convolution network. The second feature map P5, the shallow network feature map P4, the deep network feature map P6, the second shallow network feature map P3, and the second deep network feature map P7 are input into a coefficient generation network to generate a first mask coefficient. The coefficient generation network may be selected to employ RETINA NET of a shared convolutional network. In the trained defect determination model, a process of generating a plurality of first masks after the masks generate the network input feature images and a process of generating first mask coefficients after the coefficients generate the network input feature images are used as two parallel tasks, so that the speed of the process of determining the defect components in the electrical equipment by the trained defect determination model can be improved, and the recognition efficiency is improved.

By adopting the method for determining the defects of the electrical equipment, which is described by the embodiment, the characteristic diagrams with different sizes can be obtained, and the obtained characteristic diagrams with different sizes can better express the global information of the temperature abnormal region and the information of the defective parts in the temperature abnormal region, so that the accuracy of defect determination is improved.

Referring to fig. 6, fig. 6 is a flowchart illustrating another method for determining defects of an electrical device according to an embodiment of the application. As shown in fig. 6, the electrical device defect determination method may include the steps of:

601. A first infrared image of an electrical device to be identified is acquired.

602. And acquiring temperature information of the electrical equipment to be identified according to the first infrared image.

603. And determining a temperature abnormal region in the first infrared image according to the temperature information.

604. And extracting a shallow network feature map of the temperature abnormal region through the trained defect determination model, and generating a plurality of first masks corresponding to the temperature abnormal region according to the shallow network feature map.

605. In the defect determination model, a plurality of prediction frames of the defect component in the temperature anomaly area are generated according to the second feature map, the shallow network feature map and the deep network feature map, and classification confidence degrees which are in one-to-one correspondence with the prediction frames and second mask coefficients which are in one-to-one correspondence with the prediction frames are determined.

606. And screening the plurality of prediction frames according to the classification confidence degrees corresponding to the prediction frames through the defect determination model, determining a target prediction frame uniquely corresponding to the defect part, and determining a second mask coefficient corresponding to the target prediction frame as a first mask coefficient.

607. And generating a second mask of the temperature abnormal region according to the plurality of first masks and the first mask coefficients through the defect determination model.

608. And determining the image position of the defective part of the electrical equipment to be identified in the first infrared image according to the second mask through the defect determination model.

In the embodiment of the application, the prediction frame is the above prediction frame, and the classification confidence is the confidence corresponding to the prediction frame when the prediction frame is a defective component, for example, the confidence is 0.1 when the prediction frame is a defective valve, and the confidence is 0.9 when the prediction frame is a defective pipeline.

In the embodiment of the application, an NMS algorithm can be adopted to screen prediction frames according to classification confidence, specifically, 5 prediction frames are arranged for one defective component in a temperature abnormal region, and confidence levels of the 5 prediction frames corresponding to the defective component are ranked from high to low to obtain B1-B5; calculating the interaction ratio (Intersection over Union, IOU) of the 5 prediction frames through matrix operation to obtain a symmetric matrix; deleting diagonal lines and lower triangle elements of the symmetric matrix, and taking the maximum value of each column in the rest matrix to obtain IOU values of all prediction frames; and discarding the prediction frame with the IOU value larger than the threshold according to the set threshold. This is because each element in the matrix has a row number smaller than a column number, and the sequence numbers are arranged in descending order of confidence corresponding to the defective component, so that any element is larger than the threshold value, meaning that the prediction frame corresponding to this column is too overlapped with the prediction frame having a higher confidence than this prediction frame, and therefore needs to be discarded. Redundant prediction frames can be effectively removed, and the prediction frame with the highest overlapping degree with the standard prediction frame is reserved.

In the embodiment of the application, each prediction frame is screened through a defect determination model, so that a first prediction frame uniquely corresponding to each defective component, namely, a prediction frame uniquely corresponding to each defective component is determined, and a mask coefficient corresponding to the prediction frame is determined as a first mask coefficient, namely, the mask coefficient of the prediction frame relative to each first mask is determined as the first mask coefficient of the defective component corresponding to the confidence.

Referring to fig. 7, fig. 7 is a block diagram of an electrical equipment defect determining apparatus according to an embodiment of the present application. As shown in fig. 7, the electrical equipment defect determining apparatus may include:

The image acquisition module 701 is configured to acquire a first infrared image of an electrical device to be identified acquired by an infrared camera;

the temperature acquisition module 702 is configured to acquire temperature information of an electrical device to be identified according to the first infrared image;

A region determining module 703, configured to determine a temperature anomaly region in the infrared image according to the temperature information;

A feature extraction module 704, configured to extract features of the temperature anomaly region through the trained defect determination model, generate a plurality of first masks corresponding to the temperature anomaly region according to the features, and determine first mask coefficients of defective components in the temperature anomaly region;

A mask generating module 705, configured to generate, by using the defect determination model, a second mask of the temperature anomaly region according to the plurality of first masks and the first mask coefficients;

A defect determination module 706 is configured to determine, from the second mask, an image location of the defective component of the electrical device to be identified in the first infrared image, by means of the defect determination model.

It can be seen that, with the electrical equipment defect determining apparatus described in the foregoing embodiment, an image corresponding to a temperature anomaly region in a first infrared image can be determined according to temperature information obtained from a first infrared image of an electrical equipment to be identified, after features of the corresponding image are extracted by a trained defect determining model, a plurality of first masks with the same size as the corresponding image are generated according to the features, and a second mask capable of indicating a defective part in the temperature anomaly region is generated according to the plurality of first masks and the first mask coefficients, so that an image position of the defective part in the first infrared image can be accurately determined according to the second mask, thereby accurately positioning the defective part and effectively improving identification accuracy of a defective part in the electrical equipment.

Referring to fig. 8, fig. 8 is a block diagram of another defect determining apparatus for electrical equipment according to an embodiment of the present application. Wherein the electrical equipment defect determining device shown in fig. 8 is optimized by the electrical equipment defect determining device shown in fig. 7. Compared with the electrical equipment defect determining apparatus shown in fig. 7, the electrical equipment defect determining apparatus shown in fig. 8 may further include:

An image comparison module 801, configured to compare the first infrared sub-image and the temperature information of the defective component with a second infrared image of each component of the electrical device to be identified in the database and the temperature information of each component, where the second infrared image is an infrared image of each component of the electrical device to be identified in a normal state;

An information determining module 802, configured to determine component information of the defective component according to the result of the comparison.

The above-mentioned area determining module 703 may be further configured to intercept a first infrared sub-image corresponding to the defective component from the first infrared image according to the image position.

The temperature acquisition module 702 may be further configured to acquire temperature information of the defective component according to the first infrared sub-image.

Therefore, the defect determining device for the electrical equipment, which is described in the embodiment, can determine the related information of the defective component, and is beneficial to the replacement or maintenance of the subsequent defective component, so that the safety of the electrical equipment is improved.

As an alternative embodiment, the feature extraction module 704 may be further configured to:

Processing the temperature abnormal region through a plurality of convolution modules of the depth residual error network to obtain a plurality of first characteristic diagrams with different output sizes; processing the first feature map with the smallest output size through a convolution layer to obtain a second feature map; performing convolution and downsampling operation on the second feature map to obtain a deep network feature map; performing convolution processing on the first feature map with the second small output size to obtain a third feature map; and amplifying the second characteristic diagram, and summing the amplified second characteristic diagram and the third characteristic diagram to obtain the shallow network characteristic diagram.

Therefore, by adopting the electrical equipment defect determining device described in the above embodiment, the feature maps with different sizes can be obtained, and the obtained feature maps with different sizes can better express the global information of the temperature anomaly area and the information of the defect component in the temperature anomaly area, so that the accuracy of defect determination is improved.

As an alternative embodiment, the defect determination model further comprises a mask generation network and a coefficient generation network, the mask generation network and the coefficient generation network being parallel networks.

The feature extraction module 704 may be further configured to:

Inputting the shallow network feature map into a mask generation network, and generating a plurality of first masks corresponding to the temperature abnormal region based on the shallow network feature map through the mask generation network; and the second characteristic diagram, the shallow network characteristic diagram and the deep network characteristic diagram are input into a coefficient generation network together, and a first mask coefficient of a defect part in the temperature abnormal region is determined based on the second characteristic diagram, the shallow network characteristic diagram and the deep network characteristic diagram through the coefficient generation network.

Therefore, by adopting the electrical equipment defect determining device described in the above embodiment, parallel operation of a plurality of first mask acquiring processes and first mask coefficient acquiring processes can be realized, and the acquiring speed of the defect part prediction frame can be effectively improved, so that the speed of the defect determining whole process is improved, and the timeliness of safety detection is enhanced.

As an alternative embodiment, the defect determining module 706 may be further configured to:

Dividing the second mask of the temperature abnormal region through a defect determination model to obtain a third mask of each part in the temperature abnormal region; performing image binarization processing on the third mask according to a preset threshold value through a defect determination model to obtain a fourth mask; and determining the image position of the defective part of the electrical equipment to be identified in the first infrared image according to the fourth mask.

It can be seen that, with the electrical equipment defect determining apparatus described in the above embodiments, display optimization of the mask corresponding to the defective part of the electrical equipment can be achieved, so that the defective part can be better displayed and determined.

In the defect determination model, generating a plurality of prediction frames of the defect part in the temperature abnormal region according to the characteristics, and determining classification confidence degrees corresponding to the prediction frames one by one and second mask coefficients corresponding to the prediction frames one by one; screening the plurality of prediction frames according to the classification confidence degrees corresponding to the prediction frames, determining a target prediction frame uniquely corresponding to the defect part, and determining a second mask coefficient corresponding to the target prediction frame as a first mask coefficient.

Therefore, by adopting the electrical equipment defect determining device described in the above embodiment, the classification of the defective component and the prediction frame which may correspond to the defective component can be screened, and the accuracy of the obtained prediction frame of the defective component can be improved.

Referring to fig. 9, fig. 9 is a schematic structural diagram of an electronic device according to an embodiment of the application.

As shown in fig. 9, the electronic device may include:

a memory 901 storing executable program code;

a processor 902 coupled to the memory 901;

The processor 902 invokes executable program code stored in the memory 901 to perform an electrical device defect determination method of fig. 1,4 or 6.

The embodiment of the application discloses a computer-readable storage medium storing a computer program, wherein the computer program causes a computer to execute an electrical equipment defect determining method of fig. 1, 4 or 6.

The embodiment of the application also discloses an application release platform, wherein the application release platform is used for releasing a computer program product, and the computer program product is used for enabling the computer to execute part or all of the steps of the method in the method embodiments.

It should be appreciated that reference throughout this specification to "one embodiment" or "an embodiment" means that a particular feature, structure or characteristic described in connection with the embodiment is included in at least one embodiment of the present application. Thus, the appearances of the phrases "in some embodiments" or "in an embodiment" in various places throughout this specification are not necessarily all referring to the same embodiment. Furthermore, the particular features, structures, or characteristics may be combined in any suitable manner in one or more embodiments. Those skilled in the art will also appreciate that the embodiments described in the specification are alternative embodiments and that the acts and modules referred to are not necessarily required for the present application.

In various embodiments of the present application, it should be understood that the sequence numbers of the foregoing processes do not imply that the execution sequences of the processes should be determined by the functions and internal logic of the processes, and should not be construed as limiting the implementation of the embodiments of the present application.

The units described above as separate components may or may not be physically separate, and components shown as units may or may not be physical units, may be located in one place, or may be distributed over a plurality of network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the embodiment.

In addition, each functional unit in the embodiments of the present application may be integrated in one processing unit, or each unit may exist alone physically, or two or more units may be integrated in one unit. The integrated units may be implemented in hardware or in software functional units.

The integrated units described above, if implemented in the form of software functional units and sold or used as stand-alone products, may be stored in a computer-accessible memory. Based on this understanding, the technical solution of the present application, or a part contributing to the prior art or all or part of the technical solution, may be embodied in the form of a software product stored in a memory, comprising several requests for a computer device (which may be a personal computer, a server or a network device, etc., in particular may be a processor in a computer device) to execute some or all of the steps of the above-mentioned method of the various embodiments of the present application.

Those of ordinary skill in the art will appreciate that all or part of the steps of the various methods of the above embodiments may be implemented by a program that instructs associated hardware, the program may be stored in a computer readable storage medium including Read-Only Memory (ROM), random access Memory (Random Access Memory, RAM), programmable Read-Only Memory (Programmable Read-Only Memory, PROM), erasable programmable Read-Only Memory (Erasable Programmable Read Only Memory, EPROM), one-time programmable Read-Only Memory (OTPROM), electrically erasable programmable Read-Only Memory (EEPROM), compact disc Read-Only Memory (Compact Disc Read-Only Memory, CD-ROM) or other optical disk Memory, magnetic disk Memory, tape Memory, or any other medium that can be used for carrying or storing data.

The foregoing describes in detail a method for determining defects of electrical equipment and terminal equipment disclosed in the embodiments of the present application, and specific examples are applied to illustrate the principles and embodiments of the present application, and the description of the foregoing examples is only for helping to understand the method and core ideas of the present application; meanwhile, as those skilled in the art will have variations in the specific embodiments and application scope in accordance with the ideas of the present application, the present description should not be construed as limiting the present application in view of the above.

Claims

1. A method of determining defects in an electrical device, the method comprising:

Acquiring a first infrared image of the electrical equipment to be identified;

Determining the image position of the defective part of the electrical equipment to be identified in the first infrared image according to the second mask through the defect determination model;

2. The method according to claim 1, wherein the defect determination model is trained by a training image set including infrared images corresponding to when each component part in the electrical apparatus is a defective part, and the infrared images are labeled with information when each component part in the electrical apparatus is a defective part.

3. The method of claim 1, wherein the defect determination model comprises a depth residual network having a plurality of convolution modules, the extracting features of the temperature anomaly region by the trained defect determination model comprising:

4. A method according to claim 3, wherein the defect determination model further comprises a mask generation network and a coefficient generation network, the mask generation network and the coefficient generation network being parallel networks;

5. The method according to claim 1, wherein determining, by the defect determination model, an image position of a defective component of the electrical device to be identified in the first infrared image from the second mask comprises:

6. The method of claim 1, wherein determining the first mask coefficient for the defective component in the temperature anomaly region comprises:

7. An electrical equipment defect recognition apparatus, characterized by comprising:

The region determining module is used for determining a temperature abnormal region in the infrared image according to the temperature information;

The feature extraction module is used for extracting the features of the temperature abnormal region through the trained defect determination model, generating a plurality of first masks corresponding to the temperature abnormal region according to the features, and determining first mask coefficients of defective components in the temperature abnormal region;

a mask generation module, configured to generate, according to the plurality of first masks and the first mask coefficients, a second mask of the temperature anomaly region through the defect determination model;

A defect determining module, configured to determine, according to the second mask, an image position of a defective component of the electrical device to be identified in the first infrared image by using the defect determining model;

The region determining module is further configured to intercept a first infrared sub-image corresponding to the defective component from the first infrared image according to the image position;

The temperature acquisition module is further used for acquiring temperature information of the defect part according to the first infrared sub-image;

The image comparison module is used for respectively comparing the temperature information of the first infrared sub-image and the defective component with the second infrared image of each component of the electrical equipment to be identified in the database and the temperature information of each component, wherein the second infrared image is an infrared image of each component of the electrical equipment to be identified in a normal state;

And the information determining module is used for determining the component information of the defective component according to the comparison result.

8. An electronic device, comprising: a memory and a processor, the memory having stored thereon a computer program, wherein the computer program is executable by the processor to cause the processor to implement the method of any of claims 1 to 6.

9. A computer readable storage medium, characterized in that the computer readable storage medium has stored therein a computer program adapted to be loaded and executed by a processor to cause the processor to carry out the method according to any one of claims 1-6.