Method and System for Measuring Rotation Speed of Wind Turbine Blade Based on Image Processing

Abstract

The present invention provides a method and system for measuring the rotational speed of the fan blades of a wind turbine based on image processing. The design is reasonable, flexible and convenient. As long as the wind turbine generator set is operating, it can measure in real time and accurately with low dependence on equipment and is easy to implement. The method includes: obtaining continuous videos of the rotation of the fan blades of the wind turbine collected by a fixed camera, and the acquisition time is not less than the time for the fan blades to rotate half a circle; reading the number of frames of the continuous videos, dividing the frames to obtain each frame image, extracting the frame with the highest clarity from the set number of frames of images starting from the first frame as the template image, performing histogram equalization on all frame images except the set number of frames of images to obtain comparison images; performing contour extraction on the template image and the comparison images, using the Euclidean distance for image matching, and obtaining the comparison image most similar to the template image as the matching image; calculating the rotational speed of the corresponding fan blades of the wind turbine according to the frame difference between the template image and the matching image.

Description

Technical Field

[0001] The present invention relates to the technical field of wind power generation, and specifically to a method and system for measuring the rotational speed of the fan blades of a wind turbine based on image processing. Background Art

[0002] A wind turbine is a power device that converts wind energy into mechanical work. The mechanical work drives the rotor to rotate, and finally outputs alternating current. Its working principle is relatively simple. The wind wheel rotates under the action of wind, converting the kinetic energy of the wind into the mechanical energy of the wind wheel shaft. The generator rotates to generate electricity driven by the wind wheel shaft. In order to improve the reliability and safety of the wind turbine in strong winds and provide a basis for judging generator overspeed protection, it is necessary to measure the rotational speed of the wind turbine.

[0003] There are various existing methods for measuring wind speed, but they all use the method of combining sensors with embedded technology: thermal anemometers, three-cup point vortex sensors, three-cup optocoupler induction sensors; or the method of counting pulse measurements. However, since the location of the wind turbine is generally remote, the height of the wind turbine is relatively high, and the environment is relatively unfavorable for maintenance, the existing sensors or chips have problems such as inconvenient maintenance and sometimes measurement deviations when in use. Summary of the Invention

[0004] Aiming at the problems existing in the prior art, the present invention provides a method and system for measuring the rotational speed of the fan blades of a wind turbine based on image processing, which is reasonably designed, flexible and convenient. As long as the wind power generation unit is running, it can measure in real time and accurately, has a low dependence on equipment, and is easy to implement.

[0005] The present invention is realized through the following technical solutions:

[0006] A method for measuring the rotational speed of the fan blades of a wind turbine based on image processing includes:

[0007] Obtaining continuous videos of the rotation of the fan blades of the wind turbine collected by a fixed camera, and the collection time is not less than the time for the fan blades to rotate half a circle;

[0008] Reading the number of frames of the continuous video, dividing the frames to obtain each frame image, extracting the frame with the highest clarity from the set number of frames of images starting from the first frame as the template image, and performing histogram equalization on all frame images except the set number of frames of images to obtain a comparison image;

[0009] Performing contour extraction on the template image and the comparison image, and using the Euclidean distance for image matching to obtain the comparison image most similar to the template image as the matching image;

[0010] Calculating the rotational speed of the corresponding fan blades of the wind turbine according to the frame difference between the template image and the matching image.

[0011] Optionally, after reading the number of consecutive video frames and obtaining each frame image by frame division, the following grayscale processing steps for each frame image are further included:

[0012] Through the energy gradient function, calculate the sum of the squares of the differences in grayscale values between adjacent pixels in the x and y directions of each pixel point in each frame image, and use it as the gradient value of each pixel point. Accumulate all pixel gradient values as the clarity evaluation function value of each frame image.

[0013] Optionally, when performing contour extraction on the template image and the comparison image, use the sobel edge detection algorithm to perform contour extraction on the template image and the comparison image.

[0014] Optionally, the specific steps for performing image matching using the Euclidean distance are as follows:

[0015] Compress the template image and the comparison image to a preset matching size;

[0016] Calculate the Euclidean distance d between the template image and the comparison image using the following formula:

[0017]

[0018] where i is the pixel point number sequence number, and are the corresponding grayscale values of the template image and the comparison image, and N is the number of pixel points;

[0019] Select the frame of the comparison image with the smallest Euclidean distance as the frame of the comparison image most similar to the template image, which is used as the matching image.

[0020] Optionally, the specific steps for selecting the frame of the comparison image with the smallest Euclidean distance as the frame of the comparison image most similar to the template image are as follows:

[0021] Draw a two-dimensional curve graph based on the Euclidean distance between the template image and the comparison image. The horizontal axis is the number of frames, and the vertical axis is the Euclidean distance. The trough closest to zero in the two-dimensional curve graph is the comparison image with the highest similarity, which is used as the matching image;

[0022] Among them, if the acquisition time is when the fan blade rotates multiple circles, through the comparison of the trough values, multiple troughs closest to the horizontal axis can be obtained. Discard the comparison image corresponding to the trough with the largest number of frames, and obtain multiple matching images corresponding to different rotation angles.

[0023] Optionally, calculating the corresponding rotational speed of the wind turbine fan blade according to the frame difference between the template image and the matching image specifically includes:

[0024] Calculate the frame difference corresponding to the template image and the matching image, and obtain the time T required to rotate the corresponding angle according to the frame difference;

[0025] According to the obtained time, the rotational speed of the wind turbine blades corresponding to the rotational angle is obtained by the following formula;

[0026] w = r / T;

[0027] v = rR / π;

[0028] Where: w is the angular velocity of the blades, v is the linear velocity of the blades, R is the radius of the blades, and r is the rotation angle.

[0029] A measurement system for the rotational speed of the wind turbine blades based on image processing includes

[0030] An image acquisition module for acquiring a continuous video of the rotation of the wind turbine blades collected by a fixed camera, and the acquisition time is not less than the time for the blades to rotate half a turn;

[0031] An image preprocessing module for reading the number of frames of the continuous video, splitting the frames to obtain each frame of image, extracting the frame with the highest clarity from the set number of frames of images as the template image, and performing histogram equalization on all frames of images except the set number of frames of images to obtain a comparison image;

[0032] An image matching module for extracting the contours of the template image and the comparison image, performing image matching using the Euclidean distance, and obtaining the comparison image most similar to the template image as the matching image;

[0033] A rotational speed calculation module for calculating the rotational speed of the corresponding wind turbine blades according to the frame difference between the template image and the matching image.

[0034] A computer device includes:

[0035] A memory for storing a computer program;

[0036] A processor for implementing the method for measuring the rotational speed of the wind turbine blades based on image processing as described in any one of the above when executing the computer program.

[0037] A computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, it implements the method for measuring the rotational speed of the wind turbine blades based on image processing as described in any one of the above.

[0038] Compared with the prior art, the present invention has the following beneficial technical effects:

[0039] The method for measuring the rotational speed of the fan blades of a wind turbine based on image processing according to the present invention can collect videos of the rotation of the fan blades by fixing the camera from different angles. Since the three fan blades have the same shape and size, the same frame images will appear after rotating 120 degrees. The acquisition time should ensure that the rotation of the fan blades is more than half a turn, and then the two frames of images with the highest theoretical similarity can be obtained by comparison; thus, through video acquisition, rather than direct sensor output, the rotational speed of the fan blades of the wind turbine can be calculated in real time at different positions by means of computer vision for rotational speed measurement; the digital image processing technology is adopted, the method is flexible, easy to implement, and has a low cost; and there is no limitation on the shooting tools used, and images with any resolution can be used for any camera or video camera. Description of the Drawings

[0040] Figure 1 It is a flowchart of the method described in the embodiment of the present invention.

[0041] Figure 2 It is a structural block diagram of the system described in the embodiment of the present invention.

[0042] Figure 3 It is a schematic diagram of image acquisition of the wind turbine described in the embodiment of the present invention.

[0043] In the figure: 1, 2, and 3 respectively represent the three fan blades of the wind turbine. Detailed Embodiments

[0044] The following further describes the present invention in detail with specific embodiments, which are explanations rather than limitations of the present invention.

[0045] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are some, but not all, of the embodiments of the present invention. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0046] It should be noted that, without conflict, the embodiments in the present application and the features in the embodiments may be combined with each other.

[0047] The present invention may be described in the general context of computer-executable instructions executed by a computer, such as program modules. Generally, program modules include routines, programs, objects, components, data structures, etc. that perform particular tasks or implement particular abstract data types. The present invention may also be practiced in distributed computing environments where tasks are performed by remote processing devices connected through a communication network. In a distributed computing environment, program modules may be located in both local and remote computer storage media including storage devices.

[0048] In the present invention, terms such as "module", "device", "system", etc. refer to related entities applied to a computer, such as hardware, a combination of hardware and software, software, or software in execution. Specifically, for example, a component may, but is not limited to, be a process running on a processor, a processor, an object, an executable component, an execution thread, a program, and / or a computer. Also, an application program or a script program running on a server, and the server can both be components. One or more components may be in an execution process and / or thread, and the components may be localized on one computer and / or distributed between two or more computers, and can be run by various computer-readable media. The components can also communicate through local and / or remote processes according to a signal having one or more data packets, for example, a signal from data that interacts with another component in a local system, a distributed system, and / or interacts with other systems through a network in the Internet.

[0049] Finally, it should also be noted that in this article, relational terms such as first and second are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "comprising" and "including" not only include those elements, but also include other elements not expressly listed, or also include elements inherent to such a process, method, article, or device. Without further limitation, an element defined by the statement "comprising..." does not exclude the existence of additional identical elements in the process, method, article, or device including the said element.

[0050] As Figure 1 shown, a method for measuring the rotational speed of a wind turbine blade based on image processing according to the present invention includes:

[0051] Step S1, image acquisition, acquiring continuous videos of the rotation of the wind turbine blade collected by a fixed camera, and the acquisition time is not less than the time for the blade to rotate half a circle, so as to ensure that the blade rotates more than half a circle;

[0052] Step S2, Image preprocessing: Read the number of frames of the continuous video, frame the video to obtain each frame of image. Starting from the first frame, extract the frame with the highest clarity from the set number of frames as the template image. Perform histogram equalization on all frames of images except the set number of frames to obtain comparison images.

[0053] Among them, the set number of frames is preferably the first 10 frames. Thus, extract the frame with the highest clarity from the first 10 frames as the template image, perform histogram equalization on all frames of images except the first 10 frames to obtain comparison images, and improve the image quality.

[0054] Step S3, Image matching: Perform contour extraction on the template image and the comparison images, and use the Euclidean distance for image matching to obtain the comparison image most similar to the template image as the matching image.

[0055] Among them, use the sobel edge detection algorithm to perform contour extraction on the template image and the comparison images, and use the Euclidean distance for image matching to obtain the comparison image most similar to the template image as the matching image. The sobel algorithm is used because of its fast algorithm speed. Moreover, the difference between the windmill and the background is large, and the gray level changes greatly, and sobel can meet the requirements of contour extraction.

[0056] Step S4, Rotation speed calculation: Calculate the rotation speed of the wind turbine blades at the corresponding rotation angle, that is, the angular velocity and linear velocity, according to the frame difference between the template image and the matching image.

[0057] In the image preprocessing stage, read the number of frames of the collected video, perform frame processing on it, and grayscale it, denoted as frame a. This step performs histogram equalization on all frames except the first 10 frames to obtain comparison images, thereby improving the quality of the comparison images.

[0058] In the step of grayscaling it and denoting it as frame a, the energy gradient function is used. The sum of the squares of the differences in the gray values of adjacent pixels in the x and y directions is used as the gradient value of each pixel point, and the sum of all pixel gradient values is used as the clarity evaluation function value, that is, the index for evaluating which frame of image is the clearest, so as to perform clarity comparison and select the template image.

[0059] In the image matching stage, after the contour extraction, using the Euclidean distance for matching includes:

[0060] According to the clarity evaluation function value, extract the frame with the highest clarity from the first 10 frames as the template image, and use the processed images after the first 10 frames to adopt the image matching algorithm based on the Euclidean distance;

[0061] Compress the image to 100px×100px, and use the formula

[0062]

[0063] Calculate the Euclidean distance d between two images, namely the template image and the comparison image. Let i be the number of pixel points. and are the grayscale values corresponding to the two images, and obtain the frame of the image that is most similar to the template image, that is, the matching image. Since the image is a black-and-white image, there are only 0 and 255, so a unique grayscale value can be obtained for a pixel point. Since the smaller the Euclidean distance, the greater the similarity of the images.

[0064] Take the frame with the smallest Euclidean distance as the matching image and output which frame it is, denoted as the bi frame.

[0065] Draw a two-dimensional curve graph according to the Euclidean distance of the comparison image. The horizontal axis is the frame number, and the vertical axis is the Euclidean distance. The trough closest to zero on the curve graph is the matching image with the greatest similarity.

[0066] If the number of acquisitions is three circles, then by comparing the trough values, nine troughs closest to the horizontal axis can be obtained, and output their corresponding frames bi. To ensure the accuracy of the output results, multiple matching images can be obtained by discarding the acquisition of the trough of the largest frame.

[0067] According to the length of time, multiple matching images may be obtained, and the nth one can be numbered as the nth matching image.

[0068] Finally, calculate the angular velocity and linear velocity according to the frame difference. Specifically,

[0069] Output the frame difference S = b1 - a between the template image and the first matching image. Here, assume the video is N fps and the radius of the fan blade is R.

[0070] Obtain the time T = S / N (seconds) taken to rotate 120 degrees. The angular velocity of the fan blade is W = 2π / (3 * T), and the linear velocity V = 2πR / (3 * T). In the same way as above, the angular velocity and linear velocity of rotating 240 degrees, 360 degrees, etc. can be obtained.

[0071] In actual specific operations, it can be divided into the following three parts:

[0072] I. In the image preprocessing stage, read the number of frames of the collected video, divide the frames and extract the clearest frame among the first 10 frames, and perform histogram equalization on the image to improve the image quality. The method includes:

[0073] Fix the camera at any angle to collect the video of the rotation of the fan blades of the wind turbine, and ensure that the collection time is more than half a turn of the wind blades.

[0074] Take the collected video data as the input of the image preprocessing stage, perform frame reading and frame splitting operations on the video data, grayscale the frame pictures, extract the clearest frame from the first 10 frames of the video and output which frame it is, and denote it as frame a (in this step, use the energy gradient function, take the sum of the squares of the differences in grayscale values of adjacent pixels in the x and y directions as the gradient value of each pixel point, and accumulate all pixel gradient values as the clarity evaluation function value);

[0075] Perform histogram equalization on the extracted frame and all frames except the first ten frames to improve the picture quality.

[0076] II. Image Matching Stage The method includes:

[0077] Use the sobel edge detection algorithm to extract the contour of the image. Since the sobel algorithm is fast, and the difference between the wind turbine and its background is large, with large grayscale changes, sobel can meet the requirements of contour extraction;

[0078] Perform closing operation denoising on the picture to improve the matching accuracy;

[0079] Take the extracted image as the template image, and use the processed images after the first 10 frames, and adopt the image matching algorithm based on the Euclidean distance. The method is as follows: Compress the image into 100*100, and use the formula Calculate the Euclidean distance d between the two pictures, i is the number of the pixel point, and are the corresponding grayscale values of the two pictures. Here it is a black and white image, so there are only 0 and 255. The smaller the Euclidean distance, the greater the similarity of the pictures), and obtain the frame image (matching image) that is most similar to the template image;

[0080] Take the frame with the smallest Euclidean distance as the matching image and output which frame it is, and denote it as frame bi;

[0081] Use the Euclidean distance to draw a two-dimensional curve graph, with the number of frames on the horizontal axis and the Euclidean distance on the vertical axis. The trough of the curve graph approaching zero is the matching image with the greatest similarity;

[0082] To ensure the accuracy of the output result, discard the largest frame. Multiple matching images can be obtained from the trough. According to the length of time, multiple matching images may be obtained, and the nth one is the nth matching image.

[0083] III. Speed Calculation Stage The method includes:

[0084] As Figure 3As shown, in the images extracted from the video, the three fan blades of the wind turbine are identical. When the first fan blade rotates to the second position, there is a pair of images with the maximum similarity. According to the length of the video captured, the times \(T\) for the blade to rotate 120 degrees, 240 degrees, and 360 degrees can be obtained. If the video is \(S\) fps and the frame difference is \(n\), then \(T = n / S\). By calculation, the angular velocity \(w\) of the fan blade can be obtained as \(w = r / T\), and the linear velocity \(v\) as \(v = rR / \pi\). Here, \(r\) is the angle of rotation and \(R\) is the radius of the fan blade. This method can be used to develop a speed measurement software for workers to detect wind turbines. At the same time, it can be calibrated with the speed measurement device of the wind turbine itself to improve the stability of the wind turbine and monitor its safety.

[0085] Specifically, the frame difference \(S\) between the output template image and the No. 1 matching image is \(S = b1 - a\);

[0086] Assume the video is \(N\) fps and the radius of the fan blade is \(R\). The time \(T\) taken for the blade to rotate 120 degrees is \(T = S / N\) (seconds). The angular velocity \(W\) of the fan blade is \(W = 2\pi / (3T)\), and the linear velocity \(V\) is \(V = 2\pi R / (3T)\);

[0087] According to the above method, the angular velocity and linear velocity for the blade to rotate 240 degrees, 360 degrees, etc. can be obtained.

[0088] The present invention also discloses a fan blade rotation speed measurement system for wind turbines based on image processing, as Figure 2 shown, including:

[0089] An image acquisition module 201, which is used to acquire continuous videos of the rotation of the fan blades of the wind turbine captured by a fixed camera, and the acquisition time is not less than the time for the fan blades to rotate half a circle;

[0090] An image preprocessing module 202, which is used to read the number of frames of the continuous video, frame the video to obtain each frame image, extract the frame with the highest clarity from the set number of frames of images as the template image, and perform histogram equalization on all frame images except the set number of frames of images to obtain comparison images;

[0091] An image matching module 203, which is used to extract the contours of the template image and the comparison images, and perform image matching using the Euclidean distance to obtain the comparison image most similar to the template image as the matching image;

[0092] A rotation speed calculation module 203, which is used to calculate the rotation speed of the corresponding fan blades of the wind turbine according to the frame difference between the template image and the matching image.

[0093] It can be implemented by an ordinary PC, and the fixed camera, as the imaging system to be calibrated, can be a camera, a video camera, or a webcam.

[0094] Those skilled in the art should understand that the embodiments of the present invention can be provided as a method, a system, or a computer program product. Therefore, the present invention can take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware aspects. Moreover, the present invention can take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk memory, CD-ROM, optical memory, etc.) that contain computer-usable program code.

[0095] The present invention is described with reference to the flowcharts and / or block diagrams of methods, apparatuses (systems), and computer program products according to the embodiments of the present invention. It should be understood that each flow and / or block in the flowchart and / or block diagram, as well as the combination of flows 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 the processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing devices to generate a machine, such that the instructions executed by the processor of the computer or other programmable data processing devices generate means for implementing the functions specified in Figure 1 one flow or multiple flows and / or blocks Figure 1 one block or multiple blocks.

[0096] These computer program instructions can 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, such that the instructions stored in the computer-readable memory generate a manufactured article including instruction means that implement the functions specified in Figure 1 one flow or multiple flows and / or blocks Figure 1 one block or multiple blocks.

[0097] These computer program instructions can also be loaded onto a computer or other programmable data processing device, such that a series of operation steps are executed on the computer or other programmable device to generate a computer-implemented process, and thus the instructions executed on the computer or other programmable device provide steps for implementing the functions specified in Figure 1 one flow or multiple flows and / or blocks Figure 1 one block or multiple blocks.

[0098] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit them. Although the present invention has been described in detail with reference to the above embodiments, those of ordinary skill in the art should understand that: modifications or equivalent replacements can still be made to the specific embodiments of the present invention, and any modifications or equivalent replacements that do not depart from the spirit and scope of the present invention should be covered by the protection scope of the claims of the present invention.

Claims

1. A method for measuring the rotational speed of the fan blades of a wind turbine based on image processing, characterized in that, Including: Obtain the continuous video of the rotation of the three blades of the wind turbine collected by a fixed camera, and the acquisition time is not less than the time for the blade to rotate half a circle; Read the number of frames of the continuous video, frame it to obtain each frame of image, extract the frame with the highest clarity from the set number of frames of images starting from the first frame as the template image, and perform histogram equalization on all frame images except the set number of frames of images to obtain a comparison image; Perform contour extraction on the template image and the comparison image, and use the Euclidean distance for image matching to obtain the comparison image most similar to the template image as the matching image; Calculate the rotation speed of the corresponding wind turbine blade according to the frame difference between the template image and the matching image; After reading the number of frames of the continuous video and framing to obtain each frame of image, the following grayscale processing steps for each frame of image are further included Through the energy gradient function, calculate the sum of the squares of the differences in grayscale values of adjacent pixels in the x and y directions of each pixel point in each frame of image, and use it as the gradient value of each pixel point. Accumulate all pixel gradient values as the clarity evaluation function value of each frame of image; The specific steps of using the Euclidean distance for image matching are as follows Compress the template image and the comparison image to a preset matching size; Use the following formula to calculate the Euclidean distance d between the template image and the comparison image where i is the serial number of the pixel points and are the gray values corresponding to the template image and the comparison image, and N is the number of pixel points; Select the frame of the comparison image with the smallest Euclidean distance as the frame of the comparison image most similar to the template image, as the matching image.

2. The method for measuring the rotational speed of the fan blade of a wind turbine based on image processing according to claim 1, wherein, When performing contour extraction on the template image and the comparison image, use the sobel edge detection algorithm to perform contour extraction on the template image and the comparison image.

3. The method for measuring the rotational speed of the fan blade of a wind turbine based on image processing according to claim 1, wherein The specific steps of selecting the frame of the comparison image with the smallest Euclidean distance as the frame of the comparison image most similar to the template image are as follows Draw a two-dimensional curve graph according to the Euclidean distance between the template image and the comparison image. The horizontal axis is the number of frames, and the vertical axis is the Euclidean distance. The trough closest to zero in the two-dimensional curve graph is the comparison image with the greatest similarity, as the matching image; Among them, if the acquisition time is for the blade to rotate multiple circles, through the comparison of the trough values, multiple troughs closest to the horizontal axis can be obtained, discard the comparison image corresponding to the trough with the largest number of frames, and obtain multiple matching images corresponding to different rotation angles.

4. The method for measuring the rotational speed of the fan blade of a wind turbine based on image processing according to claim 3, wherein The calculation of the rotation speed of the corresponding wind turbine blade according to the frame difference between the template image and the matching image specifically includes Calculate the frame difference corresponding to the template image and the matching image, and obtain the time T required to turn the corresponding angle according to the frame difference; According to the obtained time, obtain the rotation speed of the wind turbine blade at the corresponding rotation angle from the following formula; w = r / T; v = rR / T; Where: w is the angular velocity of the blade, v is the linear velocity of the blade, R is the radius of the blade, and r is the rotation angle.

5. A blade rotation speed measurement system for a wind turbine based on image processing, characterized in that, Including An image acquisition module for obtaining the continuous video of the rotation of the three blades of the wind turbine collected by a fixed camera, and the acquisition time is not less than the time for the blade to rotate half a circle; An image preprocessing module for reading the number of frames of the continuous video, framing to obtain each frame of image, extracting the frame with the highest clarity from the set number of frames of images starting from the first frame as the template image, and performing histogram equalization on all frame images except the set number of frames of images to obtain a comparison image; It is also used for, after framing to obtain each frame of image, performing grayscale processing on each frame of image according to the following steps Through the energy gradient function, calculate the sum of the squares of the differences in gray values of adjacent pixels in the x and y directions for each pixel in each frame of the image, and use it as the gradient value of each pixel. Accumulate all pixel gradient values as the sharpness evaluation function value of each frame of the image; An image matching module for extracting the contours of the template image and the comparison image, and using the Euclidean distance for image matching to obtain the comparison image most similar to the template image as the matching image; The specific steps of using the Euclidean distance for image matching are as follows Compress the template image and the comparison image to a preset matching size; Use the following formula to calculate the Euclidean distance d between the template image and the comparison image where i is the serial number of the pixel point number, and are the gray values corresponding to the template image and the comparison image, and N is the number of pixel points; Select the frame of the comparison image with the smallest Euclidean distance as the frame of the comparison image most similar to the template image and use it as the matching image; A rotational speed calculation module for calculating the corresponding rotational speed of the wind turbine blades based on the frame difference between the template image and the matching image.

6. A computer device, characterized in that, including: A memory for storing computer programs; A processor for implementing the method for measuring the rotational speed of the wind turbine blades based on image processing as described in any one of claims 1 to 4 when executing the computer program.

7. A computer-readable storage medium, characterized in that, A computer program is stored on the computer-readable storage medium, and when the computer program is executed by the processor, the method for measuring the rotational speed of the wind turbine blades based on image processing as described in any one of claims 1 to 4 is implemented.

Images