CN115760854A - Deep learning-based power equipment defect detection method and device and electronic equipment - Google Patents
Deep learning-based power equipment defect detection method and device and electronic equipment Download PDFInfo
- Publication number
- CN115760854A CN115760854A CN202310026789.7A CN202310026789A CN115760854A CN 115760854 A CN115760854 A CN 115760854A CN 202310026789 A CN202310026789 A CN 202310026789A CN 115760854 A CN115760854 A CN 115760854A
- Authority
- CN
- China
- Prior art keywords
- image
- foreground
- power equipment
- region
- service life
- 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
- 238000001514 detection method Methods 0.000 title claims abstract description 145
- 230000007547 defect Effects 0.000 title claims abstract description 118
- 238000013135 deep learning Methods 0.000 title claims abstract description 34
- 230000007797 corrosion Effects 0.000 claims abstract description 70
- 238000005260 corrosion Methods 0.000 claims abstract description 70
- 230000011218 segmentation Effects 0.000 claims abstract description 60
- 230000009466 transformation Effects 0.000 claims abstract description 57
- 238000000034 method Methods 0.000 claims abstract description 42
- JEIPFZHSYJVQDO-UHFFFAOYSA-N iron(III) oxide Inorganic materials O=[Fe]O[Fe]=O JEIPFZHSYJVQDO-UHFFFAOYSA-N 0.000 claims description 33
- 238000012545 processing Methods 0.000 claims description 13
- 238000004590 computer program Methods 0.000 claims description 11
- 238000004364 calculation method Methods 0.000 claims description 8
- 238000011088 calibration curve Methods 0.000 claims description 3
- 230000005540 biological transmission Effects 0.000 claims description 2
- 238000004422 calculation algorithm Methods 0.000 description 8
- 238000004891 communication Methods 0.000 description 8
- 230000006870 function Effects 0.000 description 8
- 230000008569 process Effects 0.000 description 6
- 238000007726 management method Methods 0.000 description 5
- 238000010586 diagram Methods 0.000 description 4
- 230000009471 action Effects 0.000 description 3
- 238000012423 maintenance Methods 0.000 description 3
- 238000013528 artificial neural network Methods 0.000 description 2
- 238000000605 extraction Methods 0.000 description 2
- 238000003709 image segmentation Methods 0.000 description 2
- 239000004973 liquid crystal related substance Substances 0.000 description 2
- 230000003287 optical effect Effects 0.000 description 2
- 238000013473 artificial intelligence Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 238000003064 k means clustering Methods 0.000 description 1
- 238000002372 labelling Methods 0.000 description 1
- 238000010801 machine learning Methods 0.000 description 1
- 239000000203 mixture Substances 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000002093 peripheral effect Effects 0.000 description 1
- 238000011176 pooling Methods 0.000 description 1
- 230000005855 radiation Effects 0.000 description 1
- 238000000926 separation method Methods 0.000 description 1
- 238000012163 sequencing technique Methods 0.000 description 1
- 238000002834 transmittance Methods 0.000 description 1
Images
Landscapes
- Radiation Pyrometers (AREA)
Abstract
The invention relates to the field of intelligent decision making, and discloses a method and a device for detecting defects of power equipment based on deep learning and electronic equipment, wherein the method comprises the following steps: collecting service life data, infrared images and corrosion images of the power equipment, and determining a service life detection value of the power equipment; extracting image characteristics in the infrared image, identifying the region to be detected and the region type thereof from the infrared image, and determining an equipment image of the infrared image by using the region to be detected and the region type thereof; calculating the actual temperature of the power equipment; performing foreground and background segmentation on the corrosion image to obtain a segmentation foreground and a segmentation background, performing brightness transformation on the segmentation foreground to obtain a transformation foreground, identifying a corrosion region in the transformation foreground, and calculating a corrosion detection value of the corrosion region; and constructing defect factor weights of the service life detection value, the actual temperature and the corrosion detection value, and determining a defect detection result of the power equipment based on the defect factor weights. The invention can improve the accuracy of detecting the defects of the power equipment.
Description
Technical Field
The invention relates to the field of intelligent decision making, in particular to a method and a device for detecting defects of power equipment based on deep learning and electronic equipment.
Background
The electric power equipment defect detection based on deep learning refers to a process of realizing the defect detection of the electric power equipment by utilizing a machine learning algorithm.
At present, when defect detection is carried out on electric equipment based on an image processing technology, images are segmented firstly and then detected, and the method has the main defects that the images cannot be segmented well when the background of the images is complex, and the defect identification precision is not high; when the target equipment region extraction method based on Canny edge feature extraction and K-means clustering is used for realizing defect detection of electric equipment, the accuracy of a final result is insufficient due to the fact that features are designed manually, subjectivity is strong, and anti-interference capability is poor. Therefore, the accuracy of detecting the defects of the power equipment is low.
Disclosure of Invention
In order to solve the above problems, the invention provides a method and a device for detecting defects of electrical equipment based on deep learning, and an electronic device, which can improve the accuracy of detecting defects of electrical equipment.
In a first aspect, the invention provides a method for detecting defects of electrical equipment based on deep learning, which comprises the following steps:
collecting service life data, an infrared image and a corrosion image of the power equipment, and determining a service life detection value of the power equipment according to the service life data;
extracting image features in the infrared image, identifying a region to be detected and a region type thereof from the infrared image according to the image features, and determining an equipment image in the infrared image by using the region to be detected and the region type thereof;
calculating the actual temperature of the power equipment according to the equipment image;
performing foreground and background segmentation on the corrosion image to obtain a segmentation foreground and a segmentation background, performing brightness transformation on the segmentation foreground based on the segmentation background to obtain a transformation foreground, identifying a corrosion region in the transformation foreground, and calculating a corrosion detection value of the corrosion region;
and constructing defect factor weights of the service life detection value, the actual temperature and the corrosion detection value, and determining a defect detection result of the power equipment based on the defect factor weights.
In a possible implementation manner of the first aspect, the determining a lifetime detection value of the power device according to the lifetime data includes:
according to the service life data, the service life grade of the power equipment is determined;
and setting a grade index of the service life grade, and taking the grade index as the service life detection value.
In a possible implementation manner of the first aspect, the identifying, according to the image feature, the region to be detected and the region type thereof from the infrared image includes:
carrying out window sliding on the image characteristics to obtain an image anchor point;
performing bounding box regression on the image anchor points to obtain regression anchor points;
classifying the image anchor points to obtain classified anchor points;
and taking the regression anchor point as the area to be detected, and taking the classification anchor point as the area type.
In one possible implementation manner of the first aspect, the calculating an actual temperature of the electric power device according to the device image includes:
calculating an image heat of the device image using the following formula:
wherein,an image heat representing the image of the device,a gray value representing an image of the device,indicating the thermal range of the infrared imager device,indicating the thermal level of the infrared imager device,which is indicative of the degree of transmission of light,represents the emissivity of the object;
according to the image heat, calculating the actual temperature of the power equipment by using the following formula:
wherein,is indicative of the actual temperature of the electrical equipment,representing the image heat of the device image,representing the calibration curve constant of the infrared imager device.
In a possible implementation manner of the first aspect, the performing, based on the segmented background, luminance transformation on the segmented foreground to obtain a transformed foreground includes:
calculating the average brightness of the segmented background using the following formula:
wherein,represents an average brightness of the segmented background,representing the brightness of the segmented background,a red gray value representing the segmented background,a green gray value representing the segmented background,representing a blue gray value of the segmented background,expressing the pixel coordinates of the segmented background, and k expresses the number of all pixel points with the brightness not being 0;
and calculating a transformation parameter of the segmentation foreground according to the average brightness by using the following formula:
wherein,a transformation parameter representing the segmented foreground,representing the average brightness of the segmented background, and 128 representing an image brightness standard value;
according to the transformation parameters, carrying out brightness transformation on the segmentation foreground by using the following formula to obtain the transformation foreground:
wherein,the foreground of the transformation is represented and,a transformation parameter representing the segmented foreground,representing the segmentation foreground.
In one possible implementation of the first aspect, the identifying of the rusty areas in the transformed foreground comprises: calculating a rust gray value in the transformed foreground using the following formula:
wherein,represent the transformation foregroundThe value of the rust corrosion gray-scale value of (c),respectively three channels of gray values of the foreground picture,pixel coordinates representing the transformed foreground;
according to the corrosion gray value, identifying the regional color in the transformation foreground by using the following formula to obtain a color identifier:
wherein r represents the color designation,representing a rust grey value in the transformed foreground,、、、andrespectively representing orange, yellow, pink, red and dark red color marks;
determining a tarnish region in the transformed foreground based on the color identification.
In one possible implementation manner of the first aspect, the determining a defect detection result of the power device based on the defect factor weight includes:
acquiring the service life detection value, the actual temperature and the corrosion detection value;
carrying out weighted summation processing on the service life detection value, the actual temperature and the corrosion detection value by using the defect factor weight to obtain a weighted summation value;
determining a defect detection level of the electrical equipment based on the weighted summation value;
respectively weighting the service life detection value, the actual temperature and the corrosion detection value by using the defect factor weight to obtain weighted values;
determining a defect detection category of the power equipment based on the weighted processing numerical value;
and taking the defect detection grade and the defect detection category as the defect detection result.
In a second aspect, the present invention provides an apparatus for detecting defects of an electrical device based on deep learning, the apparatus including:
the detection value determining module is used for acquiring the service life data, the infrared image and the corrosion image of the power equipment and determining the service life detection value of the power equipment according to the service life data;
the device determining module is used for extracting image features in the infrared image, identifying a region to be detected and a region type thereof from the infrared image according to the image features, and determining a device image in the infrared image by using the region to be detected and the region type thereof;
the temperature calculation module is used for calculating the actual temperature of the power equipment according to the equipment image;
the rust value calculation module is used for carrying out foreground background segmentation on the rust image to obtain a segmentation foreground and a segmentation background, carrying out brightness transformation on the segmentation foreground based on the segmentation background to obtain a transformation foreground, identifying a rust region in the transformation foreground, and calculating a rust detection value of the rust region;
and the result determining module is used for constructing the defect factor weights of the service life detection value, the actual temperature and the corrosion detection value and determining the defect detection result of the power equipment based on the defect factor weights.
In a third aspect, the present invention provides an electronic device comprising:
at least one processor; and a memory communicatively coupled to the at least one processor;
wherein the memory stores a computer program executable by the at least one processor to enable the at least one processor to perform the method of deep learning based power equipment fault detection as defined in any one of the first aspects above.
In a fourth aspect, the present invention provides a computer-readable storage medium storing a computer program which, when executed by a processor, implements the deep learning-based power equipment fault detection method according to any one of the first aspects.
Compared with the prior art, the technical principle and the beneficial effects of the scheme are as follows:
the method comprises the steps of firstly collecting life data, infrared images and corrosion images of the power equipment to analyze the defect degree of the power equipment by utilizing the collected historical data of the power equipment, and further determining a life detection value of the power equipment according to the life data to take the service life of the equipment as a factor for evaluating the defect of the equipment; secondly, extracting image features in the infrared image to frame a candidate area containing equipment from the image based on the features, and further, identifying an area to be detected and an area type thereof from the infrared image according to the image features to remove the area irrelevant to the equipment in the image and identify the type of the equipment in the image, and detecting the equipment by using a fast RCNN detection algorithm to improve the detection accuracy; further, the actual temperature of the electrical equipment is calculated according to the equipment image, so that whether the temperature of the electrical equipment has defects or not is determined according to the one-to-one correspondence relationship between the gray value and the temperature value; further, the embodiment of the present invention performs foreground-background segmentation on the rusty image to detect the brightness of a background, and implements subsequent image enhancement based on the brightness of the background, and further performs brightness transformation on the segmented foreground based on the segmented background to enhance the dark details in the image, and simultaneously reduces the brightness of the image with excessively high brightness, so as to ensure the balance of the image brightness, and further, identifies the rusty region in the transformed foreground to highlight the rusty device region, and accordingly, since the region without obvious red information interferes with the grayed image, the dark red feature of the rusty region needs to be effectively distinguished from other regions, and further, calculates the rusty detection value of the rusty region to digitize the abstract rusty region, thereby facilitating numerical statistics of device defect detection; further, the defect factor weights of the service life detection value, the actual temperature and the corrosion detection value are constructed so as to distinguish the influence degree of different defect factors on the detection result by using the weights, and the accuracy of the defect detection result is improved. Therefore, the method, the device and the electronic device for detecting the defects of the electrical equipment based on the deep learning, which are provided by the embodiment of the invention, can improve the accuracy rate of detecting the defects of the electrical equipment.
Drawings
The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the invention and together with the description, serve to explain the principles of the invention.
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly described below, and it is obvious for those skilled in the art that other drawings can be obtained according to the drawings without inventive exercise.
Fig. 1 is a schematic flowchart of a method for detecting defects of an electrical device based on deep learning according to an embodiment of the present invention;
fig. 2 is a schematic flowchart illustrating a step of the method for detecting defects of electrical equipment based on deep learning in fig. 1 according to an embodiment of the present invention;
fig. 3 is a schematic flowchart illustrating another step of the method for detecting defects of electrical equipment based on deep learning in accordance with an embodiment of the present invention;
fig. 4 is a schematic block diagram of an apparatus for detecting defects of electrical equipment based on deep learning according to an embodiment of the present invention;
fig. 5 is a schematic diagram of an internal structure of an electronic device implementing a deep learning-based power device defect detection method according to an embodiment of the present invention.
Detailed Description
It should be understood that the detailed description and specific examples, while indicating the preferred embodiment of the invention, are given by way of illustration only.
The embodiment of the invention provides a method for detecting defects of electric power equipment based on deep learning, wherein an execution subject of the method for detecting defects of electric power equipment based on deep learning comprises but is not limited to at least one of electronic equipment such as a server and a terminal which can be configured to execute the method provided by the embodiment of the invention. In other words, the method for detecting defects of power equipment based on deep learning may be performed by software or hardware installed in a terminal device or a server device, and the software may be a block chain platform. The server includes but is not limited to: a single server, a server cluster, a cloud server or a cloud server cluster, and the like. The server may be an independent server, or may be a cloud server that provides basic cloud computing services such as a cloud service, a cloud database, cloud computing, a cloud function, cloud storage, a Network service, cloud communication, a middleware service, a domain name service, a security service, a Content Delivery Network (CDN), a big data and artificial intelligence platform, and the like.
Fig. 1 is a schematic flow chart of a method for detecting defects of an electrical device based on deep learning according to an embodiment of the present invention. The method for detecting the defects of the power equipment based on the deep learning, which is described in the figure 1, comprises the following steps:
s1, collecting life data, infrared images and corrosion images of the power equipment, and determining a life detection value of the power equipment according to the life data.
According to the embodiment of the invention, the service life data, the infrared image and the corrosion image of the power equipment are collected, so that the collected historical data of the power equipment is utilized to analyze the defect degree of the power equipment.
The service life data includes service time data and maintenance time limit data of the power equipment, for example, the service time data is 5 years, and the maintenance time limit data is 10 years. The infrared image is a thermal infrared image formed by receiving and recording thermal radiation energy emitted by a target object by a thermal infrared scanner. The rust image is an image shot by a camera.
Further, the embodiment of the invention determines the service life detection value of the power equipment according to the service life data, so that the service life of the equipment is used as a factor for evaluating equipment defects.
In an embodiment of the present invention, referring to fig. 2, the determining a lifetime detection value of the power device according to the lifetime data includes:
s201, defining the service life grade of the power equipment according to the service life data;
s202, setting a grade index of the service life grade, and taking the grade index as the service life detection value.
Illustratively, for example, the life data is 5 years of the service time data, and the maintenance deadline data is 10 years of the service time data; when the service time data is less than or equal to 3 years, the service life data is old equipment, when the service time data is greater than 3 years, the service life data is strong equipment, and when the service life data exceeds the service life, the service life data is severely failed equipment; setting a grade index for each service life grade, for example, setting the grade index of old equipment to be 6 to 8, setting the grade index of strong equipment to be 1 to 3, and setting the grade index of severe failure equipment to be 9 to 10.
S2, extracting image features in the infrared image, identifying a region to be detected and a region type thereof from the infrared image according to the image features, and determining an equipment image in the infrared image by using the region to be detected and the region type thereof.
According to the method and the device, the image features in the infrared image are extracted, so that the candidate area containing the equipment is framed out of the image based on the features.
Optionally, the extracting the image features in the infrared image is implemented by using a residual neural network as a base network. The residual error neural network is a ResNet network and comprises an input layer, a plurality of convolutional layers with different dimensions, and a maximum pooling layer and an output layer are sandwiched in the convolutional layers.
Furthermore, according to the image characteristics, the region to be detected and the region type thereof are identified from the infrared image so as to remove the region irrelevant to the equipment in the image, identify the type of the equipment in the image, detect the equipment by using a fast RCNN detection algorithm, and improve the detection accuracy.
In an embodiment of the present invention, referring to fig. 3, the identifying the region to be detected and the region type thereof from the infrared image according to the image feature includes:
s301, performing window sliding on the image characteristics to obtain an image anchor point;
s302, performing bounding box regression on the image anchor point to obtain a regression anchor point;
s303, carrying out anchor point classification on the image anchor points to obtain classified anchor points;
s304, taking the regression anchor points as the areas to be detected, and taking the classification anchor points as the area types.
Exemplarily, the process of performing window sliding on the image feature to obtain the image anchor point is to perform sliding scanning on the extracted feature map by using a sliding window, and configure anchor points (anchors) with different sizes as an initial candidate frame by using a central point of the sliding window as a window every time sliding is performed.
In an embodiment of the present invention, the device image is an image including a device type, and is selected from the to-be-detected region and a region type thereof.
And S3, calculating the actual temperature of the power equipment according to the equipment image.
According to the embodiment of the invention, the actual temperature of the electric power equipment is calculated according to the equipment image so as to determine whether the temperature of the electric power equipment has defects or not according to the one-to-one correspondence relationship between the gray value and the temperature value.
In an embodiment of the present invention, the calculating the actual temperature of the power device according to the device image includes: calculating an image heat of the device image using the following formula:
wherein,representing the image heat of the device image,a gray value representing an image of the device,indicating the thermal range of the infrared imager device,indicating the thermal level of the infrared imager device,which represents the transmittance of the light emitted from the light source,representing the emissivity of the object;
according to the image heat, calculating the actual temperature of the power equipment by using the following formula:
wherein,is representative of the actual temperature of the electrical equipment,representing the image heat of the device image,representing the calibration curve constant of the infrared imager device.
S4, performing foreground and background segmentation on the corrosion image to obtain a segmentation foreground and a segmentation background, performing brightness transformation on the segmentation foreground based on the segmentation background to obtain a transformation foreground, identifying a corrosion area in the transformation foreground, and calculating a corrosion detection value of the corrosion area.
The rust image is segmented into the foreground and the background so as to detect the brightness of the background, and the subsequent image enhancement is realized based on the brightness of the background.
In an embodiment of the present invention, the foreground and background segmentation is performed on the corrosion image, and the obtained segmentation foreground and segmentation background are achieved by using a Grabcut algorithm.
The Grabcut algorithm is an image segmentation algorithm realized based on graph cut, a bounding box is required to be input by a user as a segmentation target position, and the separation/segmentation of a target and a background is realized, and the realization principle is as follows: defining a rectangle(s) containing an object in the picture, the area outside the rectangle being automatically considered as background; for a user-defined rectangular region, data in the background can be used to distinguish foreground and background regions within it; modeling the background and foreground with a Gaussian Mixture Model (GMM) and labeling undefined pixels as possible foreground or background; each pixel in the image is regarded as being connected with surrounding pixels through a virtual edge, each edge has a probability of belonging to the foreground or the background, based on the similarity of the edge and the surrounding pixels in color, and each pixel (namely a node in an algorithm) is connected with a foreground or background node; after the nodes are connected (possibly with the background or the foreground), if the edges between the nodes belong to different terminals (i.e. one node belongs to the foreground and the other node belongs to the background), the edges between the nodes are cut off, and then the image parts can be segmented.
Furthermore, the embodiment of the present invention performs luminance transformation on the segmentation foreground based on the segmentation background, so as to enhance dark details in an image, reduce the luminance of an image with too high luminance, and ensure the balance of image luminance.
In an embodiment of the present invention, the performing luminance transformation on the segmentation foreground based on the segmentation background to obtain a transformation foreground includes: calculating the average brightness of the segmented background using the following formula:
wherein,representing the average brightness of the segmented background,representing the brightness of the segmented background,a red gray value representing the segmented background,a green gray value representing the segmented background,representing a blue gray value of the segmented background,expressing the pixel coordinates of the segmented background, and k expresses the number of all pixel points with the brightness not being 0;
and calculating a transformation parameter of the segmentation foreground according to the average brightness by using the following formula:
wherein,a transformation parameter representing the segmented foreground,representing the average brightness of the segmented background, and 128 representing an image brightness standard value;
according to the transformation parameters, carrying out brightness transformation on the segmentation foreground by using the following formula to obtain the transformation foreground:
wherein,the foreground of the transformation is represented and,a transformation parameter representing the segmented foreground,representing the segmentation foreground.
Furthermore, the embodiment of the present invention identifies the rusty area in the transformed foreground to highlight the rusty equipment area, and accordingly, the area without obvious red information may interfere with the grayed image, and the dark red feature of the rusty area needs to be effectively distinguished from other areas.
In an embodiment of the present invention, the identifying the rusty area in the transform foreground includes: calculating a rust gray value in the transformed foreground using the following formula:
wherein,representing a rust grey value in the transformed foreground,respectively three channel gray values of the foreground picture,pixel coordinates representing the transformed foreground;
according to the corrosion gray value, the area color in the transformation foreground is marked by using the following formula to obtain a color mark:
wherein r represents the color designation,representing a rust grey value in the transformed foreground,、、、andrespectively representing orange, yellow, pink, red and dark red color marks;
determining a tarnish region in the transformed foreground based on the color identification.
Furthermore, the embodiment of the invention calculates the rust detection value of the rust area to quantify the abstract rust area, thereby facilitating the numerical statistics of the equipment defect detection.
In an embodiment of the present invention, a corrosion detection value of the corrosion region is calculated by using the following formula:
wherein R represents a rust detection value of the rust region,is the area of the image of the rusty area after the binaryzation treatment,the area of the foreground portion after image segmentation.
S5, constructing defect factor weights of the service life detection value, the actual temperature and the corrosion detection value, and determining a defect detection result of the power equipment based on the defect factor weights.
According to the embodiment of the invention, the defect factor weights of the service life detection value, the actual temperature and the corrosion detection value are constructed so as to be used for distinguishing the influence degrees of different defect factors on the detection result by using the weights, and the accuracy of the defect detection result is improved.
In an embodiment of the invention, the constructing the defect factor weights of the life detection value, the actual temperature and the corrosion detection value includes: sequencing defect factors of the service life detection value, the actual temperature and the corrosion detection value to obtain a defect factor sequence; determining a defect factor weight for the rust detection values using the sequence of defect factors.
In an embodiment of the present invention, the determining the defect detection result of the power equipment based on the defect factor weight includes: acquiring the service life detection value, the actual temperature and the corrosion detection value; carrying out weighted summation processing on the service life detection value, the actual temperature and the corrosion detection value by using the defect factor weight to obtain a weighted summation value; determining a defect detection level of the electrical equipment based on the weighted summation value; respectively weighting the service life detection value, the actual temperature and the corrosion detection value by using the defect factor weight to obtain weighted processing numerical values; determining a defect detection category of the power equipment based on the weighted processing numerical value; and taking the defect detection grade and the defect detection category as the defect detection result.
The weighted summation value is a result of weighting the life detection value, the actual temperature and the corrosion detection value respectively and then summing the weighted summation value, and the weighted processing value is a result of weighting a certain influence factor. For example, if the life detection value, the actual temperature, and the corrosion detection value are weighted to obtain values 5, 7, and 9, respectively, and the weighted values of the corrosion detection values are found to be the largest values, the defect detection type of the power equipment is determined as the corrosion type.
It can be seen that, in the embodiment of the present invention, first, life data, an infrared image and a corrosion image of an electrical device are collected to analyze a defect degree of the electrical device by using collected historical data of the electrical device, and further, a life detection value of the electrical device is determined according to the life data to use a service life of the device as a factor for evaluating a defect of the device; secondly, the image features in the infrared image are extracted to frame a candidate area containing the equipment from the image based on the features, and further, the area to be detected and the area type of the area are identified from the infrared image according to the image features to remove the area irrelevant to the equipment from the image and identify the equipment type in the image, and the equipment is detected by using a fast RCNN detection algorithm, so that the detection accuracy is improved; further, according to the device image, the actual temperature of the electrical device is calculated, so that whether the temperature of the electrical device has a defect or not is determined according to the one-to-one correspondence relationship between the gray value and the temperature value; further, the embodiment of the present invention performs foreground-background segmentation on the rusty image to detect the brightness of a background, and implements subsequent image enhancement based on the brightness of the background, and further performs brightness transformation on the segmented foreground based on the segmented background to enhance the dark details in the image, and simultaneously reduces the brightness of the image with excessively high brightness, so as to ensure the balance of the image brightness, and further, identifies the rusty region in the transformed foreground to highlight the rusty device region, and accordingly, since the region without obvious red information interferes with the grayed image, the dark red feature of the rusty region needs to be effectively distinguished from other regions, and further, calculates the rusty detection value of the rusty region to digitize the abstract rusty region, thereby facilitating numerical statistics of device defect detection; further, the defect factor weights of the service life detection value, the actual temperature and the corrosion detection value are constructed so as to distinguish the influence degree of different defect factors on the detection result by using the weights, and the accuracy of the defect detection result is improved. Therefore, the method for detecting the defects of the electrical equipment based on the deep learning provided by the embodiment of the invention can improve the accuracy of detecting the defects of the electrical equipment.
Fig. 4 is a functional block diagram of the power equipment defect detection apparatus based on deep learning according to the present invention.
The apparatus 400 for detecting defects of electrical equipment based on deep learning according to the present invention can be installed in an electronic device. According to the implemented functions, the deep learning based power equipment defect detection apparatus may include a detection value determination module 401, an equipment determination module 402, a temperature calculation module 403, a rust value calculation module 404, and a result determination module 405. The module of the present invention, which may also be referred to as a unit, refers to a series of computer program segments that can be executed by a processor of an electronic device and that can perform a fixed function, and that are stored in a memory of the electronic device.
In the embodiment of the present invention, the functions of the modules/units are as follows:
the detection value determining module 401 is configured to collect life data, an infrared image, and a corrosion image of the power equipment, and determine a life detection value of the power equipment according to the life data;
the device determining module 402 is configured to extract image features in the infrared image, identify a region to be detected and a region type thereof from the infrared image according to the image features, and determine a device image in the infrared image by using the region to be detected and the region type thereof;
the temperature calculation module 403 is configured to calculate an actual temperature of the power device according to the device image;
the rust value calculation module 404 is configured to perform foreground-background segmentation on the rust image to obtain a segmented foreground and a segmented background, perform luminance transformation on the segmented foreground based on the segmented background to obtain a transformed foreground, identify a rust region in the transformed foreground, and calculate a rust detection value of the rust region;
the result determining module 405 is configured to construct defect factor weights of the life detection value, the actual temperature and the corrosion detection value, and determine a defect detection result of the power equipment based on the defect factor weights.
In detail, when the modules in the apparatus 400 for detecting defects of electrical equipment based on deep learning according to the embodiment of the present invention are used, the same technical means as the method for detecting defects of electrical equipment based on deep learning described in fig. 1 to 3 are adopted, and the same technical effects can be produced, which is not described herein again.
Fig. 5 is a schematic structural diagram of an electronic device implementing the method for detecting defects of an electrical device based on deep learning according to the present invention.
The electronic device may include a processor 50, a memory 51, a communication bus 52, and a communication interface 53, and may further include a computer program, such as a deep learning-based power device fault detection program, stored in the memory 51 and executable on the processor 50.
In some embodiments, the processor 50 may be composed of an integrated circuit, for example, a single packaged integrated circuit, or may be composed of a plurality of integrated circuits packaged with the same function or different functions, and includes one or more Central Processing Units (CPUs), a microprocessor, a digital Processing chip, a graphics processor, a combination of various control chips, and the like. The processor 50 is a Control Unit (Control Unit) of the electronic device, connects various components of the whole electronic device by using various interfaces and lines, and executes various functions and processes data of the electronic device by running or executing programs or modules stored in the memory 51 (for example, executing a deep learning-based power device defect detection program, etc.), and calling data stored in the memory 51.
The memory 51 includes at least one type of readable storage medium including flash memory, removable hard disks, multimedia cards, card-type memory (e.g., SD or DX memory, etc.), magnetic memory, magnetic disks, optical disks, and the like. The memory 51 may in some embodiments be an internal storage unit of the electronic device, e.g. a removable hard disk of the electronic device. The memory 51 may also be an external storage device of the electronic device in other embodiments, such as a plug-in mobile hard disk, a Smart Media Card (SMC), a Secure Digital (SD) Card, a Flash memory Card (Flash Card), and the like provided on the electronic device. Further, the memory 51 may also include both an internal storage unit and an external storage device of the electronic device. The memory 51 may be used to store not only application software installed in the electronic device and various types of data, such as codes of a database configuration connection program, but also temporarily store data that has been output or will be output.
The communication bus 52 may be a Peripheral Component Interconnect (PCI) bus or an Extended Industry Standard Architecture (EISA) bus. The bus may be divided into an address bus, a data bus, a control bus, etc. The bus is arranged to enable connection communication between the memory 51 and at least one processor 50 or the like.
The communication interface 53 is used for communication between the electronic device 5 and other devices, and includes a network interface and a user interface. Optionally, the network interface may include a wired interface and/or a wireless interface (e.g., WI-FI interface, bluetooth interface, etc.), which are typically used to establish a communication connection between the electronic device and other electronic devices. The user interface may be a Display (Display), an input unit such as a Keyboard (Keyboard), and optionally a standard wired interface, a wireless interface. Alternatively, in some embodiments, the display may be an LED display, a liquid crystal display, a touch-sensitive liquid crystal display, an OLED (Organic Light-Emitting Diode) touch device, or the like. The display, which may also be referred to as a display screen or display unit, is suitable, among other things, for displaying information processed in the electronic device and for displaying a visualized user interface.
Fig. 5 shows only an electronic device with components, and those skilled in the art will appreciate that the structure shown in fig. 5 does not constitute a limitation of the electronic device, and may include fewer or more components than shown, or some components may be combined, or a different arrangement of components.
For example, although not shown, the electronic device may further include a power supply (such as a battery) for supplying power to each component, and preferably, the power supply may be logically connected to the at least one processor 50 through a power management device, so that functions of charge management, discharge management, power consumption management and the like are realized through the power management device. The power supply may also include any component of one or more dc or ac power sources, recharging devices, power failure detection circuitry, power converters or inverters, power status indicators, and the like. The electronic device may further include various sensors, a bluetooth module, a Wi-Fi module, and the like, which are not described herein again.
It is to be understood that the embodiments described are for illustrative purposes only and that the scope of the claimed invention is not limited to this configuration.
The database configuration connection program stored in the memory 51 of the electronic device is a combination of computer programs, and when running in the processor 50, can realize:
collecting service life data, an infrared image and a corrosion image of the power equipment, and determining a service life detection value of the power equipment according to the service life data;
extracting image features in the infrared image, identifying a region to be detected and a region type thereof from the infrared image according to the image features, and determining an equipment image in the infrared image by using the region to be detected and the region type thereof;
calculating the actual temperature of the power equipment according to the equipment image;
performing foreground and background segmentation on the corrosion image to obtain a segmentation foreground and a segmentation background, performing brightness transformation on the segmentation foreground based on the segmentation background to obtain a transformation foreground, identifying a corrosion region in the transformation foreground, and calculating a corrosion detection value of the corrosion region;
and constructing defect factor weights of the service life detection value, the actual temperature and the corrosion detection value, and determining a defect detection result of the power equipment based on the defect factor weights.
Specifically, the processor 50 may refer to the description of the relevant steps in the embodiment corresponding to fig. 1 for a specific implementation method of the computer program, which is not described herein again.
Further, the electronic device integrated module/unit, if implemented in the form of a software functional unit and sold or used as a separate product, may be stored in a non-volatile computer-readable storage medium. The storage medium may be volatile or nonvolatile. For example, the computer-readable medium may include: any entity or device capable of carrying said computer program code, recording medium, U-disk, removable hard disk, magnetic disk, optical disk, computer Memory, read-Only Memory (ROM).
The present invention also provides a storage medium storing a computer program which, when executed by a processor of an electronic device, enables:
collecting service life data, an infrared image and a corrosion image of the power equipment, and determining a service life detection value of the power equipment according to the service life data;
extracting image features in the infrared image, identifying a region to be detected and a region type thereof from the infrared image according to the image features, and determining an equipment image in the infrared image by using the region to be detected and the region type thereof;
calculating the actual temperature of the power equipment according to the equipment image;
performing foreground-background segmentation on the corrosion image to obtain a segmentation foreground and a segmentation background, performing brightness transformation on the segmentation foreground based on the segmentation background to obtain a transformation foreground, identifying a corrosion region in the transformation foreground, and calculating a corrosion detection value of the corrosion region;
and constructing defect factor weights of the service life detection value, the actual temperature and the corrosion detection value, and determining a defect detection result of the power equipment based on the defect factor weights.
In the several embodiments provided in the present invention, it should be understood that the disclosed apparatus, device and method may be implemented in other manners. For example, the above-described apparatus embodiments are merely illustrative, and for example, the division of the modules is only one logical functional division, and other divisions may be realized in practice.
The modules described as separate parts may or may not be physically separate, and parts displayed as modules may or may not be physical units, may be located in one place, or may be distributed on a plurality of network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the solution of this embodiment.
In addition, functional modules in the embodiments of the present invention may be integrated into one processing unit, or each unit may exist alone physically, or two or more units are integrated into one unit. The integrated unit can be realized in a form of hardware, or in a form of hardware plus a software functional module.
It will be evident to those skilled in the art that the invention is not limited to the details of the foregoing illustrative embodiments, and that the present invention may be embodied in other specific forms without departing from the spirit or essential attributes thereof.
The present embodiments are therefore to be considered in all respects as illustrative and not restrictive, the scope of the invention being indicated by the appended claims rather than by the foregoing description, and all changes which come within the meaning and range of equivalency of the claims are therefore intended to be embraced therein. Any reference signs in the claims shall not be construed as limiting the claim concerned.
It is noted that, in this document, relational terms such as "first" and "second," and the like, are used solely to distinguish one entity or action from another entity or action without necessarily requiring or implying any actual such relationship or order between such entities or actions. Also, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without further limitation, an element defined by the phrases "comprising a," "8230," "8230," or "comprising" does not exclude the presence of additional like elements in a process, method, article, or apparatus that comprises the element.
The above description is merely illustrative of particular embodiments of the invention that enable those skilled in the art to understand or practice the invention. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the generic principles defined herein may be applied to other embodiments without departing from the spirit or scope of the invention. Thus, the present invention is not intended to be limited to the embodiments shown herein but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.
Claims (10)
1. A method for detecting defects of electric power equipment based on deep learning is characterized by comprising the following steps:
collecting service life data, an infrared image and a corrosion image of the power equipment, and determining a service life detection value of the power equipment according to the service life data;
extracting image features in the infrared image, identifying a region to be detected and a region type thereof from the infrared image according to the image features, and determining an equipment image in the infrared image by using the region to be detected and the region type thereof;
calculating the actual temperature of the power equipment according to the equipment image;
performing foreground-background segmentation on the corrosion image to obtain a segmentation foreground and a segmentation background, performing brightness transformation on the segmentation foreground based on the segmentation background to obtain a transformation foreground, identifying a corrosion region in the transformation foreground, and calculating a corrosion detection value of the corrosion region;
and constructing defect factor weights of the service life detection value, the actual temperature and the corrosion detection value, and determining a defect detection result of the power equipment based on the defect factor weights.
2. The method of claim 1, wherein determining a life detection value for the power device from the life data comprises:
according to the service life data, the service life grade of the power equipment is determined;
and setting a grade index of the service life grade, and taking the grade index as the service life detection value.
3. The method according to claim 1, wherein the identifying the region to be detected and the region type thereof from the infrared image according to the image features comprises:
carrying out window sliding on the image characteristics to obtain an image anchor point;
performing bounding box regression on the image anchor points to obtain regression anchor points;
carrying out anchor point classification on the image anchor points to obtain classified anchor points;
and taking the regression anchor point as the area to be detected, and taking the classification anchor point as the area type.
4. The method of claim 1, wherein calculating the actual temperature of the electrical device from the device image comprises:
calculating an image heat of the device image using the following formula:
wherein,represents the aboveThe image heat of the device image is,a gray value representing an image of the device,indicating the thermal range of the infrared imager device,indicating the thermal level of the infrared imager device,which is indicative of the degree of transmission of light,representing the emissivity of the object;
according to the image heat, calculating the actual temperature of the power equipment by using the following formula:
5. The method of claim 1, wherein the luminance transformation of the segmented foreground based on the segmented background to obtain a transformed foreground comprises:
calculating the average brightness of the segmented background using the following formula:
wherein,representing the average brightness of the segmented background,representing the brightness of the segmented background,a red gray value representing the segmented background,a green gray value representing the segmented background,representing a blue gray value of the segmented background,expressing the pixel coordinates of the segmented background, and k expresses the number of all pixel points with the brightness not being 0;
and calculating a transformation parameter of the segmentation foreground according to the average brightness by using the following formula:
wherein,a transformation parameter representing the segmented foreground,representing the average brightness of the segmented background, and 128 representing an image brightness standard value;
according to the transformation parameters, carrying out brightness transformation on the segmentation foreground by using the following formula to obtain the transformation foreground:
6. The method of claim 1, wherein the identifying of the rusty areas in the transformed foreground comprises: calculating a rust gray value in the transformed foreground using the following formula:
wherein,representing a rust gray value in the transformed foreground,respectively three channels of gray values of the foreground picture,pixel coordinates representing the transformed foreground;
according to the corrosion gray value, the area color in the transformation foreground is marked by using the following formula to obtain a color mark:
wherein r represents the color designation,representing a rust grey value in the transformed foreground,、、、andrespectively representing orange, yellow, pink, red and dark red color marks;
determining a tarnish region in the transformed foreground based on the color identification.
7. The method of claim 1, wherein determining the fault detection result of the power equipment based on the fault factor weight comprises:
acquiring the service life detection value, the actual temperature and the corrosion detection value;
carrying out weighted summation processing on the service life detection value, the actual temperature and the corrosion detection value by using the defect factor weight to obtain a weighted summation value;
determining a defect detection level of the power equipment based on the weighted summation value;
respectively weighting the service life detection value, the actual temperature and the corrosion detection value by using the defect factor weight to obtain weighted processing numerical values;
determining a defect detection category of the power equipment based on the weighted processing numerical value;
and taking the defect detection grade and the defect detection category as the defect detection result.
8. A method and a device for detecting defects of electric power equipment based on deep learning are characterized by comprising the following steps:
the detection value determining module is used for acquiring the service life data, the infrared image and the corrosion image of the power equipment and determining the service life detection value of the power equipment according to the service life data;
the device determining module is used for extracting image features in the infrared image, identifying a region to be detected and a region type thereof from the infrared image according to the image features, and determining a device image in the infrared image by using the region to be detected and the region type thereof;
the temperature calculation module is used for calculating the actual temperature of the power equipment according to the equipment image;
the rust value calculation module is used for carrying out foreground background segmentation on the rust image to obtain a segmentation foreground and a segmentation background, carrying out brightness transformation on the segmentation foreground based on the segmentation background to obtain a transformation foreground, identifying a rust region in the transformation foreground, and calculating a rust detection value of the rust region;
and the result determining module is used for constructing defect factor weights of the service life detection value, the actual temperature and the corrosion detection value and determining a defect detection result of the power equipment based on the defect factor weights.
9. An electronic device, characterized in that the electronic device comprises:
at least one processor; and the number of the first and second groups,
a memory communicatively coupled to the at least one processor; wherein,
the memory stores a computer program executable by the at least one processor to enable the at least one processor to perform the method of deep learning based power equipment fault detection as claimed in any one of claims 1 to 7.
10. A computer-readable storage medium storing a computer program, wherein the computer program, when executed by a processor, implements the deep learning-based power equipment fault detection method according to any one of claims 1 to 7.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202310026789.7A CN115760854A (en) | 2023-01-09 | 2023-01-09 | Deep learning-based power equipment defect detection method and device and electronic equipment |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202310026789.7A CN115760854A (en) | 2023-01-09 | 2023-01-09 | Deep learning-based power equipment defect detection method and device and electronic equipment |
Publications (1)
Publication Number | Publication Date |
---|---|
CN115760854A true CN115760854A (en) | 2023-03-07 |
Family
ID=85348757
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202310026789.7A Pending CN115760854A (en) | 2023-01-09 | 2023-01-09 | Deep learning-based power equipment defect detection method and device and electronic equipment |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN115760854A (en) |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN117710366A (en) * | 2024-02-02 | 2024-03-15 | 杭州百子尖科技股份有限公司 | Quality inspection method and device for thermos cup and storage medium |
Citations (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN108492292A (en) * | 2018-03-20 | 2018-09-04 | 西安工程大学 | Conducting wire broken lot detection method based on infrared image processing |
US20190171787A1 (en) * | 2017-12-01 | 2019-06-06 | Applied Materials, Inc. | Using design proximity index and defect-to-design proximity ratio to control semiconductor processes and achieve enhanced yield |
CN110726725A (en) * | 2019-10-23 | 2020-01-24 | 许昌许继软件技术有限公司 | Transmission line hardware corrosion detection method and device |
CN111563693A (en) * | 2020-05-20 | 2020-08-21 | 深圳达实智能股份有限公司 | Method and device for scoring health value of rail transit equipment and storage medium |
CN111640112A (en) * | 2020-06-11 | 2020-09-08 | 云从科技集团股份有限公司 | Image detection method, system, platform, device, medium, and image processing apparatus |
CN114792319A (en) * | 2022-06-23 | 2022-07-26 | 国网浙江省电力有限公司电力科学研究院 | Transformer substation inspection method and system based on transformer substation image |
CN115294025A (en) * | 2022-07-05 | 2022-11-04 | 深圳市商汤科技有限公司 | Defect detection method and device, electronic equipment and storage medium |
-
2023
- 2023-01-09 CN CN202310026789.7A patent/CN115760854A/en active Pending
Patent Citations (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20190171787A1 (en) * | 2017-12-01 | 2019-06-06 | Applied Materials, Inc. | Using design proximity index and defect-to-design proximity ratio to control semiconductor processes and achieve enhanced yield |
CN108492292A (en) * | 2018-03-20 | 2018-09-04 | 西安工程大学 | Conducting wire broken lot detection method based on infrared image processing |
CN110726725A (en) * | 2019-10-23 | 2020-01-24 | 许昌许继软件技术有限公司 | Transmission line hardware corrosion detection method and device |
CN111563693A (en) * | 2020-05-20 | 2020-08-21 | 深圳达实智能股份有限公司 | Method and device for scoring health value of rail transit equipment and storage medium |
CN111640112A (en) * | 2020-06-11 | 2020-09-08 | 云从科技集团股份有限公司 | Image detection method, system, platform, device, medium, and image processing apparatus |
CN114792319A (en) * | 2022-06-23 | 2022-07-26 | 国网浙江省电力有限公司电力科学研究院 | Transformer substation inspection method and system based on transformer substation image |
CN115294025A (en) * | 2022-07-05 | 2022-11-04 | 深圳市商汤科技有限公司 | Defect detection method and device, electronic equipment and storage medium |
Non-Patent Citations (1)
Title |
---|
涂铭等: "《基于深度神经网络技术的高分遥感图像处理及应用》", pages: 140 - 142 * |
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN117710366A (en) * | 2024-02-02 | 2024-03-15 | 杭州百子尖科技股份有限公司 | Quality inspection method and device for thermos cup and storage medium |
CN117710366B (en) * | 2024-02-02 | 2024-05-14 | 杭州百子尖科技股份有限公司 | Quality inspection method and device for thermos cup and storage medium |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN110060237B (en) | Fault detection method, device, equipment and system | |
CN113781485B (en) | Intelligent detection method and device for PCB defect types, electronic equipment and medium | |
CN112137591B (en) | Target object position detection method, device, equipment and medium based on video stream | |
CN115131283B (en) | Defect detection and model training method, device, equipment and medium for target object | |
CN112507923B (en) | Certificate copying detection method and device, electronic equipment and medium | |
CN113610934B (en) | Image brightness adjustment method, device, equipment and storage medium | |
CN111695609A (en) | Target damage degree determination method, target damage degree determination device, electronic device, and storage medium | |
CN112507934A (en) | Living body detection method, living body detection device, electronic apparatus, and storage medium | |
CN114758249B (en) | Target object monitoring method, device, equipment and medium based on field night environment | |
US20170178341A1 (en) | Single Parameter Segmentation of Images | |
CN112528908B (en) | Living body detection method, living body detection device, electronic equipment and storage medium | |
CN112749653A (en) | Pedestrian detection method, device, electronic equipment and storage medium | |
CN114627435B (en) | Intelligent light adjusting method, device, equipment and medium based on image recognition | |
CN116168351A (en) | Inspection method and device for power equipment | |
CN114267064A (en) | Face recognition method and device, electronic equipment and storage medium | |
CN113554008A (en) | Method and device for detecting static object in area, electronic equipment and storage medium | |
CN115294483A (en) | Small target identification method and system for complex scene of power transmission line | |
CN115760854A (en) | Deep learning-based power equipment defect detection method and device and electronic equipment | |
CN114240924A (en) | Power grid equipment quality evaluation method based on digitization technology | |
CN111931721B (en) | Method and device for detecting color and number of annual inspection label and electronic equipment | |
CN112017065A (en) | Vehicle loss assessment and claim settlement method and device and computer readable storage medium | |
CN115909467A (en) | Human face living body detection method, device, equipment and medium in motion state scene | |
CN111402185A (en) | Image detection method and device | |
CN115205596A (en) | Image classification method, device, equipment and storage medium | |
CN113792801A (en) | Method, device and equipment for detecting dazzling degree of human face and storage medium |
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 | ||
RJ01 | Rejection of invention patent application after publication | ||
RJ01 | Rejection of invention patent application after publication |
Application publication date: 20230307 |