CN115272284A - Power transmission line defect identification method based on image quality evaluation - Google Patents
Power transmission line defect identification method based on image quality evaluation Download PDFInfo
- Publication number
- CN115272284A CN115272284A CN202210988364.XA CN202210988364A CN115272284A CN 115272284 A CN115272284 A CN 115272284A CN 202210988364 A CN202210988364 A CN 202210988364A CN 115272284 A CN115272284 A CN 115272284A
- Authority
- CN
- China
- Prior art keywords
- transmission line
- image
- power transmission
- image quality
- processing
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/0002—Inspection of images, e.g. flaw detection
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T5/00—Image enhancement or restoration
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/13—Edge detection
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/136—Segmentation; Edge detection involving thresholding
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/20—Image preprocessing
- G06V10/22—Image preprocessing by selection of a specific region containing or referencing a pattern; Locating or processing of specific regions to guide the detection or recognition
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
- G06V10/764—Arrangements for image or video recognition or understanding using pattern recognition or machine learning using classification, e.g. of video objects
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
- G06V10/82—Arrangements for image or video recognition or understanding using pattern recognition or machine learning using neural networks
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Evolutionary Computation (AREA)
- Health & Medical Sciences (AREA)
- General Health & Medical Sciences (AREA)
- Software Systems (AREA)
- Computing Systems (AREA)
- Artificial Intelligence (AREA)
- Multimedia (AREA)
- Medical Informatics (AREA)
- Databases & Information Systems (AREA)
- Quality & Reliability (AREA)
- Life Sciences & Earth Sciences (AREA)
- Biomedical Technology (AREA)
- Biophysics (AREA)
- Computational Linguistics (AREA)
- Data Mining & Analysis (AREA)
- Molecular Biology (AREA)
- General Engineering & Computer Science (AREA)
- Mathematical Physics (AREA)
- Image Analysis (AREA)
Abstract
The invention discloses a power transmission line defect identification method based on image quality evaluation, which is applied to a power transmission line defect identification device and comprises the following steps: acquiring a power transmission line inspection image; extracting and processing the power transmission line inspection image through a candidate area generation network to obtain a plurality of candidate area images; performing image enhancement processing and image quality evaluation processing on each candidate region image to obtain a plurality of to-be-identified power transmission line images with image quality meeting preset image quality indexes; carrying out feature extraction processing on the plurality of power transmission line images to be identified according to a preset convolutional neural network to obtain convolutional neural network features of the plurality of power transmission line images to be identified; and carrying out classification and identification processing on the convolutional neural network characteristics according to the classifier, and determining defect type information of a plurality of images of the power transmission line to be identified. According to the scheme of the embodiment of the application, the success rate and the accuracy of defect identification can be improved by improving the image quality of the image of the power transmission line to be identified.
Description
Technical Field
The present disclosure relates to the field of defect identification, and more particularly, to a method, an apparatus, a device, and a storage medium for identifying defects of a power transmission line based on image quality evaluation.
Background
With the continuous development of the social living standard, the rising of the power demand causes the power transmission line to tend to be complicated. The transmission line is in complex and various environments, and is influenced by various factors in the actual working and running process, so that accidents such as short circuit, short circuit and damage of transmission equipment occur. These fault defects all contribute to the operation of the transmission line. These factors present significant difficulties to the repair and maintenance of the transmission line.
In the correlation technique, along with the intelligent level of electric wire netting constantly improves, use unmanned aerial vehicle to patrol and examine a large amount of pictures of patrolling and examining of collection. For collected inspection pictures, the calibration of target components and the classification of defects are mostly completed based on a manual interpretation mode, and the manual interpretation mode has high labor intensity and low working efficiency, so that the situations of untimely defect inspection of the power transmission line and poor defect inspection effect often occur.
Disclosure of Invention
The present application is directed to solving, at least in part, one of the technical problems in the related art. Therefore, the application provides a method, a device, equipment and a storage medium for identifying the defects of the power transmission line based on image quality evaluation, and the success rate and the accuracy of defect identification can be improved by improving the image quality of the images of the power transmission line to be identified.
In a first aspect, an embodiment of the present application provides a method for identifying defects of a power transmission line based on image quality evaluation, which is applied to a device for identifying defects of a power transmission line, and includes:
acquiring a power transmission line inspection image; extracting and processing the power transmission line inspection image through a candidate area generation network to obtain a plurality of candidate area images; performing image enhancement processing and image quality evaluation processing on each candidate region image to obtain a plurality of to-be-identified power transmission line images with image quality meeting preset image quality indexes; performing feature extraction processing on the plurality of power transmission line images to be identified according to a preset convolutional neural network to obtain convolutional neural network features of the plurality of power transmission line images to be identified; and carrying out classification and identification processing on the convolutional neural network characteristics according to a classifier, and determining defect type information of a plurality of images of the power transmission line to be identified.
The power transmission line defect identification method based on image quality evaluation of the embodiment of the application at least has the following beneficial effects: acquiring a power transmission line inspection image; extracting and processing the power transmission line inspection image through a candidate area generation network to obtain a plurality of candidate area images; performing image enhancement processing and image quality evaluation processing on each candidate region image to obtain a plurality of to-be-identified power transmission line images with image quality meeting preset image quality indexes; performing feature extraction processing on the plurality of electric transmission line images to be identified according to a preset convolutional neural network to obtain convolutional neural network features of the plurality of electric transmission line images to be identified; and carrying out classification and identification processing on the convolutional neural network characteristics according to a classifier, and determining defect type information of a plurality of images of the power transmission line to be identified. According to the scheme of the embodiment of the application, the power transmission line inspection image is obtained; extracting and processing the power transmission line inspection image through a candidate area generation network to obtain a plurality of candidate area images; performing image enhancement processing and image quality evaluation processing on each candidate region image to obtain a plurality of to-be-identified power transmission line images with image quality meeting preset image quality indexes; performing feature extraction processing on the multiple power transmission line images to be identified according to a preset convolutional neural network to obtain convolutional neural network features of the multiple power transmission line images to be identified; classifying and identifying the features of the convolutional neural network according to a classifier, and determining the defect type information of a plurality of images of the power transmission line to be identified, namely: according to the scheme, the power transmission line defect recognition device can improve the image quality of the image of the power transmission line to be recognized, so that the success rate and the accuracy of defect recognition are improved, the efficiency of defect troubleshooting of the power transmission line in the acceptance phase of completion engineering is improved, and a reliable reference is provided for subsequent overhaul of engineering personnel.
Optionally, in an embodiment of the present application, the performing image enhancement processing and image quality evaluation processing on the power transmission line inspection image to obtain a to-be-identified power transmission line image with image quality meeting a preset image quality index includes: performing the image enhancement processing on the power transmission line inspection image to obtain a first power transmission line enhanced image; performing the image quality evaluation processing on the first power transmission line enhanced image to generate first image quality evaluation information, wherein the first image quality evaluation information represents the image quality of the power transmission line enhanced image; and determining the first power transmission line enhanced image as a power transmission line image to be identified under the condition that the first image quality evaluation information meets a preset image quality index.
Optionally, in an embodiment of the present application, the performing the image quality evaluation processing on the first power transmission line enhanced image to generate first image quality evaluation information further includes: performing quality feature extraction processing on the first power transmission line enhanced image to determine a plurality of quality features; performing single quality evaluation processing on the first power transmission line enhanced image aiming at each quality feature to obtain a plurality of single quality feature scores; carrying out scoring weighting processing according to a preset weighting coefficient and the plurality of single quality feature scores to obtain a first image quality evaluation total score; generating the first image quality assessment information according to the plurality of individual quality feature scores and the first image quality assessment total score.
Optionally, in an embodiment of the application, the performing image enhancement processing and image quality evaluation processing on the power transmission line inspection image to obtain a to-be-identified power transmission line image whose image quality meets a preset image quality index further includes: under the condition that the first image quality evaluation information does not meet the preset image quality index, determining substandard quality characteristic information which does not meet the preset image quality index according to the first image quality evaluation information; performing image feature enhancement processing on the first transmission line enhanced image according to the substandard quality feature information to obtain a second transmission line enhanced image; performing image quality evaluation processing on the second transmission line enhanced image to obtain second image quality evaluation information; and determining the second power transmission line enhanced image as the power transmission line image to be identified under the condition that the second image quality evaluation information meets a preset image quality index.
Optionally, in an embodiment of the present application, the image enhancement processing includes at least one of: defogging processing, filtering processing and sharpening processing.
Optionally, in an embodiment of the present application, the performing defect identification processing on the image of the power transmission line to be identified to obtain defect information includes: extracting the images of the power transmission line to be identified to obtain a plurality of candidate areas; performing feature extraction processing on the candidate areas according to a preset convolutional neural network to obtain convolutional neural network features of the candidate areas; classifying, identifying and processing the convolutional neural network characteristics according to a classifier, and determining defect type information of the candidate regions; performing frame regression correction processing on the candidate region with the determined defect type information to determine defect position information;
and obtaining the defect information according to the defect type information and the defect position information.
Optionally, in an embodiment of the application, the extracting the to-be-identified power transmission line image to obtain a plurality of candidate regions further includes: and carrying out size scaling processing on the candidate areas, and setting the size scaling of the candidate areas to be a preset size.
In a second aspect, an embodiment of the present application provides an apparatus for identifying defects of a power transmission line, including: the image acquisition module is used for acquiring the inspection image of the power transmission line; the image processing module is used for extracting and processing the power transmission line inspection image through the candidate area generation network to obtain a plurality of candidate area images; performing image enhancement processing and image quality evaluation processing on each candidate region image to obtain a plurality of to-be-identified power transmission line images with image quality meeting preset image quality indexes; the defect identification module is used for extracting and processing the characteristics of the plurality of electric transmission line images to be identified according to a preset convolutional neural network to obtain convolutional neural network characteristics of the plurality of electric transmission line images to be identified; and carrying out classification and identification processing on the convolutional neural network characteristics according to a classifier, and determining defect type information of a plurality of electric transmission line images to be identified.
The power transmission line defect identification device of the embodiment of the application at least has the following beneficial effects: the method comprises the steps that a transmission line defect recognition device obtains a transmission line inspection image; carrying out image enhancement processing and image quality evaluation processing on the power transmission line inspection image to obtain a power transmission line image to be identified, wherein the image quality of the power transmission line image meets a preset image quality index; and carrying out defect identification processing on the image of the power transmission line to be identified to obtain defect information. According to the scheme of the embodiment of the application, the power transmission line defect identification device acquires a power transmission line inspection image, then performs image enhancement processing and image quality evaluation processing on the power transmission line inspection image to obtain a to-be-identified power transmission line image with the image quality meeting a preset image quality index, and finally performs defect identification processing on the to-be-identified power transmission line image to obtain defect information, namely: according to the scheme of the embodiment of the application, the power transmission line defect identification device can improve the success rate and accuracy of defect identification by improving the image quality of the image of the power transmission line to be identified, is favorable for improving the efficiency of defect troubleshooting of the power transmission line in the engineering completion acceptance stage, and provides reliable reference for subsequent overhaul of engineers.
In a third aspect, an embodiment of the present application provides a device for identifying defects of a power transmission line, including: a memory, a processor and a computer program stored on the memory and executable on the processor, the processor implementing the defect identification method according to the first aspect when executing the computer program.
In a fourth aspect, the present application provides a computer-readable storage medium storing computer-executable instructions for causing a computer to perform the defect identification method according to the first aspect.
Drawings
The accompanying drawings are included to provide a further understanding of the invention, and are incorporated in and constitute a part of this specification, illustrate embodiments of the invention and together with the example serve to explain the principles of the invention.
Fig. 1 is a schematic block diagram of a defect identification apparatus for a power transmission line according to an embodiment of the present application;
fig. 2 is a schematic flowchart of a method for identifying defects of a power transmission line based on image quality evaluation according to an embodiment of the present application;
FIG. 3 is a flow chart illustrating a specific method of step S230 in FIG. 2;
FIG. 4 is a schematic flow chart of another specific method of step S230 in FIG. 2;
FIG. 5 is a flowchart illustrating a defect identification method according to another embodiment of the present invention;
fig. 6 is a schematic diagram of a power transmission line defect identification device based on image quality evaluation according to an embodiment of the present application.
Detailed Description
In order to make the objects, technical solutions and advantages of the present application more clearly understood, the present application is further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the present application and are not intended to limit the present application.
It should be noted that although functional block divisions are provided in the system drawings and logical orders are shown in the flowcharts, in some cases, the steps shown and described may be performed in different orders than the block divisions in the systems or in the flowcharts. The terms first, second and the like in the description and in the claims, and the drawings described above, are used for distinguishing between similar elements and not necessarily for describing a particular sequential or chronological order.
With the continuous development of the social living standard, the rising of the power demand prompts the power transmission line to be complicated. The transmission line is in complex and various environments, and is influenced by various factors in the actual working and running process, so that accidents such as short circuit, short circuit and damage of transmission equipment occur. These fault defects all have a serious impact on the operation of the transmission line. These factors present great difficulties to the repair and maintenance of the transmission line.
In the correlation technique, along with the intelligent level of electric wire netting constantly improves, use unmanned aerial vehicle to patrol and examine a large amount of pictures of patrolling and examining of collection. For the collected inspection pictures, the calibration of the target component and the classification of the defects are mostly completed based on a manual interpretation mode, and the manual interpretation mode has the disadvantages of high labor intensity, low working efficiency, untimely defect inspection of the power transmission line and poor defect inspection effect.
Based on the above, the application provides a power transmission line defect identification method and device based on image quality evaluation, a power transmission line defect identification device and a computer readable storage medium, in the stage of completion of engineering acceptance, the power transmission line defect identification device obtains a power transmission line inspection image, then performs image enhancement processing and image quality evaluation processing on the power transmission line inspection image to obtain a power transmission line image to be identified, wherein the image quality of the power transmission line image to be identified meets the preset image quality index, and finally performs defect identification processing on the power transmission line image to be identified to obtain defect information, namely: according to the scheme of the embodiment of the application, the power transmission line defect identification device can improve the success rate and accuracy of defect identification by improving the image quality of the image of the power transmission line to be identified, is favorable for improving the efficiency of defect troubleshooting of the power transmission line in the engineering completion acceptance stage, and provides reliable reference for subsequent overhaul of engineers.
The embodiments of the present application will be further explained with reference to the drawings.
Referring to fig. 1, fig. 1 is a schematic block diagram of a power transmission line defect identification apparatus according to an embodiment of the present application. This transmission line defect recognition device includes: the image processing system comprises an image acquisition module 110, an image processing module 120 and a defect identification module 130, wherein the image processing module 120 is respectively connected with the image acquisition module 110 and the defect identification module 130 in a communication way.
The image acquisition module 110 is configured to acquire an inspection image of the power transmission line; specifically, the image obtaining module 110 can be in communication connection with an external image capturing device, obtain the power transmission line inspection image sent by the external image capturing device, and send the power transmission line inspection image to the image processing module 120.
The image processing module 120 is configured to extract and process the power transmission line inspection image through the candidate area generation network to obtain a plurality of candidate area images; performing image enhancement processing and image quality evaluation processing on each candidate region image to obtain a plurality of to-be-identified power transmission line images with image quality meeting preset image quality indexes; specifically, the image processing module 120 receives the power transmission line inspection image sent by the image obtaining module 110, performs image enhancement processing and image quality evaluation processing on the power transmission line inspection image, and sends the obtained power transmission line image to be identified, of which the image quality meets a preset image quality index, to the defect identification module, so that the image quality of the power transmission line image to be identified can be improved, and the success rate and accuracy of subsequent defect identification are improved.
And the defect identification module 130 is configured to perform defect identification processing on the image of the power transmission line to be identified to obtain defect information. Specifically, the defect identification module 130 performs feature extraction processing on the plurality of to-be-identified power transmission line images according to a preset convolutional neural network to obtain convolutional neural network features of the plurality of to-be-identified power transmission line images; and carrying out classification and identification processing on the convolutional neural network characteristics according to the classifier, and determining defect type information of a plurality of images of the power transmission line to be identified. The defect identification module 130 receives the image of the transmission line to be identified sent by the image processing module 120, performs defect identification processing on the image of the transmission line to be identified, and can perform defect identification processing based on the image of the transmission line to be identified with better image quality, so that the success rate and accuracy of defect identification are improved, the efficiency of transmission line defect investigation in the acceptance stage of engineering completion is improved, and reliable reference is provided for subsequent overhaul of engineering personnel.
According to the transmission line defect identification device provided by the embodiment of the application, in the process of completion acceptance of engineering, an engineer can effectively eliminate the defects of the transmission line by using the transmission line defect identification device, the transmission line defect identification device can obtain the transmission line image to be identified with better image quality based on the obtained transmission line inspection image, and finally, the transmission line image to be identified is subjected to defect identification processing to obtain defect information.
The device function module schematic and the application scenario described in the embodiment of the present application are for more clearly illustrating the technical solution of the embodiment of the present application, and do not constitute a limitation to the technical solution provided in the embodiment of the present application, and it is known by those skilled in the art that, along with the evolution of the device function module and the occurrence of a new application scenario, the technical solution provided in the embodiment of the present application is also applicable to similar technical problems. Those skilled in the art will appreciate that the system functional blocks illustrated in fig. 1 are not limiting of the embodiments of the present application and may include more or fewer blocks than those illustrated, or some blocks may be combined, or a different arrangement of blocks may be combined.
Referring to fig. 2, fig. 2 is a schematic flowchart of a power transmission line defect identification method based on image quality evaluation according to an embodiment of the present application; the power transmission line defect identification method based on image quality evaluation is applied to the power transmission line defect identification device shown in fig. 1, and the power transmission line defect identification method includes, but is not limited to, steps S210 to S250.
Step S210: and acquiring a power transmission line inspection image.
In this step, the transmission line defect recognition device acquires a transmission line inspection image. Specifically, the unmanned aerial vehicle patrols the transmission line, acquires a transmission line patrol image, and sends the transmission line patrol image to the transmission line defect identification device, so that the transmission line defect identification device acquires the transmission line patrol image. It can be understood that the power transmission line inspection image can be acquired through the intelligent robot line inspection, the power transmission line inspection image can be shot manually, and the power transmission line inspection image is recorded into the power transmission line defect identification device. Therefore, the method for acquiring the inspection image of the power transmission line by the power transmission line defect identification device is not particularly limited. It can be understood that the number of the obtained inspection images of the power transmission line can be one or more, and the application does not specifically limit the number.
Step S220: and extracting and processing the power transmission line inspection image through a candidate area generation network to obtain a plurality of candidate area images.
In the step, the power transmission line defect identification device extracts and processes the power transmission line inspection image through the candidate area generation network to obtain a plurality of candidate area images. It can be understood that the candidate area image, that is, the area of the power transmission line inspection image, may include the target detection device of the power transmission line, such as a tower and a power transmission line. The candidate area image may be obtained using information such as edges, textures, colors, etc. in the image. And the acquisition of the candidate area image is equivalent to the rough detection of the identification target, so that the pressure of subsequent defect identification work is favorably reduced, and the detection precision of defect identification is improved.
Specifically, performing primary feature extraction processing on the power transmission line inspection image to obtain a power transmission line feature map; and inputting the transmission line characteristic diagram into a candidate region generation network, and outputting rectangular candidate region images with various scales and aspect ratios. Specifically, in the candidate area generation network, a sliding window is acquired, and the sliding window defines 9 reference rectangular frames (anchor points); sliding on the power transmission line characteristic diagram by using a sliding window to obtain 9 reference rectangular frames; mapping the features of each position passed by the sliding window into a 256-dimensional feature vector to obtain a plurality of feature vectors; inputting each feature vector into a first fully-connected layer and a second fully-connected layer; each reference rectangular frame corresponds to 4 correction parameters respectively, and the second full-connection layer outputs 4 × 9=36 correction parameters in total; correcting one reference rectangular frame by using 4 correction parameters to obtain a candidate area; the first fully-connected layer outputs 2 × 9=18 scores, and each candidate region corresponds to 2 scores, which are the first score and the second score, respectively. The first score represents the possibility that the candidate area contains the transmission line target detection equipment, and the second score represents the possibility that the candidate area does not contain the transmission line target detection equipment; and determining the candidate area with the first score larger than the second score as a candidate area image containing power transmission line target detection equipment such as a tower and a power transmission line.
Specifically, the obtaining of the candidate region by correcting one reference rectangular frame by using 4 correction parameters includes: the candidate area generation network outputs 4 correction parameters for each reference rectangular frame: t1, t2, t3, t4; and correcting the reference rectangular frame by using four correction parameters to obtain a candidate area, wherein the adopted correction formula is as follows: the central abscissa X = w1 × t1+ X1 of the candidate region; the central ordinate Y = h1 × t2+ Y1 of the candidate region; width W = W1exp (t 3) of the candidate region; height H = H1exp (t 4) of the candidate region; wherein x1, y1, w1, h1 respectively represent the central abscissa, central ordinate, width and height of the reference rectangular frame.
In another embodiment of the application, the power transmission line defect identification device may further extract and process the image of the power transmission line to be identified based on a selective search algorithm to obtain a plurality of candidate area images. Or obtaining the candidate area image suspected to have the defect by a method of artificial experience knowledge and based on color space decomposition. The candidate region image is extracted by finding out the possible positions of defects in the image in advance, and the extraction of the candidate region image utilizes the information of textures, edges, colors and the like in the image.
Step S230: and performing image enhancement processing and image quality evaluation processing on each candidate region image to obtain a plurality of to-be-identified power transmission line images with image quality meeting preset image quality indexes.
Step S240: and performing feature extraction processing on the plurality of to-be-identified power transmission line images according to a preset convolutional neural network to obtain convolutional neural network features of the plurality of to-be-identified power transmission line images.
In the step, the power transmission line defect recognition device inputs a plurality of power transmission line images to be recognized into a preset convolutional neural network trained in advance, and the preset convolutional neural network performs feature extraction processing to obtain convolutional neural network features of the plurality of power transmission line images to be recognized. In one embodiment of the application, the size of the plurality of images of the power transmission line to be identified, which are input into the preset convolutional neural network, is a preset size.
Step S250: and carrying out classification and identification processing on the convolutional neural network characteristics according to the classifier, and determining defect type information of a plurality of images of the power transmission line to be identified.
In this step, the power transmission line defect recognition device performs classification recognition processing on the convolutional neural network features according to the classifier, and determines defect type information of a plurality of power transmission line images to be recognized, specifically, the power transmission line defect recognition device performs classification recognition processing on the convolutional neural network features by using an SVM (Support Vector Machine) classifier, and determines defect type information of a plurality of candidate areas. Namely: and classifying each candidate region by the SVM classifier so as to determine whether the plurality of candidate regions of the image of the power transmission line to be identified belong to a certain defect type.
Through the steps S210 to S250, the power transmission line defect identification device acquires the power transmission line inspection image, then performs image enhancement processing and image quality evaluation processing on the power transmission line inspection image to obtain a power transmission line image to be identified, the image quality of which meets the preset image quality index, and finally performs defect identification processing on the power transmission line image to be identified to obtain defect information. In the process of acceptance in construction completion, the power transmission line defect identification device can improve the success rate and accuracy of defect identification by improving the image quality of the image of the power transmission line to be identified, is favorable for improving the efficiency of defect investigation of the power transmission line in the stage of acceptance in construction completion, and provides reliable reference for subsequent overhaul of engineers.
Referring to fig. 3, fig. 3 is a flowchart illustrating a specific method of step S230 in fig. 2 according to the present application. Step S230: and performing image enhancement processing and image quality evaluation processing on each candidate region image to obtain a plurality of to-be-identified power transmission line images with image quality meeting a preset image quality index, wherein the steps include but are not limited to step S310, step S320 and step S330.
Step S310: and performing image enhancement processing on each candidate region image to obtain a first transmission line enhanced image.
In this step, the transmission line defect recognition device performs image enhancement processing on each candidate region image to obtain a first transmission line enhanced image. Optionally, in an embodiment of the present application, the image enhancement processing includes at least one of: defogging processing, filtering processing and sharpening processing.
Specifically, it can be understood that in the process of collecting the power transmission line inspection image, the collected power transmission line inspection image has a condition of unsatisfactory quality due to the influence of factors such as a collection angle, a weather condition, an unmanned aerial vehicle motion state, and a surrounding complex environment interference, and therefore, the power transmission line inspection image needs to be subjected to image enhancement processing to enhance the power transmission line inspection image and improve the quality of the power transmission line inspection image, so as to perform subsequent defect identification processing and defect identification processing.
Particularly, most of the power transmission lines are arranged in the field, and when the power transmission lines meet heavy fog weather and sand-dust weather, more noise can be introduced into collected power transmission line inspection images. By the defogging processing, each candidate area image can be clearer, and the image quality is improved. And in the process of transmitting the inspection image of the transmission line, gaussian noise or pulse noise can be generated, the noise can be filtered to a certain degree through filtering processing, the image quality is improved, and the quality of each candidate area image can be influenced. And the imaging system of unmanned aerial vehicle receives influence such as motion, attitude change, mechanical oscillation in flight process, leads to patrolling and examining the image that the in-process obtained and produces distortion and motion blur, and the image of field collection can have the blurred problem of edge usually simultaneously, through carrying out preliminary sharpening to every candidate area image, can strengthen the edge or the profile of image so that carry out subsequent defect identification and handle.
It will be appreciated that the image enhancement process may also include an image suppression process, a luminance or color transformation process, or a geometric transformation process. Therefore, the present application does not limit the specific processing included in the image enhancement processing performed by the power transmission line defect identification apparatus, as long as it can preliminarily improve the quality of each candidate region image to a certain extent.
Step S320: and performing image quality evaluation processing on the first power transmission line enhanced image to generate first image quality evaluation information, wherein the first image quality evaluation information represents the image quality of the power transmission line enhanced image.
In the step, the transmission line defect identification device carries out image quality evaluation processing on the first transmission line enhanced image to generate first image quality evaluation information, the first image quality evaluation information represents the image quality of the transmission line enhanced image, the quality of the first transmission line enhanced image can be preliminarily evaluated, and the efficiency and the quality of subsequent defect identification processing can be improved.
Optionally, in an embodiment of the present application, the performing image quality evaluation processing on the first power transmission line enhanced image to generate first image quality evaluation information further includes: performing quality feature extraction processing on the first power transmission line enhanced image to determine a plurality of quality features; performing single quality evaluation processing on the first transmission line enhanced image according to each quality characteristic to obtain a plurality of single quality characteristic scores; carrying out scoring weighting processing according to a preset weighting coefficient and a plurality of single quality feature scores to obtain a first image quality evaluation total score; and generating first image quality evaluation information according to the plurality of single quality feature scores and the first image quality evaluation total score.
Specifically, firstly, the transmission line defect recognition device performs quality feature extraction processing on the first transmission line enhanced image, and in the process of determining a plurality of quality features, the first transmission line enhanced image can be input into one or more preset feature extraction networks in parallel to perform feature extraction to obtain a plurality of quality features. It can be understood that the feature extraction may also be performed by a non-reference image quality evaluation algorithm, and the method of extracting the quality feature is not particularly limited in the present application.
Specifically, the quality characteristics include: image sharpness, color richness, contrast, brightness, color vividness, distortion, and resolution, among others. At least two quality characteristics are obtained, but the type of the obtained quality characteristics is not particularly limited.
And then, the transmission line defect identification device carries out single quality evaluation processing on the first transmission line enhanced image according to each quality characteristic to obtain a plurality of single quality characteristic scores. Specifically, when the quality characteristic is image definition, performing edge detection on the first transmission line enhanced image in a single quality evaluation processing process to obtain an edge detection result of each pixel point in the first transmission line enhanced image; and calculating the variance of the edge detection result of each pixel point in the first transmission line enhanced image to obtain the definition value of the first transmission line enhanced image. It will be appreciated that the higher the image sharpness, the higher the individual quality feature score, i.e. sharpness score value, and the lower the image sharpness, the lower the individual quality feature score. When the quality feature is color richness, the higher the individual quality feature score. When the quality feature is a distortion degree, the smaller the distortion degree, the higher the individual quality feature score, and the larger the distortion degree, the higher the individual quality feature score. For the specific scoring process for other quality features, detailed description is omitted here. The power transmission line defect identification device carries out quality evaluation on the first power transmission line enhanced image based on different quality characteristics to obtain a plurality of single quality characteristic scores.
Then, the power transmission line defect identification device carries out scoring weighting processing according to the preset weighting coefficient and the plurality of single quality feature scores to obtain a first image quality evaluation total score, and it can be understood that in the process of scoring the quality of the first power transmission line enhanced image, the importance degrees of different quality features are different, so that the corresponding weighting coefficient needs to be set in advance for the single quality feature score corresponding to the quality feature, and the weighting coefficient reflects the influence degree of the quality feature on the image quality. Firstly, normalization processing is carried out on each single quality feature score, then, weighting summation is carried out on the multiple single quality feature scores after normalization processing according to a preset weighting coefficient of each single quality feature score, and a first image quality assessment total score of the first power transmission line enhanced image can be obtained. The quality characteristics of multiple dimensions are comprehensively considered, the quality score of the enhanced image of the first power transmission line is determined, so that the quality of the enhanced image of the first power transmission line is evaluated, the evaluation of the enhanced image of the first power transmission line is more objective, and the accuracy and the speed of the image quality evaluation can be improved.
Step S330: and under the condition that the first image quality evaluation information meets a preset image quality index, determining the first power transmission line enhanced image as a power transmission line image to be identified.
In this step, the power transmission line defect identification device needs to judge the first image quality evaluation information according to a preset image quality index, and determines the first power transmission line enhanced image as the power transmission line image to be identified under the condition that the first image quality evaluation information of the first power transmission line enhanced image meets the preset image quality index. It can be understood that the preset image quality index includes a single quality feature score threshold corresponding to different quality features and an image quality evaluation total score threshold, and the judgment can be made only if a plurality of single quality feature scores are greater than the corresponding single quality feature score thresholds and the image quality evaluation total score is greater than the image quality evaluation total score threshold: the first image quality evaluation information of the first power transmission line enhanced image meets a preset image quality index. The values of the plurality of individual quality feature score thresholds may be different. The image quality evaluation total score threshold value can also be determined according to image quality requirements required by subsequent defect identification processing, and specific values of the single quality feature score threshold value and the image quality evaluation total score threshold value are not specifically limited.
Referring to fig. 4, fig. 4 is a flowchart illustrating another specific method of step S230 in fig. 2. Step S230: and performing image enhancement processing and image quality evaluation processing on each candidate region image to obtain a plurality of to-be-identified power transmission line images with image quality meeting a preset image quality index, wherein the steps include but are not limited to step S410, step S430 and step S440.
Step S410: under the condition that the first image quality evaluation information does not meet the preset image quality index, determining substandard quality characteristic information which does not meet the preset image quality index according to the first image quality evaluation information;
step S420: performing image feature enhancement processing on the first transmission line enhanced image according to the substandard quality feature information to obtain a second transmission line enhanced image;
step S430: performing image quality evaluation processing on the second transmission line enhanced image to obtain second image quality evaluation information;
step S440: and under the condition that the second image quality evaluation information meets the preset image quality index, determining the second transmission line enhanced image as the transmission line image to be identified.
Through the steps S410 to S440, after the power transmission line defect identification apparatus needs to determine the first image quality evaluation information according to the preset image quality index, under the condition that the first image quality evaluation information does not meet the preset image quality index, firstly, determining unqualified quality feature information which does not meet the preset image quality index according to the first image quality evaluation information, for example, when definition score based on definition does not reach a single quality feature score threshold value, the definition is an unqualified quality feature, and the unqualified quality feature information may include one or more unqualified quality features; then, performing image feature enhancement processing on the first transmission line enhanced image according to the substandard quality feature information to obtain a second transmission line enhanced image, specifically, determining substandard quality features according to the obtained substandard quality feature information, selecting corresponding image feature enhancement processing according to the substandard quality features, and processing the first transmission line enhanced image to obtain a second transmission line enhanced image, wherein the process is equivalent to accurate image enhancement processing, and the selected specific image feature enhancement processing is different on the basis of different substandard quality features, for example, when the substandard quality features are definition, the definition enhancement processing is performed; and performing image quality evaluation processing on the second power transmission line enhanced image to obtain second image quality evaluation information, and determining the second power transmission line enhanced image as the power transmission line image to be identified under the condition that the second image quality evaluation information meets a preset image quality index. And if the second image quality evaluation information still does not meet the preset image quality index, generating a quality evaluation report based on the second image quality evaluation information, and informing engineers that part of the transmission line inspection images are not suitable for subsequent defect identification processing, so as to remind the engineers to intervene in the processing in time.
Referring to fig. 5, fig. 5 is a schematic flowchart of a defect identification method according to another embodiment of the present invention, and after performing classification identification processing on the convolutional neural network features according to a classifier and determining defect type information of a plurality of to-be-identified power transmission line images, the method further includes, but is not limited to, step S510 and step S520.
Step S510: performing frame regression correction processing on the to-be-identified power transmission line image with the determined defect type information to determine defect position information;
in this step, the power transmission line defect identification device performs frame regression correction processing on the power transmission line image to be identified, which is determined to have defect type information, and determines defect position information. Specifically, the power transmission line defect identification device adjusts the position of the window through frame regression (Bounding-BoxRegression) correction processing, and performs frame regression on the image of the power transmission line to be identified, which corresponds to the convolutional neural network characteristics conforming to a certain defect type, so that the image of the power transmission line to be identified is closer to the real window position, accurate positioning of the defect is realized, and defect position information is obtained.
Step S520: and obtaining defect information according to the defect type information and the defect position information.
Referring to fig. 6, fig. 6 is a schematic diagram of a power transmission line defect identification device based on image quality evaluation according to an embodiment of the present application. The power transmission line defect identification device 600 based on image quality evaluation according to the embodiment of the present application includes one or more processors 610 and a memory 620, and one processor 610 and one memory 620 are taken as an example in fig. 6. The processor 610 and the memory 620 may be connected by a bus or other means, such as by a bus in FIG. 6.
The memory 620, as a non-transitory computer-readable storage medium, may be used to store non-transitory software programs as well as non-transitory computer-executable programs. Further, the memory 620 may include high-speed random access memory, and may also include non-transitory memory, such as at least one magnetic disk storage device, flash memory device, or other non-transitory solid state storage device. In some embodiments, the memory 620 may optionally include a memory 620 remotely disposed from the processor 610, and these remote memories 620 may be connected to the image quality evaluation-based transmission line defect identification apparatus 600 via a network. Examples of such networks include, but are not limited to, the internet, intranets, local area networks, mobile communication networks, and combinations thereof.
Those skilled in the art will appreciate that the configuration of the apparatus shown in fig. 6 does not constitute a limitation of the transmission line defect identifying apparatus 600 based on image quality evaluation, and may include more or less components than those shown, or some components in combination, or a different arrangement of components.
Non-transitory software programs and instructions required to implement the power transmission line defect identification method applied to the power transmission line defect identification apparatus 600 based on image quality evaluation in the above-described embodiment are stored in the memory 620, and when executed by the processor 610, the power transmission line defect identification method applied to the power transmission line defect identification apparatus 600 based on image quality evaluation in the above-described embodiment is performed, for example, the method steps S210 to S250 in fig. 2, the method steps S310 to S330 in fig. 3, the method steps S410 to S440 in fig. 4, and the method steps S510 to S520 in fig. 5 described above are performed.
The above-described embodiments of the apparatus are merely illustrative, wherein the units illustrated as separate components may or may not be physically separate, i.e. may be located in one place, or may also be distributed over a plurality of network elements. Some or all of the modules may be selected according to actual needs to achieve the purpose of the solution of this embodiment.
Furthermore, an embodiment of the present application further provides a computer-readable storage medium, which stores computer-executable instructions, which are executed by one or more processors, for example, by one of the processors 610 in fig. 6, and can cause the one or more processors 610 to execute the control method in the method embodiment, for example, execute the method steps S210 to S250 in fig. 2, the method steps S310 to S330 in fig. 3, the method steps S410 to S440 in fig. 4, and the method steps S510 to S520 in fig. 5 described above.
It will be understood by those of ordinary skill in the art that all or some of the steps, systems, and methods disclosed above may be implemented as software, firmware, hardware, or suitable combinations thereof. Some or all of the physical components may be implemented as software executed by a processor, such as a central processing unit, digital signal processor, or microprocessor, or as hardware, or as an integrated circuit, such as an application specific integrated circuit. Such software may be distributed on computer readable media, which may include computer storage media (or non-transitory media) and communication media (or transitory media). The term computer storage media includes volatile and nonvolatile, removable and non-removable media implemented in any method or technology for storage of information such as computer readable instructions, data structures, program modules or other data, as is well known to those skilled in the art. Computer storage media includes, but is not limited to, RAM, ROM, EEPROM, flash memory or other memory technology, CD-ROM, digital Versatile Disks (DVD) or other optical disk storage, magnetic cassettes, magnetic tape, magnetic disk storage or other magnetic storage devices, or any other medium which can be used to store the desired information and which can accessed by a computer. In addition, communication media typically embodies computer readable instructions, data structures, program modules or other data in a modulated data signal such as a carrier wave or other transport mechanism and includes any information delivery media as known to those skilled in the art.
Claims (10)
1. A power transmission line defect identification method based on image quality evaluation is characterized by being applied to a power transmission line defect identification device and comprising the following steps:
acquiring a power transmission line inspection image;
extracting and processing the power transmission line inspection image through a candidate area generation network to obtain a plurality of candidate area images;
performing image enhancement processing and image quality evaluation processing on each candidate region image to obtain a plurality of to-be-identified power transmission line images with image quality meeting preset image quality indexes;
performing feature extraction processing on the plurality of electric transmission line images to be identified according to a preset convolutional neural network to obtain convolutional neural network features of the plurality of electric transmission line images to be identified;
and carrying out classification and identification processing on the convolutional neural network characteristics according to a classifier, and determining defect type information of a plurality of images of the power transmission line to be identified.
2. The method for identifying the defects of the power transmission line according to claim 1, wherein the step of performing image enhancement processing and image quality evaluation processing on each candidate region image to obtain a plurality of power transmission line images to be identified, the image quality of which meets a preset image quality index, comprises the steps of:
performing the image enhancement processing on each candidate region image to obtain a first power transmission line enhanced image;
performing the image quality evaluation processing on the first power transmission line enhanced image to generate first image quality evaluation information, wherein the first image quality evaluation information represents the image quality of the power transmission line enhanced image;
and determining the first transmission line enhanced image as a transmission line image to be identified under the condition that the first image quality evaluation information meets a preset image quality index.
3. The method for identifying defects in an electric transmission line according to claim 2, wherein the performing the image quality evaluation processing on the first electric transmission line enhanced image to generate first image quality evaluation information further comprises:
performing quality feature extraction processing on the first power transmission line enhanced image to determine a plurality of quality features;
performing single quality evaluation processing on the first power transmission line enhanced image aiming at each quality feature to obtain a plurality of single quality feature scores;
carrying out scoring weighting processing according to a preset weighting coefficient and the plurality of single quality feature scores to obtain a first image quality evaluation total score;
generating the first image quality assessment information according to the plurality of individual quality feature scores and the first image quality assessment total score.
4. The method for identifying the defects of the power transmission line according to claim 3, wherein the image enhancement processing and the image quality evaluation processing are performed on each candidate region image to obtain a plurality of power transmission line images to be identified, the image quality of which meets a preset image quality index, and the method further comprises the following steps:
under the condition that the first image quality evaluation information does not meet the preset image quality index, determining substandard quality characteristic information which does not meet the preset image quality index according to the first image quality evaluation information;
performing image feature enhancement processing on the first transmission line enhanced image according to the substandard quality feature information to obtain a second transmission line enhanced image;
performing image quality evaluation processing on the second transmission line enhanced image to obtain second image quality evaluation information;
and determining the second transmission line enhanced image as the transmission line image to be identified under the condition that the second image quality evaluation information meets a preset image quality index.
5. The method according to claim 4, wherein the image enhancement process comprises at least one of: defogging processing, filtering processing and sharpening processing.
6. The method for identifying the defects of the power transmission line according to claim 1, wherein after the convolutional neural network features are classified and identified according to a classifier and the defect type information of the plurality of images of the power transmission line to be identified is determined, the method further comprises the following steps:
performing frame regression correction processing on the candidate region with the determined defect type information to determine defect position information;
and obtaining defect information according to the defect type information and the defect position information.
7. The method for identifying the defects of the power transmission line according to claim 6, wherein the extracting the power transmission line inspection image through the candidate area generating network to obtain a plurality of candidate area images further comprises:
and carrying out size scaling processing on the plurality of images of the power transmission line to be identified, and scaling the size of the plurality of images of the power transmission line to be identified to be a preset size.
8. A transmission line defect recognition device, characterized by comprising:
the image acquisition module is used for acquiring the inspection image of the power transmission line;
the image processing module is used for extracting and processing the power transmission line inspection image through the candidate area generation network to obtain a plurality of candidate area images; performing image enhancement processing and image quality evaluation processing on each candidate region image to obtain a plurality of to-be-identified power transmission line images with image quality meeting preset image quality indexes;
the defect identification module is used for extracting the features of the plurality of images of the power transmission line to be identified according to a preset convolutional neural network to obtain convolutional neural network features of the plurality of images of the power transmission line to be identified; and carrying out classification and identification processing on the convolutional neural network characteristics according to a classifier, and determining defect type information of a plurality of images of the power transmission line to be identified.
9. An electric transmission line defect identifying device, characterized by comprising: the method comprises the following steps: memory, processor and computer program stored on the memory and executable on the processor, characterized in that the processor implements the method for defect recognition according to any of claims 1 to 7 when executing the computer program.
10. A computer-readable storage medium storing computer-executable instructions for causing a computer to perform the defect identification method of any one of claims 1 to 7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210988364.XA CN115272284A (en) | 2022-08-17 | 2022-08-17 | Power transmission line defect identification method based on image quality evaluation |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210988364.XA CN115272284A (en) | 2022-08-17 | 2022-08-17 | Power transmission line defect identification method based on image quality evaluation |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN115272284A true CN115272284A (en) | 2022-11-01 |
Family
ID=83752770
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210988364.XA Pending CN115272284A (en) | 2022-08-17 | 2022-08-17 | Power transmission line defect identification method based on image quality evaluation |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115272284A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116228774A (en) * | 2023-05-10 | 2023-06-06 | 国网山东省电力公司菏泽供电公司 | Substation inspection image defect identification method and system based on image quality evaluation |
-
2022
- 2022-08-17 CN CN202210988364.XA patent/CN115272284A/en active Pending
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116228774A (en) * | 2023-05-10 | 2023-06-06 | 国网山东省电力公司菏泽供电公司 | Substation inspection image defect identification method and system based on image quality evaluation |
| CN116228774B (en) * | 2023-05-10 | 2023-09-08 | 国网山东省电力公司菏泽供电公司 | Substation inspection image defect identification method and system based on image quality evaluation |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US11887064B2 (en) | Deep learning-based system and method for automatically determining degree of damage to each area of vehicle | |
| CN109784333B (en) | Three-dimensional target detection method and system based on point cloud weighted channel characteristics | |
| CN111008961B (en) | Transmission line equipment defect detection method and system, equipment and medium thereof | |
| CN111222395A (en) | Target detection method and device and electronic equipment | |
| CN108664839B (en) | Image processing method and device | |
| CN109460754A (en) | A kind of water surface foreign matter detecting method, device, equipment and storage medium | |
| CN109886947A (en) | The high-tension bus-bar defect inspection method of convolutional neural networks based on region | |
| CN110599453A (en) | Panel defect detection method and device based on image fusion and equipment terminal | |
| CN112598922B (en) | Parking space detection method, device, equipment and storage medium | |
| CN116665080B (en) | Unmanned aerial vehicle deteriorated insulator detection method and system based on target recognition | |
| CN117677969A (en) | Defect detection method and device | |
| CN111274964A (en) | Detection method for analyzing water surface pollutants based on visual saliency of unmanned aerial vehicle | |
| CN113034378A (en) | Method for distinguishing electric automobile from fuel automobile | |
| CN110334775B (en) | Unmanned aerial vehicle line fault identification method and device based on width learning | |
| CN115272284A (en) | Power transmission line defect identification method based on image quality evaluation | |
| CN112288682A (en) | Electric power equipment defect positioning method based on image registration | |
| CN115409789A (en) | Power transmission line engineering defect detection method based on image semantic segmentation | |
| CN116485802B (en) | Insulator flashover defect detection method, device, equipment and storage medium | |
| CN117409083A (en) | Cable terminal identification method and device based on infrared image and improved YOLOV5 | |
| Nayan et al. | Detection of objects from noisy images | |
| Kaimkhani et al. | UAV with Vision to Recognise Vehicle Number Plates | |
| CN115984759A (en) | Substation switch state identification method and device, computer equipment and storage medium | |
| CN116071713A (en) | Zebra crossing determination method, device, electronic equipment and medium | |
| CN115937492A (en) | Transformer equipment infrared image identification method based on feature identification | |
| CN112927197B (en) | Method, device, equipment and storage medium for detecting rust of air conditioner outdoor unit bracket |
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 |