Fan defect detection method and system based on machine vision

Abstract

The invention discloses a fan defect detection method and system based on machine vision, and belongs to the technical field of image data processing, and the method comprises the steps: obtaining a historical shot image of a fan blade, generating a training data set, building a deep learning network architecture based on a convolutional neural network, and building a fan blade motion recognition model. Carrying out calculation processing by using a dark channel prior algorithm, obtaining a dynamic adjustment weight, training an AOD-Net model based on the dynamic adjustment weight, establishing a fan blade fuzzy correction model, carrying out real-time shooting on the fan blade, and carrying out motion identification correction by using a fan blade motion identification model. The fan blade fuzzy correction model is used for dust fog removal, the defect detection model is established, and the defect detection model is used for defect detection, so that the fan blade moving at a high speed can be accurately identified, the outline details of the fan blade are clear, and the state of the fan blade can be timely and accurately analyzed during defect detection. And the working performance of the fan blade is improved.

Description

Technical Field

[0001] The present application belongs to the field of image data processing technology, and specifically relates to a fan defect detection method and system based on machine vision. Background Art

[0002] Wind power generation is a renewable energy technology that converts wind energy into electrical energy. As a renewable, environmentally friendly, cost-effective energy form, it is of great significance to achieving sustainable development goals. The core component of a wind turbine is the wind turbine blade, also known as the rotor blade. Its working performance directly affects the working efficiency and service life of the wind turbine. Therefore, it is necessary to promptly detect and deal with defects on the surface of the wind turbine blade to improve its working performance.

[0003] Traditional methods of detecting blade damage and faults involve observation using a telescope and manual inspection by rappelling a rope. This method involves high-altitude work, resulting in high costs. Furthermore, manual inspections are labor-intensive, leading to long inspection times and low efficiency.

[0004] Due to the low efficiency and accuracy of manual inspection, in order to address the above defects, a method for inspecting wind blades based on surface image data of wind turbine blades based on machine vision theory and image processing technology has been proposed. However, some wind turbines are located in harsh environments, such as images taken in dusty, foggy, rainy and snowy weather. At the same time, the wind turbine blades are in motion, which will make the blade details in the captured images unclear, reducing the accuracy of the detection effect. Summary of the Invention

[0005] To solve the above problems and technical defects, the present application adopts the following technical solution, a fan defect detection method based on machine vision, comprising the following steps:

[0006] Step 1: Obtain historical images of wind turbine blades, pre-process the historical images, and generate a training data set;

[0007] Step 2: Establish a deep learning network architecture based on the convolutional neural network, use the training data set for training, and establish a wind turbine blade motion recognition model;

[0008] Step 3: Use the dark channel prior algorithm to calculate and process the historical images in the training data set to obtain dynamic adjustment weights, train the AOD-Net model based on the dynamic adjustment weights, and establish a fan blade blur correction model;

[0009] Step 4: Take real-time photos of the wind turbine blades, use the wind turbine blade motion recognition model to perform motion recognition correction on the real-time images of the wind turbine blades, and use the wind turbine blade blur correction model to remove dust and fog;

[0010] Step 5: Establish a defect detection model and use the defect detection model to perform defect detection on the wind turbine blade image after motion recognition correction and dust and fog removal.

[0011] Preferably, the establishing of the wind turbine blade motion recognition model includes:

[0012] The historical images in the training dataset are divided into continuous image segments of fixed length, and the input data dimensions are adjusted according to the data dimensions to meet the use requirements of the 3D convolutional neural network;

[0013] The first two C3 modules in the 3D convolutional neural network retain 2D convolution to capture dynamic features, and the 2D convolutions of the last two C3 modules are replaced with 3D convolution to extract spatiotemporal features;

[0014] Mark and filter the dynamic features and input them into the recurrent neural network;

[0015] In the fully connected layer of the recurrent neural network, the dynamic features are mapped to the probability distribution of the taken photos. The loop iteration is performed according to the sequence length of the probability distribution. The wind blade motion recognition model is established by combining the loop structure and spatiotemporal features.

[0016] Preferably, the calculation processing of the dark channel priori algorithm includes:

[0017] In the fast dark channel of historical images, guided filtering is used to accelerate the calculation and implement minimum value filtering to obtain the dark channel transmittance map;

[0018] Preset pixel ratio, select the brightest pixel in the dark channel transmittance map according to the preset pixel ratio, and use the selected pixel as the highest intensity pixel in the corresponding original image;

[0019] Calculate the transmittance, use the atmospheric scattering model combined with the transmittance calculation results to generate a preliminary deblurred image;

[0020] The clarity of the preliminary deblurred image is calculated using the Laplace variance algorithm, the noise level of the preliminary deblurred image is calculated using the block area variance analysis, and the contrast of the preliminary deblurred image is calculated using the image grayscale histogram;

[0021] The clarity, noise level, and contrast are normalized according to preset weights to generate dynamic adjustment weights.

[0022] Furthermore, the training of the AOD-Net model based on the dynamic weight adjustment includes:

[0023] A 3-layer fully connected network is established. In the input layer, the statistical features of the quality index and dark channel transmittance map are calculated.

[0024] Activate the linear rectification function in the hidden layer and calculate the weight dimension;

[0025] In the output layer, the weight offset of each convolutional layer of the AOD-Net model is calculated based on the statistical features and weight dimensions;

[0026] The backbone parameters of the AOD-Net model are fixed, and the weighted loss is calculated by combining the dehazing loss and quality indicators. The weight offset is corrected based on the weighted loss to obtain the dynamically adjusted weights after training.

[0027] Furthermore, the establishment of the fan blade fuzzy correction model includes:

[0028] The preliminary deblurred image and the original blurred image are spliced ​​in the channel dimension, and the channel-dimensional spliced ​​data is received at the input layer of the AOD-Net model;

[0029] Dynamically adjusted weights are input into the last three convolutional layers, the data spliced ​​into the channel dimension is modified based on the dynamically adjusted weights, and the modified deblurred image is output;

[0030] Compare the deblurred image with the corresponding clear image, and judge the deblurred image based on the comparison result. If the deblurred image does not meet the requirements, retrain it until it meets the requirements;

[0031] The dynamic adjustment weights corresponding to all deblurred images that meet the requirements are obtained, the dynamic adjustment weights are marked, and a fan blade blur correction model is established based on the marked dynamic adjustment weights.

[0032] Furthermore, the performing motion recognition correction on the real-time captured wind turbine blade image includes:

[0033] Use edge detection and contour detection algorithms to determine multiple corner points of the device object when it is static, and determine the target shape and size based on the corner points;

[0034] Determine the corner points of the device object in the image frame of the captured video, calculate the determined corner points with the target shape and size, and obtain the perspective transformation matrix of the device object;

[0035] Apply the perspective transformation matrix to the target device object in the image frame to deform the device object back to its original shape and size.

[0036] Furthermore, the process of establishing the defect detection model includes:

[0037] Obtain historical defect images of wind turbine blades, perform feature extraction on the historical defect images, obtain defect features, and generate a wind turbine blade image defect set;

[0038] Perform dimensionality reduction on the defect set of wind turbine blade images to generate image defect subsets, perform training calculations on the image defect subsets to generate multiple defect sub-classifiers, and perform weighted fusion on the defect sub-classifiers to generate a defect classifier.

[0039] A defect classifier is used to detect defects in the real-time images of wind turbine blades to determine whether the images contain defect features. If so, the defects are classified to determine the defect type of the wind turbine blades.

[0040] A fan defect detection system based on machine vision, comprising:

[0041] A data storage module is used to obtain historical images of wind turbine blades, pre-process the historical images, and generate a training data set;

[0042] A motion recognition module is used to establish a deep learning network architecture based on a convolutional neural network, use a training data set for training, establish a wind blade motion recognition model, and use the wind blade motion recognition model to perform motion recognition correction on real-time images of wind blades;

[0043] A blur correction module is used to calculate and process historical images in the training data set using a dark channel prior algorithm to obtain dynamic adjustment weights, train the AOD-Net model based on the dynamic adjustment weights, establish a fan blade blur correction model, and use the fan blade blur correction model to remove dust and fog from real-time fan blade images;

[0044] The defect detection module is used to take real-time photos of wind turbine blades, establish a defect detection model, and use the defect detection model to perform defect detection on wind turbine blade images after motion recognition correction and dust and fog removal.

[0045] An electronic device includes a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the program, the contents of the above-mentioned method for detecting defects in a wind turbine based on machine vision are implemented.

[0046] A computer-readable storage medium stores a computer program, which, when executed by a processor, implements the contents of the above-mentioned method for detecting defects in a wind turbine based on machine vision.

[0047] Compared with the prior art, the beneficial effects of this application are:

[0048] This application establishes a fan blade motion recognition model to perform motion recognition on running fan blades, and can accurately identify high-speed moving fan blades. It also establishes a fan blade blur correction model to remove dust and fog from external factors such as sand, fog, rain and snow, thereby improving image quality and making the contour details of the fan blades clear. When performing defect detection, it can timely and accurately analyze the status of the fan blades and improve the working performance of the fan blades. BRIEF DESCRIPTION OF THE DRAWINGS

[0049] In the attached figure:

[0050] Figure 1 A schematic diagram of the method steps of an embodiment of the present application;

[0051] Figure 2 A schematic diagram of the system structure of an embodiment of the present application;

[0052] Figure 3 This is a schematic diagram of the device structure of an embodiment of the present application. DETAILED DESCRIPTION

[0053] In order to make the purpose, technical solutions and advantages of the embodiments of the present application clearer, the technical solutions in the embodiments of the present application will be clearly and completely described below in combination with the drawings in the embodiments of the present application. Obviously, the described embodiments are part of the embodiments of the present application, not all of the embodiments. Generally, the components of the embodiments of the present application described and shown in the drawings here can be arranged and designed in various different configurations.

[0054] Example 1

[0055] like Figure 1 As shown, a fan defect detection method based on machine vision includes the following steps:

[0056] Obtain historical images of wind turbine blades, pre-process the historical images, and generate a training data set;

[0057] A deep learning network architecture was established based on a convolutional neural network, and a training dataset was used for training to establish a wind turbine blade motion recognition model.

[0058] Establishing a wind turbine blade motion recognition model includes:

[0059] The historical images in the training dataset are divided into continuous image segments of fixed length, and the input data dimensions are adjusted according to the data dimensions to meet the use requirements of the 3D convolutional neural network;

[0060] The first two C3 modules in the 3D convolutional neural network retain 2D convolution to capture dynamic features, and the 2D convolutions of the last two C3 modules are replaced with 3D convolution to extract spatiotemporal features;

[0061] Mark and filter the dynamic features and input them into the recurrent neural network;

[0062] In the fully connected layer of the recurrent neural network, the dynamic features are mapped to the probability distribution of the taken photos. The loop iteration is performed according to the sequence length of the probability distribution. The wind blade motion recognition model is established by combining the loop structure and spatiotemporal features.

[0063] The dark channel prior algorithm is used to calculate and process the historical images in the training data set to obtain dynamic adjustment weights. The AOD-Net model is trained based on the dynamic adjustment weights to establish a fan blade blur correction model.

[0064] The calculation process performed by the dark channel priori algorithm includes:

[0065] In the fast dark channel of historical images, guided filtering is used to accelerate the calculation and implement minimum value filtering to obtain the dark channel transmittance map;

[0066] Preset pixel ratio, select the brightest pixel in the dark channel transmittance map according to the preset pixel ratio, and use the selected pixel as the highest intensity pixel in the corresponding original image;

[0067] Calculate the transmittance, use the atmospheric scattering model combined with the transmittance calculation results to generate a preliminary deblurred image;

[0068] The clarity of the preliminary deblurred image is calculated using the Laplace variance algorithm, the noise level of the preliminary deblurred image is calculated using the block area variance analysis, and the contrast of the preliminary deblurred image is calculated using the image grayscale histogram;

[0069] The clarity, noise level, and contrast are normalized according to preset weights to generate dynamic adjustment weights.

[0070] Training the AOD-Net model based on dynamically adjusted weights includes:

[0071] A 3-layer fully connected network is established. In the input layer, the statistical features of the quality index and dark channel transmittance map are calculated.

[0072] Activate the linear rectification function in the hidden layer and calculate the weight dimension;

[0073] In the output layer, the weight offset of each convolutional layer of the AOD-Net model is calculated based on the statistical features and weight dimensions;

[0074] The backbone parameters of the AOD-Net model are fixed, and the weighted loss is calculated by combining the dehazing loss and quality indicators. The weight offset is corrected based on the weighted loss to obtain the dynamically adjusted weights after training.

[0075] Establishing the fan blade fuzzy correction model includes:

[0076] The preliminary deblurred image and the original blurred image are spliced ​​in the channel dimension, and the channel-dimensional spliced ​​data is received at the input layer of the AOD-Net model;

[0077] Dynamically adjusted weights are input into the last three convolutional layers, the data spliced ​​into the channel dimension is modified based on the dynamically adjusted weights, and the modified deblurred image is output;

[0078] Compare the deblurred image with the corresponding clear image, and judge the deblurred image based on the comparison result. If the deblurred image does not meet the requirements, retrain it until it meets the requirements;

[0079] The dynamic adjustment weights corresponding to all deblurred images that meet the requirements are obtained, the dynamic adjustment weights are marked, and a fan blade blur correction model is established based on the marked dynamic adjustment weights.

[0080] The wind turbine blades are photographed in real time, and the wind turbine blade motion recognition model is used to perform motion recognition correction on the real-time photographed wind turbine blade images, and the wind turbine blade blur correction model is used to remove dust and fog;

[0081] The motion recognition correction of the wind turbine blade images captured in real time includes;

[0082] Use edge detection and contour detection algorithms to determine multiple corner points of the device object when it is static, and determine the target shape and size based on the corner points;

[0083] Determine the corner points of the device object in the image frame of the captured video, calculate the determined corner points with the target shape and size, and obtain the perspective transformation matrix of the device object;

[0084] Apply the perspective transformation matrix to the target device object in the image frame to deform the device object back to its original shape and size.

[0085] A defect detection model is established and used to detect defects in wind turbine blade images after motion recognition correction and dust and fog removal.

[0086] The process of establishing a defect detection model includes:

[0087] Obtain historical defect images of wind turbine blades, perform feature extraction on the historical defect images, obtain defect features, and generate a wind turbine blade image defect set;

[0088] Perform dimensionality reduction on the defect set of wind turbine blade images to generate image defect subsets, perform training calculations on the image defect subsets to generate multiple defect sub-classifiers, and perform weighted fusion on the defect sub-classifiers to generate a defect classifier.

[0089] A defect classifier is used to detect defects in the real-time images of wind turbine blades to determine whether the images contain defect features. If so, the defects are classified to determine the defect type of the wind turbine blades.

[0090] Example 2

[0091] like Figure 2 As shown, a fan defect detection system based on machine vision includes:

[0092] A data storage module is used to obtain historical images of wind turbine blades, pre-process the historical images, and generate a training data set;

[0093] A motion recognition module is used to establish a deep learning network architecture based on a convolutional neural network, use a training data set for training, establish a wind blade motion recognition model, and use the wind blade motion recognition model to perform motion recognition correction on real-time images of wind blades;

[0094] A blur correction module is used to calculate and process historical images in the training data set using a dark channel prior algorithm to obtain dynamic adjustment weights, train the AOD-Net model based on the dynamic adjustment weights, establish a fan blade blur correction model, and use the fan blade blur correction model to remove dust and fog from real-time fan blade images;

[0095] The defect detection module is used to take real-time photos of wind turbine blades, establish a defect detection model, and use the defect detection model to perform defect detection on wind turbine blade images after motion recognition correction and dust and fog removal.

[0096] Example 3

[0097] like Figure 3 As shown, from a hardware perspective, the present application provides an embodiment of an electronic device that implements all or part of the content of a wind turbine defect detection method based on machine vision. The electronic device includes a service processor and a distributed memory. The service processor is connected to the memory. The distributed memory stores a service self-management program configured to store machine-readable instructions. The service processor executes the service self-management program. When the instructions are executed by the processor, the wind turbine defect detection method based on machine vision as described above is implemented.

[0098] From a hardware perspective, in order to effectively improve the flexibility, versatility, and efficiency of data acquisition, this application provides an embodiment of an electronic device that implements all or part of the machine vision-based wind turbine defect detection method. The electronic device specifically includes the following:

[0099] A processor, a memory, a communications interface, and a bus; wherein the processor, the memory, and the communications interface communicate with each other via the bus; the communications interface is used to implement information transmission between the core business system of the machine vision-based wind turbine defect detection method, user terminals, and related databases and other related devices; the logic controller can be a desktop computer, a tablet computer, a mobile terminal, etc., but this embodiment is not limited thereto.

[0100] In this embodiment, the logic controller can be implemented with reference to the embodiment of the fan defect detection method based on machine vision in the embodiment, the content of which is incorporated herein and the repeated parts are not repeated.

[0101] It is understandable that the user terminal may include a smart phone, a tablet electronic device, a network set-top box, a portable computer, a desktop computer, a personal digital assistant (PDA), a vehicle-mounted device, a smart wearable device, etc., wherein the smart wearable device may include smart glasses, smart watches, smart bracelets, etc.

[0102] In practical applications, portions of the machine vision-based wind turbine defect detection method may be executed on the electronic device as described above, or all operations may be performed on the client device. The specific selection may be based on the processing capabilities of the client device and the limitations of the user's usage scenario, and this application does not impose any restrictions on this. If all operations are performed on the client device, the client device may also include a processor.

[0103] The above-mentioned client device may have a communication module (i.e., a communication unit), which can communicate with a remote server to realize data transmission with the server. The server may include a server on the task scheduling center side, and other implementation scenarios may also include a server on an intermediate platform, such as a server on a third-party server platform that has a communication link with the task scheduling center server. The server may include a single computer device, or a server cluster consisting of multiple servers, or a server structure of a distributed device.

[0104] Example 4

[0105] An embodiment of the present application also provides a computer-readable storage medium capable of implementing the machine vision-based wind turbine defect detection method in the above-mentioned embodiment, where the execution subject is a server or a client. The computer-readable storage medium stores a computer program, which, when executed by a processor, implements all the contents of the machine vision-based wind turbine defect detection method in the above-mentioned embodiment, where the execution subject is a server or a client.

[0106] The embodiments of the present application may be provided as methods, apparatuses, or computer program products. Thus, the present application may take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Moreover, the present application may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0107] The present application is described with reference to the flowcharts and / or block diagrams of the methods, devices (apparatus), and computer program products according to the embodiments of the present application. It should be understood that each process and / or block in the flowchart and / or block diagram, as well as the combination of the processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 a process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0108] These computer program instructions may also be stored in a computer readable memory that can direct a computer or other programmable data processing device to work in a specific manner, so that the instructions stored in the computer readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 a process or multiple processes and / or boxes Figure 1 The function specified in one or more boxes.

[0109] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operational steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing the instructions executed on the computer or other programmable device for implementing the process. Figure 1 a process or multiple processes and / or boxes Figure 1 The steps for the function specified in one or more boxes.

[0110] The above-described embodiments merely represent preferred implementations of the present application. While the descriptions are relatively specific and detailed, they should not be construed as limiting the scope of the present application. It should be noted that a person skilled in the art would be able to make numerous variations, improvements, and substitutions without departing from the spirit of the present application, all of which fall within the scope of protection of the present application.

Claims

1. A fan defect detection method based on machine vision, characterized in that: The following steps are involved: Step 1: Obtain historical images of wind turbine blades, pre-process the historical images, and generate a training data set; Step 2: Establish a deep learning network architecture based on the convolutional neural network, use the training data set for training, and establish a wind turbine blade motion recognition model; Step 3: Use the dark channel prior algorithm to calculate and process the historical images in the training data set to obtain dynamic adjustment weights, train the AOD-Net model based on the dynamic adjustment weights, and establish a fan blade blur correction model; Step 4: Take real-time photos of the wind turbine blades, use the wind turbine blade motion recognition model to perform motion recognition correction on the real-time images of the wind turbine blades, and use the wind turbine blade blur correction model to remove dust and fog; Step 5: Establish a defect detection model and use the defect detection model to perform defect detection on the wind turbine blade image after motion recognition correction and dust and fog removal.

2. The method for detecting fan defects based on machine vision according to claim 1, characterized in that: The establishment of the fan blade motion recognition model includes: The historical images in the training dataset are divided into continuous image segments of fixed length, and the input data dimensions are adjusted according to the data dimensions to meet the use requirements of the 3D convolutional neural network; The first two C3 modules in the 3D convolutional neural network retain 2D convolution to capture dynamic features, and the 2D convolutions of the last two C3 modules are replaced with 3D convolution to extract spatiotemporal features; Mark and filter the dynamic features and input them into the recurrent neural network; In the fully connected layer of the recurrent neural network, the dynamic features are mapped to the probability distribution of the taken photos. The loop iteration is performed according to the sequence length of the probability distribution. The wind blade motion recognition model is established by combining the loop structure and spatiotemporal features.

3. The method for detecting fan defects based on machine vision according to claim 1, characterized in that: The calculation processing of the dark channel priori algorithm includes: In the fast dark channel of historical images, guided filtering is used to accelerate the calculation and implement minimum value filtering to obtain the dark channel transmittance map; Preset pixel ratio, select the brightest pixel in the dark channel transmittance map according to the preset pixel ratio, and use the selected pixel as the highest intensity pixel in the corresponding original image; Calculate the transmittance, use the atmospheric scattering model combined with the transmittance calculation results to generate a preliminary deblurred image; The clarity of the preliminary deblurred image is calculated using the Laplace variance algorithm, the noise level of the preliminary deblurred image is calculated using the block area variance analysis, and the contrast of the preliminary deblurred image is calculated using the image grayscale histogram; The clarity, noise level, and contrast are normalized according to preset weights to generate dynamic adjustment weights.

4. The method for detecting fan defects based on machine vision according to claim 3, characterized in that: The training of the AOD-Net model based on the dynamic weight adjustment includes: A 3-layer fully connected network is established. In the input layer, the statistical features of the quality index and dark channel transmittance map are calculated. Activate the linear rectification function in the hidden layer and calculate the weight dimension; In the output layer, the weight offset of each convolutional layer of the AOD-Net model is calculated based on the statistical features and weight dimensions; The backbone parameters of the AOD-Net model are fixed, and the weighted loss is calculated by combining the dehazing loss and quality indicators. The weight offset is corrected based on the weighted loss to obtain the dynamically adjusted weights after training.

5. The method for detecting fan defects based on machine vision according to claim 4, characterized in that: The establishment of the fan blade fuzzy correction model includes: The preliminary deblurred image and the original blurred image are spliced ​​in the channel dimension, and the channel-dimensional spliced ​​data is received at the input layer of the AOD-Net model; Dynamically adjusted weights are input into the last three convolutional layers, the data spliced ​​into the channel dimension is modified based on the dynamically adjusted weights, and the modified deblurred image is output; Compare the deblurred image with the corresponding clear image, and judge the deblurred image based on the comparison result. If the deblurred image does not meet the requirements, retrain it until it meets the requirements; The dynamic adjustment weights corresponding to all deblurred images that meet the requirements are obtained, the dynamic adjustment weights are marked, and a fan blade blur correction model is established based on the marked dynamic adjustment weights.

6. The method for detecting fan defects based on machine vision according to claim 1, characterized in that: The performing motion recognition correction on the real-time captured wind turbine blade image comprises: Use edge detection and contour detection algorithms to determine multiple corner points of the device object when it is static, and determine the target shape and size based on the corner points; Determine the corner points of the device object in the image frame of the captured video, calculate the determined corner points with the target shape and size, and obtain the perspective transformation matrix of the device object; Apply the perspective transformation matrix to the target device object in the image frame to deform the device object back to its original shape and size.

7. The method for detecting fan defects based on machine vision according to claim 1, characterized in that: The process of establishing the defect detection model includes: Obtain historical defect images of wind turbine blades, perform feature extraction on the historical defect images, obtain defect features, and generate a wind turbine blade image defect set; Perform dimensionality reduction on the defect set of wind turbine blade images to generate image defect subsets, perform training calculations on the image defect subsets to generate multiple defect sub-classifiers, and perform weighted fusion on the defect sub-classifiers to generate a defect classifier. A defect classifier is used to detect defects in the real-time images of wind turbine blades to determine whether the images contain defect features. If so, the defects are classified to determine the defect type of the wind turbine blades.

8. A fan defect detection system based on machine vision, characterized in that: include: A data storage module is used to obtain historical images of wind turbine blades, pre-process the historical images, and generate a training data set; A motion recognition module is used to establish a deep learning network architecture based on a convolutional neural network, use a training data set for training, establish a wind blade motion recognition model, and use the wind blade motion recognition model to perform motion recognition correction on real-time images of wind blades; A blur correction module is used to calculate and process historical images in the training data set using a dark channel prior algorithm to obtain dynamic adjustment weights, train the AOD-Net model based on the dynamic adjustment weights, establish a fan blade blur correction model, and use the fan blade blur correction model to remove dust and fog from real-time fan blade images; The defect detection module is used to take real-time photos of wind turbine blades, establish a defect detection model, and use the defect detection model to perform defect detection on wind turbine blade images after motion recognition correction and dust and fog removal.

9. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein: When the processor executes the program, the content of the fan defect detection method based on machine vision described in claim 1 is implemented.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the content of the fan defect detection method based on machine vision described in claim 1 is implemented.

Images